29:null 2a:["$","title",null,{"children":"Nanopore Sequencing Scripts | WabushOnce | Observable"}] 2f:T564,# Nanopore Sequencing Scripts - If the assembly tools are not installed then Install Assembly Tools ### Nanopore POD5-to-FastA: - [**Dorado Version Check**](#DoradoVersion): Ensure the latest Dorado Base-caller is Used - [**Dorado Basecalling**](#Dorado): Convert Nanopore voltage signals (.pod5) to reads (.fastq) - [**Demultiplexing**](#Demultiplexing): Split barcoded reads into individual barcode folders - [**Merge Barcodes**](#Merge_Barcodes): Combine multiple barcode .fasta files into one .fasta file - [**Remove Host Reads**](#Remove_Host): Remove Host Reads from .fastQ files - [**NanoComp or NanoPlot**](#NanoPlot): Assess the quality of the sequence reads - [**Filter Reads**](#Filter_Filtlong): Filter reads using FiltLong or Chopper - [**Flye**](#Flye): Use Flye to assemble by barcode and polish using Racon & Medaka - [**View Read Coverage**](#Create_BAM): Create .bam and .sam files to view read coverage - [**Pharokka**](#PharokkaAnnotation): Annotate phage FastA genomes - [**Batch Rename**](#BatchRename): Batch Rename Folders and Files ## Analysis BlastN Fasta Contigs: - Blast Contigs tool opens a FastA, displays all contigs contigs, and quickly BlastN each contig on PhagesDB or NCBI. --- 30:T9ee,viewof doradoChecker = { const releasesUrl = "https://github.com/nanoporetech/dorado/releases/latest"; const buttonStyle = ` padding: 0.5em 0.85em; border: 1px solid #bbb; border-radius: 6px; background: #f7f7f7; cursor: pointer; font-size: 0.9em; `; const latest = html`Not checked`; const checkBtn = html``; const openBtn = html``; const copyBtn = html``; const cmd = `conda activate reads-assembly-core dorado --version`; const codeBox = html`

${cmd}

`; function parseVersion(str) { const m = (str || "").match(/[0-9]+(?:\.[0-9]+)*/); return m ? m[0] : "unknown"; } checkBtn.onclick = async () => { latest.textContent = "checking..."; try { const resp = await fetch( "https://api.github.com/repos/nanoporetech/dorado/releases/latest" ); const data = await resp.json(); latest.textContent = parseVersion(data.tag_name); } catch (err) { latest.textContent = "check failed"; } }; openBtn.onclick = () => { window.open(releasesUrl, "_blank", "noopener,noreferrer"); }; copyBtn.onclick = () => { navigator.clipboard.writeText(cmd); copyBtn.textContent = "✅ Copied"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL code"; }, 1200); }; return html`

Dorado version checker

Use the latest basecaller for improved accuracy.
Latest Dorado version: ${latest}

${checkBtn} ${openBtn}

To check your installed version, copy and run the code below in WSL:

${copyBtn} ${codeBox}

`; }31:T11f7,viewof doradoLatestInstaller = { const envInput = Inputs.text({ label: "🧪 Conda environment name", value: "reads-assembly-core", style: "width: 320px" }); const textarea = html``; const copyBtn = html``; function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const envName = envInput.value.trim() || "reads-assembly-core"; const innerScript = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Dorado install failed at line $LINENO (exit $status)"; exit $status' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:\${PATH:-}" CONDA_ENV=${shQuote(envName)} echo "🧪 Installing/updating Dorado in conda env: $CONDA_ENV" if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Could not find conda.sh under $HOME/miniconda or $HOME/miniconda3" exit 1 fi conda activate "$CONDA_ENV" if [ -z "\${CONDA_PREFIX:-}" ]; then echo "❌ CONDA_PREFIX is empty after activation." exit 1 fi echo "📦 Active conda env path: $CONDA_PREFIX" INSTALLED_VER="none" if [ -x "$CONDA_PREFIX/dorado-current/bin/dorado" ]; then RAW=$("$CONDA_PREFIX/dorado-current/bin/dorado" --version 2>&1 | grep -Eo "^[0-9]+(\\.[0-9]+)*" | head -n1 || true) if [ -n "$RAW" ]; then INSTALLED_VER="$RAW" fi fi echo "🔎 Installed Dorado version: $INSTALLED_VER" echo "🌐 Checking latest Dorado release..." LATEST_URL=$(curl -fLsS -o /dev/null -w "%{url_effective}" https://github.com/nanoporetech/dorado/releases/latest) LATEST_TAG=\${LATEST_URL##*/} LATEST_VER=\${LATEST_TAG#v} echo "🌐 Latest Dorado version: $LATEST_VER" ver_gt () { [ -n "\${1:-}" ] && [ -n "\${2:-}" ] || return 1 [ "$1" = "$2" ] && return 1 [ "$(printf "%s\\n%s\\n" "$1" "$2" | sort -V | tail -n1)" = "$1" ] } NEED_INSTALL=0 if [ "$INSTALLED_VER" = "none" ]; then NEED_INSTALL=1 elif ver_gt "$LATEST_VER" "$INSTALLED_VER"; then NEED_INSTALL=1 fi if [ "$NEED_INSTALL" -eq 0 ]; then echo "✅ Dorado is already up to date." "$CONDA_PREFIX/dorado-current/bin/dorado" --version || true exit 0 fi WORKDIR="$CONDA_PREFIX/.dorado-install-tmp" mkdir -p "$WORKDIR" cd "$WORKDIR" TARBALL="dorado-\${LATEST_VER}-linux-x64.tar.gz" URL="https://cdn.oxfordnanoportal.com/software/analysis/\${TARBALL}" echo "⬇️ Downloading: $URL" curl -fL --retry 3 --retry-delay 2 -o "$TARBALL" "$URL" echo "📦 Extracting..." rm -rf "dorado-\${LATEST_VER}-linux-x64" tar -xzf "$TARBALL" NEW_DIR="$CONDA_PREFIX/dorado-\${LATEST_VER}-linux-x64" rm -rf "$NEW_DIR" mv "dorado-\${LATEST_VER}-linux-x64" "$NEW_DIR" ln -sfn "$NEW_DIR" "$CONDA_PREFIX/dorado-current" mkdir -p "$CONDA_PREFIX/etc/conda/activate.d" cat > "$CONDA_PREFIX/etc/conda/activate.d/dorado.sh" <<'EOF' export DORADO_HOME="$CONDA_PREFIX/dorado-current" export PATH="$DORADO_HOME/bin:$PATH" export LD_LIBRARY_PATH="$DORADO_HOME/lib:\${LD_LIBRARY_PATH:-}" EOF export DORADO_HOME="$CONDA_PREFIX/dorado-current" export PATH="$DORADO_HOME/bin:$PATH" export LD_LIBRARY_PATH="$DORADO_HOME/lib:\${LD_LIBRARY_PATH:-}" echo "🧪 Verifying Dorado..." which dorado dorado --version rm -rf "$WORKDIR" echo echo "✅ Done. Dorado is installed in:" echo " $CONDA_PREFIX/dorado-current" echo echo "To use it later:" echo " conda activate $CONDA_ENV" echo " dorado --version"`; textarea.value = `bash <<'DORADO_INSTALLER' ${innerScript} DORADO_INSTALLER`; } copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy Dorado Installer"), 1500); }; envInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ Install/update Dorado directly from pasted WSL code

${envInput}

Copy the generated block, paste it directly into WSL, and press Enter.

${textarea} ${copyBtn}

`; }32:T491d,viewof doradoBasecallDemuxScript = { const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 320px; max-width: 100%;" }); const modeInput = Inputs.select( new Map([ ["Custom barcodes", "custom"], ["ONT kit barcodes", "builtin"] ]), { label: "🏷️ Barcoding mode", value: "custom" } ); const classificationStageInput = Inputs.select( new Map([ ["Basecall only — retain barcodes (recommended)", "demux"], ["Basecall and demux — trim barcodes", "basecaller"] ]), { label: "🔬 Dorado workflow", value: "demux" } ); const kitInput = Inputs.text({ label: "🧬 Kit name", value: "PL27f", style: "width: 320px; max-width: 100%;" }); const builtinPresetInput = Inputs.select( new Map([ [ "SQK-RBK114-24 + SUP v4.3.0 + 5mC/5hmC all-contexts", "dna_r10.4.1_e8.2_400bps_sup@v4.3.0_5mC_5hmC@v1" ] ]), { label: "📈 Built-in model preset", value: "dna_r10.4.1_e8.2_400bps_sup@v4.3.0_5mC_5hmC@v1" } ); const modelInput = Inputs.text({ label: "📈 Custom model", value: "dna_r10.4.1_e8.2_400bps_sup@v5.2.0", style: "width: 520px; max-width: 100%;" }); const qscoreInput = Inputs.range([0, 30], { label: "🎯 Minimum Q-score", value: 10, step: 1 }); const pod5FolderInput = Inputs.text({ label: "📁 Folder of POD5 files (Windows path)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Data\\Pod5_Files", style: "width: 100%;" }); const barcodeArrangementInput = Inputs.text({ label: "🧬 Barcode arrangement TOML (Windows path)", placeholder: "C:\\Path\\To\\M13-pl27f.toml", style: "width: 100%;" }); const barcodeSequencesInput = Inputs.text({ label: "🧬 Barcode sequences FASTA (Windows path)", placeholder: "C:\\Path\\To\\pl27f.fasta", style: "width: 100%;" }); const bamNameInput = Inputs.text({ label: "📄 BAM file name", value: "Dorado_Untrimmed.bam", style: "width: 420px; max-width: 100%;" }); const demuxFolderInput = Inputs.text({ label: "📂 Demux output folder name", value: "Demux_Output", style: "width: 420px; max-width: 100%;" }); const emitFastqInput = Inputs.toggle({ label: "📤 Emit FASTQ during demux (off = demultiplexed BAM files)", value: true }); const emitSummaryInput = Inputs.toggle({ label: "📊 Emit barcode summary file", value: true }); const bothEndsInput = Inputs.toggle({ label: "🔒 Require barcode on both ends during classification", value: false }); const writeBamToLinuxHomeInput = Inputs.toggle({ label: "⚡ Write temporary BAM to Linux home first", value: true }); const textarea = html``; const copyButton = html``; const workflowNote = html`

`; function toWSLPath(winPath) { return (winPath || "") .trim() .replace(/^([A-Za-z]):/, (_, drive) => `/mnt/${drive.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(value) { return `'${String(value).replace(/'/g, `'\\''`)}'`; } function formatCommand(parts) { return parts .map((part, index) => { return index === 0 ? part : ` ${part}`; }) .join(" \\\n"); } const builtinPresetWrap = html`

${builtinPresetInput}

`; const modelWrap = html`

${modelInput}

`; const barcodeArrangementWrap = html`

${barcodeArrangementInput}

`; const barcodeSequencesWrap = html`

${barcodeSequencesInput}

`; /* * Left column containing the user controls. */ const controlsColumn = html`

${envInput} ${modeInput} ${classificationStageInput} ${workflowNote} ${kitInput} ${builtinPresetWrap} ${modelWrap} ${qscoreInput} ${pod5FolderInput} ${barcodeArrangementWrap} ${barcodeSequencesWrap} ${bamNameInput} ${demuxFolderInput} ${emitFastqInput} ${emitSummaryInput} ${bothEndsInput} ${writeBamToLinuxHomeInput}

The BAM is used for demultiplexing and then moved into the demux output folder. Only one BAM copy is retained.

Copy the generated script, paste it into WSL, and press Enter.

`; /* * Right column containing the generated WSL script. */ const outputColumn = html`

${textarea}

`; const columns = html`

${controlsColumn} ${outputColumn}

`; columns.classList.add("dorado-two-column-layout"); const copyButtonRow = html`

${copyButton}

`; function matchTextareaHeightToControls() { requestAnimationFrame(() => { const controlsHeight = Math.ceil( controlsColumn.getBoundingClientRect().height ); if (controlsHeight > 0) { textarea.style.height = `${controlsHeight}px`; } }); } function refreshVisibility() { const isCustom = modeInput.value === "custom"; const classifyDuringDemux = classificationStageInput.value === "demux"; barcodeArrangementWrap.style.display = isCustom ? "" : "none"; barcodeSequencesWrap.style.display = isCustom ? "" : "none"; modelWrap.style.display = isCustom ? "" : "none"; builtinPresetWrap.style.display = isCustom ? "none" : ""; workflowNote.innerHTML = classifyDuringDemux ? ` Recommended workflow: basecall with --no-trim to create a reusable untrimmed BAM. Dorado then classifies and separates the reads during dorado demux. After demultiplexing, the untrimmed BAM is moved into the demux output folder. ` : ` Trimmed workflow: classify and trim the barcodes during basecalling. Dorado then separates the existing barcode assignments using dorado demux --no-classify. After demultiplexing, the trimmed BAM is moved into the demux output folder. `; matchTextareaHeightToControls(); } function updateScript() { const envName = envInput.value.trim() || "reads-assembly-core"; const mode = modeInput.value || "custom"; const classificationStage = classificationStageInput.value || "demux"; const classifyDuringDemux = classificationStage === "demux"; const kitName = mode === "builtin" ? kitInput.value.trim() || "SQK-RBK114-24" : kitInput.value.trim() || "PL27f"; const modelName = mode === "builtin" ? builtinPresetInput.value || "dna_r10.4.1_e8.2_400bps_sup@v4.3.0_5mC_5hmC@v1" : modelInput.value.trim() || "dna_r10.4.1_e8.2_400bps_sup@v5.2.0"; const pod5WinPath = pod5FolderInput.value.trim(); const tomlWinPath = barcodeArrangementInput.value.trim(); const fastaWinPath = barcodeSequencesInput.value.trim(); const bamName = bamNameInput.value.trim() || (classifyDuringDemux ? "Dorado_Untrimmed.bam" : "Dorado_Trimmed.bam"); const demuxFolder = demuxFolderInput.value.trim() || "Demux_Output"; if (!modelName || !pod5WinPath) { textarea.value = ""; copyButton.disabled = true; refreshVisibility(); return; } if (mode === "custom" && (!tomlWinPath || !fastaWinPath)) { textarea.value = ""; copyButton.disabled = true; refreshVisibility(); return; } const pod5LinuxPath = toWSLPath(pod5WinPath).replace(/\/+$/, ""); const tomlLinuxPath = toWSLPath(tomlWinPath); const fastaLinuxPath = toWSLPath(fastaWinPath); const demuxDir = `${pod5LinuxPath}/${demuxFolder}`; /* * The BAM may first be written to Linux home for faster * basecalling output. It is moved into Demux_Output after * successful demultiplexing. */ const workBam = writeBamToLinuxHomeInput.value ? `$HOME/${bamName}` : `${pod5LinuxPath}/${bamName}`; const finalBam = `${demuxDir}/${bamName}`; /* * Barcode arguments are placed on whichever Dorado command * performs barcode classification. */ const barcodeArgs = [ `--kit-name ${shQuote(kitName)}`, ...(mode === "custom" ? [ `--barcode-arrangement ${shQuote(tomlLinuxPath)}`, `--barcode-sequences ${shQuote(fastaLinuxPath)}` ] : []), ...(bothEndsInput.value ? ["--barcode-both-ends"] : []) ]; /* * Recommended workflow: * Basecaller retains barcode sequences using --no-trim. * * Trimmed workflow: * Basecaller performs barcode classification and trimming. */ const basecallerParts = [ "dorado basecaller", "-r", "-v", `--min-qscore ${qscoreInput.value}`, ...(classifyDuringDemux ? ["--no-trim"] : barcodeArgs), shQuote(modelName), shQuote(pod5LinuxPath) ]; /* * Recommended workflow: * Demux performs barcode classification. * * Trimmed workflow: * Demux uses the existing barcode assignments. */ const demuxParts = [ "dorado demux", ...(classifyDuringDemux ? barcodeArgs : ["--no-classify"]), ...(emitFastqInput.value ? ["--emit-fastq"] : []), ...(emitSummaryInput.value ? ["--emit-summary"] : []), `--output-dir "$DEMUX_DIR"`, `"$WORK_BAM"` ]; const basecallerCommand = formatCommand(basecallerParts); const demuxCommand = formatCommand(demuxParts); const workBamAssignment = workBam.startsWith("$HOME/") ? `WORK_BAM="${workBam}"` : `WORK_BAM=${shQuote(workBam)}`; const finalBamAssignment = `FINAL_BAM=${shQuote(finalBam)}`; const customFileChecks = mode === "custom" ? ` if [[ ! -f ${shQuote(tomlLinuxPath)} ]]; then echo "❌ Barcode TOML file not found:" echo " ${tomlLinuxPath}" pause_if_terminal exit 1 fi if [[ ! -f ${shQuote(fastaLinuxPath)} ]]; then echo "❌ Barcode FASTA file not found:" echo " ${fastaLinuxPath}" pause_if_terminal exit 1 fi ` : ""; const summaryMessage = emitSummaryInput.value ? ` echo echo "Barcode summary:" echo " Written inside $DEMUX_DIR" ` : ""; const bamDescription = classifyDuringDemux ? "Reusable untrimmed BAM" : "Trimmed and classified BAM"; textarea.value = `# 🧬 Dorado basecall + demux (WSL) #!/usr/bin/env bash set -Eeuo pipefail pause_if_terminal() { if [[ -t 0 ]]; then echo echo "Press Enter to close..." read -r _ fi } trap ' status=$? echo echo "❌ Script failed at line $LINENO (exit $status)" pause_if_terminal exit "$status" ' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" # ============================================================ # User-selected workflow # ============================================================ WORKFLOW=${shQuote(classificationStage)} MODEL=${shQuote(modelName)} POD5_DIR=${shQuote(pod5LinuxPath)} DEMUX_DIR=${shQuote(demuxDir)} ${workBamAssignment} ${finalBamAssignment} # ============================================================ # Locate and activate Conda # ============================================================ if [[ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [[ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Could not find conda.sh" pause_if_terminal exit 1 fi conda activate ${shQuote(envName)} # ============================================================ # Check Dorado and input files # ============================================================ if ! command -v dorado >/dev/null 2>&1; then echo "❌ dorado not found" pause_if_terminal exit 1 fi if [[ ! -d "$POD5_DIR" ]]; then echo "❌ POD5 folder not found:" echo " $POD5_DIR" pause_if_terminal exit 1 fi ${customFileChecks} mkdir -p "$(dirname "$WORK_BAM")" mkdir -p "$DEMUX_DIR" echo echo "============================================================" echo "Dorado basecalling and demultiplexing" echo "============================================================" echo echo "Dorado version:" dorado --version echo echo "Workflow:" echo " $WORKFLOW" echo echo "POD5 folder:" echo " $POD5_DIR" echo echo "Temporary BAM location:" echo " $WORK_BAM" echo echo "Final BAM location:" echo " $FINAL_BAM" echo echo "Demux folder:" echo " $DEMUX_DIR" echo # ============================================================ # Step 1: Basecalling # ============================================================ echo "🧬 Basecalling..." ${basecallerCommand} \\ > "$WORK_BAM" echo echo "✅ Basecalling complete." # Check the BAM if samtools is installed. if command -v samtools >/dev/null 2>&1; then echo echo "🔍 Checking BAM integrity with samtools..." samtools quickcheck -v "$WORK_BAM" echo "✅ BAM passed the samtools check." else echo echo "⚠️ samtools was not found." echo " Skipping the BAM integrity check." fi # ============================================================ # Step 2: Barcode classification and demultiplexing # ============================================================ echo echo "📂 Demultiplexing..." ${demuxCommand} echo echo "✅ Demultiplexing complete." # ============================================================ # Step 3: Move the BAM into the demux output folder # ============================================================ echo echo "📦 Moving the BAM into the demux output folder..." if [[ "$WORK_BAM" != "$FINAL_BAM" ]]; then mv -f "$WORK_BAM" "$FINAL_BAM" fi echo "✅ ${bamDescription}:" echo " $FINAL_BAM" echo echo "============================================================" echo "✅ Complete" echo "============================================================" echo echo "Demultiplexed output:" echo " $DEMUX_DIR" echo echo "${bamDescription}:" echo " $FINAL_BAM" ${summaryMessage} pause_if_terminal`; copyButton.disabled = false; refreshVisibility(); matchTextareaHeightToControls(); } copyButton.onclick = async () => { try { await navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; setTimeout(() => { copyButton.textContent = "📋 Copy Script"; }, 1500); } catch (error) { copyButton.textContent = "❌ Copy failed"; setTimeout(() => { copyButton.textContent = "📋 Copy Script"; }, 1500); } }; [ envInput, modeInput, classificationStageInput, kitInput, builtinPresetInput, modelInput, qscoreInput, pod5FolderInput, barcodeArrangementInput, barcodeSequencesInput, bamNameInput, demuxFolderInput, emitFastqInput, emitSummaryInput, bothEndsInput, writeBamToLinuxHomeInput ].forEach((input) => { input.addEventListener("input", updateScript); }); modeInput.addEventListener("input", () => { if (modeInput.value === "builtin") { kitInput.value = "SQK-RBK114-24"; } else { kitInput.value = "PL27f"; } refreshVisibility(); updateScript(); }); classificationStageInput.addEventListener("input", () => { const currentName = bamNameInput.value.trim(); if ( classificationStageInput.value === "demux" && (currentName === "" || currentName === "Dorado_Output.bam" || currentName === "Dorado_Trimmed.bam") ) { bamNameInput.value = "Dorado_Untrimmed.bam"; } if ( classificationStageInput.value === "basecaller" && (currentName === "" || currentName === "Dorado_Untrimmed.bam") ) { bamNameInput.value = "Dorado_Trimmed.bam"; } refreshVisibility(); updateScript(); }); /* * Automatically match the textarea height to the controls. * This also updates when controls are hidden, revealed or wrap. */ const resizeObserver = new ResizeObserver(() => { matchTextareaHeightToControls(); }); resizeObserver.observe(controlsColumn); const responsiveStyle = html``; const content = html`

⬇ Dorado basecall + demux WSL script

${columns} ${copyButtonRow}

`; content.addEventListener("toggle", () => { if (content.open) { matchTextareaHeightToControls(); } }); refreshVisibility(); updateScript(); requestAnimationFrame(() => { matchTextareaHeightToControls(); }); return html` ${responsiveStyle} ${content} `; }33:T5f15,viewof doradoBasecallDemuxScriptUpdated = { const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 320px; max-width: 100%;" }); const modeInput = Inputs.select( new Map([ ["Custom barcodes", "custom"], ["ONT kit barcodes", "builtin"] ]), { label: "🏷️ Barcoding mode", value: "custom" } ); const classificationStageInput = Inputs.select( new Map([ ["Basecall only — retain barcodes (recommended)", "demux"], ["Basecall and demux — trim barcodes", "basecaller"] ]), { label: "🔬 Dorado workflow", value: "demux" } ); const kitInput = Inputs.text({ label: "🧬 Kit name", value: "PL27f", style: "width: 320px; max-width: 100%;" }); const builtinPresetInput = Inputs.select( new Map([ [ "SQK-RBK114-24 + SUP v4.3.0 + 5mC/5hmC all-contexts", "dna_r10.4.1_e8.2_400bps_sup@v4.3.0_5mC_5hmC@v1" ] ]), { label: "📈 Built-in model preset", value: "dna_r10.4.1_e8.2_400bps_sup@v4.3.0_5mC_5hmC@v1" } ); const modelInput = Inputs.text({ label: "📈 Custom model", value: "dna_r10.4.1_e8.2_400bps_sup@v5.2.0", style: "width: 520px; max-width: 100%;" }); const qscoreInput = Inputs.range([0, 30], { label: "🎯 Minimum Q-score", value: 10, step: 1 }); const pod5FolderInput = Inputs.text({ label: "📁 Folder of POD5 files (Windows path)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Data\\Pod5_Files", style: "width: 100%;" }); const barcodeArrangementInput = Inputs.text({ label: "🧬 Barcode arrangement TOML (Windows path)", placeholder: "C:\\Path\\To\\M13-pl27f.toml", style: "width: 100%;" }); const barcodeSequencesInput = Inputs.text({ label: "🧬 Barcode sequences FASTA (Windows path)", placeholder: "C:\\Path\\To\\pl27f.fasta", style: "width: 100%;" }); const barcodePhageMapInput = Inputs.text({ label: "🧾 Barcode → phage CSV (Windows path)", placeholder: "C:\\Path\\To\\barcode_phage_map.csv", style: "width: 100%;" }); const renameToPhageInput = Inputs.toggle({ label: "🦠 Rename demultiplexed files with phage names", value: true }); const bamNameInput = Inputs.text({ label: "📄 BAM file name", value: "Dorado_Untrimmed.bam", style: "width: 420px; max-width: 100%;" }); const demuxFolderInput = Inputs.text({ label: "📂 Demux output folder name", value: "Demux_Output", style: "width: 420px; max-width: 100%;" }); const emitFastqInput = Inputs.toggle({ label: "📤 Emit FASTQ during demux (off = demultiplexed BAM files)", value: true }); const emitSummaryInput = Inputs.toggle({ label: "📊 Emit barcode summary file", value: true }); const bothEndsInput = Inputs.toggle({ label: "🔒 Require barcode on both ends during classification", value: false }); const writeBamToLinuxHomeInput = Inputs.toggle({ label: "⚡ Write temporary BAM to Linux home first", value: true }); const textarea = html``; const copyButton = html``; const workflowNote = html`

${builtinPresetInput}

`; const modelWrap = html`

${modelInput}

`; const barcodeArrangementWrap = html`

${barcodeArrangementInput}

`; const barcodeSequencesWrap = html`

${barcodeSequencesInput}

`; const barcodePhageMapWrap = html`

${barcodePhageMapInput}

`; /* * Left column containing the user controls. */ const controlsColumn = html`

${envInput} ${modeInput} ${classificationStageInput} ${workflowNote} ${kitInput} ${builtinPresetWrap} ${modelWrap} ${qscoreInput} ${pod5FolderInput} ${barcodeArrangementWrap} ${barcodeSequencesWrap} ${barcodePhageMapWrap} ${renameToPhageInput} ${bamNameInput} ${demuxFolderInput} ${emitFastqInput} ${emitSummaryInput} ${bothEndsInput} ${writeBamToLinuxHomeInput}

Barcode/phage CSV example: barcode,phage. Files are renamed after demultiplexing while keeping both the barcode and phage name for traceability.

The BAM is used for demultiplexing and then moved into the demux output folder. Only one BAM copy is retained.

Copy the generated script, paste it into WSL, and press Enter.

`; /* * Right column containing the generated WSL script. */ const outputColumn = html`

${textarea}

`; const columns = html`

${controlsColumn} ${outputColumn}

`; columns.classList.add("dorado-two-column-layout"); const copyButtonRow = html`

${copyButton}

`; function matchTextareaHeightToControls() { requestAnimationFrame(() => { const controlsHeight = Math.ceil( controlsColumn.getBoundingClientRect().height ); if (controlsHeight > 0) { textarea.style.height = `${controlsHeight}px`; } }); } function refreshVisibility() { const isCustom = modeInput.value === "custom"; const classifyDuringDemux = classificationStageInput.value === "demux"; barcodeArrangementWrap.style.display = isCustom ? "" : "none"; barcodeSequencesWrap.style.display = isCustom ? "" : "none"; barcodePhageMapWrap.style.display = renameToPhageInput.value ? "" : "none"; modelWrap.style.display = isCustom ? "" : "none"; builtinPresetWrap.style.display = isCustom ? "none" : ""; workflowNote.innerHTML = classifyDuringDemux ? ` Recommended workflow: basecall with --no-trim to create a reusable untrimmed BAM. Dorado then classifies and separates the reads during dorado demux. After demultiplexing, the untrimmed BAM is moved into the demux output folder. ` : ` Trimmed workflow: classify and trim the barcodes during basecalling. Dorado then separates the existing barcode assignments using dorado demux --no-classify. After demultiplexing, the trimmed BAM is moved into the demux output folder. `; matchTextareaHeightToControls(); } function updateScript() { const envName = envInput.value.trim() || "reads-assembly-core"; const mode = modeInput.value || "custom"; const classificationStage = classificationStageInput.value || "demux"; const classifyDuringDemux = classificationStage === "demux"; const kitName = mode === "builtin" ? kitInput.value.trim() || "SQK-RBK114-24" : kitInput.value.trim() || "PL27f"; const modelName = mode === "builtin" ? builtinPresetInput.value || "dna_r10.4.1_e8.2_400bps_sup@v4.3.0_5mC_5hmC@v1" : modelInput.value.trim() || "dna_r10.4.1_e8.2_400bps_sup@v5.2.0"; const pod5WinPath = pod5FolderInput.value.trim(); const tomlWinPath = barcodeArrangementInput.value.trim(); const fastaWinPath = barcodeSequencesInput.value.trim(); const barcodePhageMapWinPath = barcodePhageMapInput.value.trim(); const bamName = bamNameInput.value.trim() || (classifyDuringDemux ? "Dorado_Untrimmed.bam" : "Dorado_Trimmed.bam"); const demuxFolder = demuxFolderInput.value.trim() || "Demux_Output"; if (!modelName || !pod5WinPath) { textarea.value = ""; copyButton.disabled = true; refreshVisibility(); return; } if (mode === "custom" && (!tomlWinPath || !fastaWinPath)) { textarea.value = ""; copyButton.disabled = true; refreshVisibility(); return; } if (renameToPhageInput.value && !barcodePhageMapWinPath) { textarea.value = ""; copyButton.disabled = true; refreshVisibility(); return; } const pod5LinuxPath = toWSLPath(pod5WinPath).replace(/\/+$/, ""); const tomlLinuxPath = toWSLPath(tomlWinPath); const fastaLinuxPath = toWSLPath(fastaWinPath); const barcodePhageMapLinuxPath = toWSLPath(barcodePhageMapWinPath); const demuxDir = `${pod5LinuxPath}/${demuxFolder}`; /* * The BAM may first be written to Linux home for faster * basecalling output. It is moved into Demux_Output after * successful demultiplexing. */ const workBam = writeBamToLinuxHomeInput.value ? `$HOME/${bamName}` : `${pod5LinuxPath}/${bamName}`; const finalBam = `${demuxDir}/${bamName}`; /* * Barcode arguments are placed on whichever Dorado command * performs barcode classification. */ const barcodeArgs = [ `--kit-name ${shQuote(kitName)}`, ...(mode === "custom" ? [ `--barcode-arrangement ${shQuote(tomlLinuxPath)}`, `--barcode-sequences ${shQuote(fastaLinuxPath)}` ] : []), ...(bothEndsInput.value ? ["--barcode-both-ends"] : []) ]; /* * Recommended workflow: * Basecaller retains barcode sequences using --no-trim. * * Trimmed workflow: * Basecaller performs barcode classification and trimming. */ const basecallerParts = [ "dorado basecaller", "-r", "-v", `--min-qscore ${qscoreInput.value}`, ...(classifyDuringDemux ? ["--no-trim"] : barcodeArgs), shQuote(modelName), shQuote(pod5LinuxPath) ]; /* * Recommended workflow: * Demux performs barcode classification. * * Trimmed workflow: * Demux uses the existing barcode assignments. */ const demuxParts = [ "dorado demux", ...(classifyDuringDemux ? barcodeArgs : ["--no-classify"]), ...(emitFastqInput.value ? ["--emit-fastq"] : []), ...(emitSummaryInput.value ? ["--emit-summary"] : []), `--output-dir "$DEMUX_DIR"`, `"$WORK_BAM"` ]; const basecallerCommand = formatCommand(basecallerParts); const demuxCommand = formatCommand(demuxParts); const workBamAssignment = workBam.startsWith("$HOME/") ? `WORK_BAM="${workBam}"` : `WORK_BAM=${shQuote(workBam)}`; const finalBamAssignment = `FINAL_BAM=${shQuote(finalBam)}`; const customFileChecks = mode === "custom" ? ` if [[ ! -f ${shQuote(tomlLinuxPath)} ]]; then echo "❌ Barcode TOML file not found:" echo " ${tomlLinuxPath}" pause_if_terminal exit 1 fi if [[ ! -f ${shQuote(fastaLinuxPath)} ]]; then echo "❌ Barcode FASTA file not found:" echo " ${fastaLinuxPath}" pause_if_terminal exit 1 fi ` : ""; const phageMapFileCheck = renameToPhageInput.value ? ` if [[ ! -f ${shQuote(barcodePhageMapLinuxPath)} ]]; then echo "❌ Barcode → phage CSV not found:" echo " ${barcodePhageMapLinuxPath}" pause_if_terminal exit 1 fi ` : ""; const phageRenameStep = renameToPhageInput.value ? ` # ============================================================ # Step 3: Rename demultiplexed files using barcode → phage map # ============================================================ echo echo "🦠 Applying barcode → phage names..." python3 - "$DEMUX_DIR" ${shQuote(barcodePhageMapLinuxPath)} <<'PY' import csv import re import sys from pathlib import Path demux_dir = Path(sys.argv[1]) map_path = Path(sys.argv[2]) def clean_name(value): value = (value or "").strip() value = re.sub(r"[^A-Za-z0-9._-]+", "_", value) value = re.sub(r"_+", "_", value).strip("_.-") return value def normalize_header(value): return re.sub(r"[^a-z0-9]+", "_", (value or "").strip().lower()).strip("_") def barcode_variants(barcode): b = (barcode or "").strip() if not b: return [] vals = {b, b.lower()} m = re.fullmatch(r"(?:barcode)?0*(\d+)", b, flags=re.I) if m: n = int(m.group(1)) vals.update({ f"barcode{n:02d}", f"barcode{n:03d}", f"barcode{n}", }) return sorted(vals, key=len, reverse=True) with map_path.open(newline="", encoding="utf-8-sig") as fh: reader = csv.DictReader(fh) if not reader.fieldnames: raise SystemExit("CSV has no header row.") normalized = {normalize_header(h): h for h in reader.fieldnames} barcode_col = next((normalized[k] for k in ( "barcode", "barcode_id", "barcode_name", "barcode_number" ) if k in normalized), None) phage_col = next((normalized[k] for k in ( "phage", "phage_name", "sample", "sample_name" ) if k in normalized), None) if not barcode_col or not phage_col: raise SystemExit( "CSV must contain a barcode column and a phage-name column. " "Accepted examples: barcode,phage or barcode,phage_name." ) mapping = [] seen = set() for row in reader: barcode = (row.get(barcode_col) or "").strip() phage = clean_name(row.get(phage_col)) if not barcode or not phage: continue key = barcode.lower() if key in seen: raise SystemExit(f"Duplicate barcode in CSV: {barcode}") seen.add(key) mapping.append((barcode, phage, barcode_variants(barcode))) if not mapping: raise SystemExit("No valid barcode/phage rows were found in the CSV.") files = [p for p in demux_dir.iterdir() if p.is_file()] renamed = [] unmatched = [] for path in files: lower_name = path.name.lower() if "unclassified" in lower_name: continue if path.name.startswith("Dorado_"): continue if "summary" in lower_name: continue match = None for barcode, phage, variants in mapping: if any(re.search(rf"(? barcode27_Winz.fastq suffixes = "".join(path.suffixes) clean_barcode = clean_name(barcode) new_name = f"{clean_barcode}_{phage}{suffixes}" target = path.with_name(new_name) if target.exists() and target != path: raise SystemExit(f"Refusing to overwrite existing file: {target.name}") if target != path: path.rename(target) renamed.append((path.name, target.name)) print(f"Renamed {len(renamed)} demultiplexed file(s).") for old, new in renamed: print(f" {old} -> {new}") if unmatched: print() print("Files not matched to the barcode/phage CSV:") for name in unmatched: print(f" {name}") PY echo "✅ Phage naming complete." ` : ""; const summaryMessage = emitSummaryInput.value ? ` echo echo "Barcode summary:" echo " Written inside $DEMUX_DIR" ` : ""; const bamDescription = classifyDuringDemux ? "Reusable untrimmed BAM" : "Trimmed and classified BAM"; textarea.value = `# 🧬 Dorado basecall + demux (WSL) #!/usr/bin/env bash set -Eeuo pipefail pause_if_terminal() { if [[ -t 0 ]]; then echo echo "Press Enter to close..." read -r _ fi } trap ' status=$? echo echo "❌ Script failed at line $LINENO (exit $status)" pause_if_terminal exit "$status" ' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" # ============================================================ # User-selected workflow # ============================================================ WORKFLOW=${shQuote(classificationStage)} MODEL=${shQuote(modelName)} POD5_DIR=${shQuote(pod5LinuxPath)} DEMUX_DIR=${shQuote(demuxDir)} BARCODE_PHAGE_MAP=${shQuote(barcodePhageMapLinuxPath)} ${workBamAssignment} ${finalBamAssignment} # ============================================================ # Locate and activate Conda # ============================================================ if [[ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [[ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Could not find conda.sh" pause_if_terminal exit 1 fi conda activate ${shQuote(envName)} # ============================================================ # Check Dorado and input files # ============================================================ if ! command -v dorado >/dev/null 2>&1; then echo "❌ dorado not found" pause_if_terminal exit 1 fi if [[ ! -d "$POD5_DIR" ]]; then echo "❌ POD5 folder not found:" echo " $POD5_DIR" pause_if_terminal exit 1 fi ${customFileChecks} ${phageMapFileCheck} mkdir -p "$(dirname "$WORK_BAM")" mkdir -p "$DEMUX_DIR" echo echo "============================================================" echo "Dorado basecalling and demultiplexing" echo "============================================================" echo echo "Dorado version:" dorado --version echo echo "Workflow:" echo " $WORKFLOW" echo echo "POD5 folder:" echo " $POD5_DIR" echo echo "Temporary BAM location:" echo " $WORK_BAM" echo echo "Final BAM location:" echo " $FINAL_BAM" echo echo "Demux folder:" echo " $DEMUX_DIR" echo # ============================================================ # Step 1: Basecalling # ============================================================ echo "🧬 Basecalling..." ${basecallerCommand} \\ > "$WORK_BAM" echo echo "✅ Basecalling complete." # Check the BAM if samtools is installed. if command -v samtools >/dev/null 2>&1; then echo echo "🔍 Checking BAM integrity with samtools..." samtools quickcheck -v "$WORK_BAM" echo "✅ BAM passed the samtools check." else echo echo "⚠️ samtools was not found." echo " Skipping the BAM integrity check." fi # ============================================================ # Step 2: Barcode classification and demultiplexing # ============================================================ echo echo "📂 Demultiplexing..." ${demuxCommand} echo echo "✅ Demultiplexing complete." ${phageRenameStep} # ============================================================ # Step 4: Move the BAM into the demux output folder # ============================================================ echo echo "📦 Moving the BAM into the demux output folder..." if [[ "$WORK_BAM" != "$FINAL_BAM" ]]; then mv -f "$WORK_BAM" "$FINAL_BAM" fi echo "✅ ${bamDescription}:" echo " $FINAL_BAM" echo echo "============================================================" echo "✅ Complete" echo "============================================================" echo echo "Demultiplexed output:" echo " $DEMUX_DIR" echo echo "${bamDescription}:" echo " $FINAL_BAM" ${summaryMessage} pause_if_terminal`; copyButton.disabled = false; refreshVisibility(); matchTextareaHeightToControls(); } copyButton.onclick = async () => { try { await navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; setTimeout(() => { copyButton.textContent = "📋 Copy Script"; }, 1500); } catch (error) { copyButton.textContent = "❌ Copy failed"; setTimeout(() => { copyButton.textContent = "📋 Copy Script"; }, 1500); } }; [ envInput, modeInput, classificationStageInput, kitInput, builtinPresetInput, modelInput, qscoreInput, pod5FolderInput, barcodeArrangementInput, barcodeSequencesInput, barcodePhageMapInput, renameToPhageInput, bamNameInput, demuxFolderInput, emitFastqInput, emitSummaryInput, bothEndsInput, writeBamToLinuxHomeInput ].forEach((input) => { input.addEventListener("input", updateScript); }); modeInput.addEventListener("input", () => { if (modeInput.value === "builtin") { kitInput.value = "SQK-RBK114-24"; } else { kitInput.value = "PL27f"; } refreshVisibility(); updateScript(); }); classificationStageInput.addEventListener("input", () => { const currentName = bamNameInput.value.trim(); if ( classificationStageInput.value === "demux" && (currentName === "" || currentName === "Dorado_Output.bam" || currentName === "Dorado_Trimmed.bam") ) { bamNameInput.value = "Dorado_Untrimmed.bam"; } if ( classificationStageInput.value === "basecaller" && (currentName === "" || currentName === "Dorado_Untrimmed.bam") ) { bamNameInput.value = "Dorado_Trimmed.bam"; } refreshVisibility(); updateScript(); }); /* * Automatically match the textarea height to the controls. * This also updates when controls are hidden, revealed or wrap. */ const resizeObserver = new ResizeObserver(() => { matchTextareaHeightToControls(); }); resizeObserver.observe(controlsColumn); const responsiveStyle = html``; const content = html`

⬇ Dorado basecall + demux WSL script

${columns} ${copyButtonRow}

⚠️ Please enter a valid path ending in .bam

`; const bamInput = Inputs.text({ label: "📄 Path to Dorado BAM file (include FileName.bam)", placeholder: "C:\\Users\\kiera\\Desktop\\Nanopore_Results\\2025-Jun-24_Run1\\Dorado_Output.bam", style: "width: 600px" }); const outFolderInput = Inputs.text({ label: "📁 Output folder (relative to BAM file)", placeholder: "Demux_Output", style: "width: 600px" }); const envInput = Inputs.text({ label: "🧪 Conda environment name", value: "reads-assembly-core", style: "width: 600px" }); const textarea = html``; const copyButton = html``; copyButton.onclick = () => { navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; setTimeout(() => (copyButton.textContent = "📋 Copy"), 1500); }; function toWSLPath(path) { return path .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateCommand() { const rawBam = bamInput.value.trim(); const outFolder = outFolderInput.value.trim() || "Demux_Output"; const envName = envInput.value.trim() || "reads-assembly-core"; const isValid = rawBam.toLowerCase().endsWith(".bam"); if (!isValid) { textarea.value = ""; copyButton.disabled = true; warning.style.display = "block"; return; } warning.style.display = "none"; copyButton.disabled = false; const linuxBamPath = toWSLPath(rawBam); const bamDir = linuxBamPath.substring(0, linuxBamPath.lastIndexOf("/")); const outputDir = `${bamDir}/${outFolder}`; textarea.value = `# 🧬 Dorado Demux Command (for WSL) set -euo pipefail export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" conda activate ${shQuote(envName)} if ! command -v dorado >/dev/null 2>&1; then echo "❌ dorado is not available in the active conda environment." echo " Expected environment: ${envName}" exit 1 fi if [ ! -f ${shQuote(linuxBamPath)} ]; then echo "❌ BAM file not found:" echo " ${linuxBamPath}" exit 1 fi mkdir -p ${shQuote(outputDir)} dorado demux --no-classify --emit-fastq \\ -o ${shQuote(outputDir)} \\ ${shQuote(linuxBamPath)} echo echo "✅ Demultiplexing complete." echo "📁 Output folder: ${outputDir}"`; } bamInput.addEventListener("input", generateCommand); outFolderInput.addEventListener("input", generateCommand); envInput.addEventListener("input", generateCommand); generateCommand(); return html`

⬇ Dorado demux WSL script

${warning} ${bamInput} ${outFolderInput} ${envInput}

Copy the generated script, paste it into WSL, and press Enter.

${copyButton}

${textarea}

`; }35:T1184,viewof fastqConcatScript23 = { const inputPath = Inputs.text({ label: "📂 Folder path (Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\Phage_580", style: "width: 600px" }); const recursiveInput = Inputs.toggle({ label: "🔁 Recreate subfolder structure inside Concatenated-FASTQ", value: true }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy WSL Script"), 1500); }; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function generateScript() { const winPath = inputPath.value.trim(); const recursive = recursiveInput.value; if (!winPath) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const basePath = toWSLPath(winPath); textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; echo "Press Enter to close..."; read -r _; exit $status' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" BASE_DIR="${basePath}" RECURSIVE=${recursive ? 1 : 0} OUTPUT_ROOT="$BASE_DIR/Concatenated-FASTQ" if [ ! -d "$BASE_DIR" ]; then echo "❌ Directory not found: $BASE_DIR" echo "Press Enter to close..." read -r _ exit 1 fi mkdir -p "$OUTPUT_ROOT" process_folder() { local folder="$1" local rel_path local dest_dir local parent_folder_name local out_file local files=() rel_path="\${folder#"$BASE_DIR"/}" parent_folder_name="$(basename "$folder")" if [ "$folder" = "$BASE_DIR" ]; then dest_dir="$OUTPUT_ROOT" else dest_dir="$OUTPUT_ROOT/$rel_path" fi shopt -s nullglob files=( "$folder"/*.fastq ) shopt -u nullglob if (( \${#files[@]} == 0 )); then echo "⚠ No .fastq files found in: $folder" return 0 fi mkdir -p "$dest_dir" # Output file is named after the folder containing the FASTQ files. out_file="$dest_dir/\${parent_folder_name}.fastq" echo echo "📁 Source folder: $folder" echo "📂 Output folder: $dest_dir" echo "➕ Concatenating \${#files[@]} .fastq file(s) into:" echo " $out_file" cat "\${files[@]}" > "$out_file" echo "✅ Done: $out_file" } if [ "$RECURSIVE" -eq 1 ]; then echo "🔁 Recreating folder structure under: $OUTPUT_ROOT" found_folder_with_fastq=0 while IFS= read -r -d '' folder; do [ "$folder" = "$OUTPUT_ROOT" ] && continue before_count="$found_folder_with_fastq" process_folder "$folder" shopt -s nullglob check_files=( "$folder"/*.fastq ) shopt -u nullglob if (( \${#check_files[@]} > 0 )); then found_folder_with_fastq=1 fi done < <(find "$BASE_DIR" -type d ! -path "$OUTPUT_ROOT" ! -path "$OUTPUT_ROOT/*" -print0) if [ "$found_folder_with_fastq" -eq 0 ]; then echo "⚠ No folders containing .fastq files found under: $BASE_DIR" fi else echo "📄 Processing only this folder: $BASE_DIR" process_folder "$BASE_DIR" fi echo echo "🎉 Finished concatenation." echo "📂 Output root: $OUTPUT_ROOT" echo "Press Enter to close..." read -r _ `; } inputPath.addEventListener("input", generateScript); recursiveInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ FASTQ concatenation WSL script

${inputPath} ${recursiveInput}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

${textarea}

`; }36:T1c04,viewof minimap2DehostScript = { const hostFastaInput = Inputs.text({ label: "🧬 Host genome FASTA (Windows path)", placeholder: "C:\\Users\\YourName\\Desktop\\host.fasta", style: "width: 600px" }); const readsFolderInput = Inputs.text({ label: "📁 Folder containing FASTQ files or sample subfolders", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results", style: "width: 600px" }); const modeInput = Inputs.select( [ "One folder: combine FASTQs into one dehosted file", "Parse immediate subfolders separately" ], { label: "Search mode", value: "Parse immediate subfolders separately" } ); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 300px" }); const threadsInput = Inputs.number({ label: "🧵 Threads", value: 4, min: 1, max: 64 }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy WSL Script"), 1500); }; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const hostWin = hostFastaInput.value.trim(); const readsWin = readsFolderInput.value.trim(); const env = envInput.value.trim() || "reads-assembly-core"; const threads = threadsInput.value || 1; const parseSubfolders = modeInput.value === "Parse immediate subfolders separately"; if (!hostWin || !readsWin) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const host = toWSLPath(hostWin); const reads = toWSLPath(readsWin); textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO with exit status $status"; echo "Press Enter to close..."; read -r _; exit "$status"' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" HOST_FASTA=${shQuote(host)} READS_DIR=${shQuote(reads)} THREADS=${threads} ENV_NAME=${shQuote(env)} PARSE_SUBFOLDERS=${parseSubfolders ? 1 : 0} # Load conda if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Could not find conda.sh" read -r _ exit 1 fi conda activate "$ENV_NAME" # Restore system PATH in case conda activation changes it export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" echo "HOST_FASTA = $HOST_FASTA" echo "READS_DIR = $READS_DIR" echo "THREADS = $THREADS" echo "ENV = $ENV_NAME" echo "PARSE_SUBFOLDERS = $PARSE_SUBFOLDERS" for tool in minimap2 samtools; do if ! command -v "$tool" >/dev/null 2>&1; then echo "❌ Missing tool: $tool" read -r _ exit 1 fi done if [ ! -f "$HOST_FASTA" ]; then echo "❌ Host FASTA not found: $HOST_FASTA" read -r _ exit 1 fi if [ ! -d "$READS_DIR" ]; then echo "❌ Reads folder not found: $READS_DIR" read -r _ exit 1 fi OUT_DIR="$READS_DIR/dehosted_output" mkdir -p "$OUT_DIR" INDEX="$OUT_DIR/host.mmi" if [ ! -f "$INDEX" ]; then echo "📦 Indexing host genome..." minimap2 -d "$INDEX" "$HOST_FASTA" else echo "ℹ️ Using existing host index: $INDEX" fi dehost_folder() { local input_folder="$1" local output_folder="$2" local sample="$3" mkdir -p "$output_folder" mapfile -t fq_files < <( find "$input_folder" -maxdepth 1 -type f \$ \\ -iname "*.fastq" -o \\ -iname "*.fq" -o \\ -iname "*.fastq.gz" -o \\ -iname "*.fq.gz" \\ \$ | sort ) if (( \${#fq_files[@]} == 0 )); then echo "⚠ No FASTQ files found in: $input_folder" return 0 fi sample="\${sample// /_}" tmp_fastq="$output_folder/\${sample}_combined.fastq" out_fastq="$output_folder/NoHost_\${sample}.fastq" echo echo "📁 Sample folder: $input_folder" echo "📁 Output folder: $output_folder" echo "✅ Found \${#fq_files[@]} FASTQ file(s)" : > "$tmp_fastq" echo "📦 Combining FASTQ files for sample: $sample" for f in "\${fq_files[@]}"; do echo " + $f" if [[ "$f" == *.gz ]]; then zcat "$f" >> "$tmp_fastq" else cat "$f" >> "$tmp_fastq" fi done echo "🧬 Removing host reads for sample: $sample" minimap2 -ax map-ont -t "$THREADS" "$INDEX" "$tmp_fastq" | \\ samtools view -b -f 4 - | \\ samtools fastq - > "$out_fastq" rm -f "$tmp_fastq" echo "✅ Wrote: $out_fastq" } if [ "$PARSE_SUBFOLDERS" -eq 1 ]; then echo echo "🔁 Parsing immediate subfolders separately..." found_folder=0 for folder in "$READS_DIR"/*/; do [ -d "$folder" ] || continue folder_clean="\${folder%/}" sample="$(basename "$folder_clean")" # Skip output folder if script is re-run if [ "$folder_clean" = "$OUT_DIR" ]; then continue fi found_folder=1 dehost_folder "$folder_clean" "$OUT_DIR/$sample" "$sample" done if [ "$found_folder" -eq 0 ]; then echo "⚠ No subfolders found under: $READS_DIR" echo "If your FASTQ files are directly in this folder, choose one-folder mode." fi else echo echo "📄 Processing one folder only..." sample="$(basename "$READS_DIR")" dehost_folder "$READS_DIR" "$OUT_DIR" "$sample" fi echo echo "🎉 Finished dehosting" echo "📁 Output folder: $OUT_DIR" echo "Press Enter to close..." read -r _ `; } hostFastaInput.addEventListener("input", generateScript); readsFolderInput.addEventListener("input", generateScript); modeInput.addEventListener("input", generateScript); envInput.addEventListener("input", generateScript); threadsInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ Minimap2 host-read removal WSL script

${hostFastaInput} ${readsFolderInput} ${modeInput} ${envInput} ${threadsInput}

Save as dehost_reads.sh, then run in WSL.

${copyBtn}

${textarea}

`; }37:T1fef,viewof runNanoCompFromBarcodeMapWSL = { const folderInput = Inputs.text({ label: "Windows FASTQ parent folder", placeholder: "C:\\Users\\kiera\\Desktop\\...\\fastq_pass", style: "width: 600px" }); const csvInput = Inputs.text({ label: "Optional CSV path — Column A = Barcode, Column B = Phage Name", placeholder: "C:\\Users\\kiera\\Desktop\\barcode_to_phage.csv", style: "width: 600px" }); const csvHelp = html`

CSV format when Sample naming is unchecked:
Column A = Barcode Folder Name Format
Column B = Phage Name

Example:

barcode001,Olympi
barcode002,Tadpole
barcode003,Seldom

This will match folders such as PL27f_barcode001, barcode001, or text_barcode001.

`; const useFolderNamesInput = Inputs.checkbox( ["Use folder name for sample names instead of CSV"], { label: "Sample naming", value: [] } ); const recursiveInput = Inputs.checkbox( ["Search subfolders for barcode FASTQ files"], { label: "Options", value: ["Search subfolders for barcode FASTQ files"] } ); const threadsInput = Inputs.range([1, 32], { label: "Threads", value: 4, step: 1 }); const scriptArea = html``; const copyBtn = html``; function convertPath(winPath) { if (!winPath) return "/path/to/file_or_folder"; return winPath .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\\\/g, "/") .replace(/\\/g, "/"); } function shellEscapeDoubleQuotes(s) { return String(s).replace(/"/g, '\\"'); } function updateScript() { const inputDir = convertPath(folderInput.value || ""); const csvPath = csvInput.value ? convertPath(csvInput.value) : ""; const useFolderNames = useFolderNamesInput.value.includes( "Use folder name for sample names instead of CSV" ) ? "true" : "false"; const depthArgs = recursiveInput.value.includes( "Search subfolders for barcode FASTQ files" ) ? "-mindepth 2" : "-mindepth 2 -maxdepth 2"; const threads = threadsInput.value || 4; const script = `#!/bin/bash set -Eeuo pipefail export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" hash -r INPUT_DIR="${shellEscapeDoubleQuotes(inputDir)}" CSV_MAP="${shellEscapeDoubleQuotes(csvPath)}" USE_FOLDER_NAMES=${useFolderNames} OUTPUT_DIR="$INPUT_DIR/nanocomp_output" THREADS=${threads} echo "INPUT_DIR=$INPUT_DIR" echo "CSV_MAP=$CSV_MAP" echo "USE_FOLDER_NAMES=$USE_FOLDER_NAMES" echo "OUTPUT_DIR=$OUTPUT_DIR" eval "$(conda shell.bash hook)" conda activate reads-assembly-core || { echo "❌ Failed to activate reads-assembly-core" read exit 1 } if [ ! -d "$INPUT_DIR" ]; then echo "❌ ERROR: Input directory does not exist" read exit 1 fi if [ "$USE_FOLDER_NAMES" = false ] && [ -n "$CSV_MAP" ] && [ ! -f "$CSV_MAP" ]; then echo "❌ ERROR: CSV mapping file does not exist" read exit 1 fi mkdir -p "$OUTPUT_DIR" fastq_files=() sample_names=() declare -A seen_names declare -A seen_files clean_name() { echo "$1" | sed 's/^\\xef\\xbb\\xbf//' | tr -d '\\r' | xargs } barcode_number() { echo "$1" | grep -oE '[0-9]+' | tail -n1 | sed 's/^0*//' } strip_fastq_ext() { local base="$1" base="\${base%.fastq.gz}" base="\${base%.fq.gz}" base="\${base%.fastq}" base="\${base%.fq}" echo "$base" } add_sample() { local file="$1" local name="$2" if [ -z "$file" ] || [ -z "$name" ]; then return 0 fi if [ -n "\${seen_files[$file]:-}" ]; then echo "⚠️ Skipping duplicate FASTQ file: $file" return 0 fi if [ -n "\${seen_names[$name]:-}" ]; then echo "⚠️ Skipping duplicate sample name: $name" return 0 fi fastq_files+=("$file") sample_names+=("$name") seen_files["$file"]=1 seen_names["$name"]=1 echo "✅ Added sample: $name -> $file" } find_fastq_files() { find "$INPUT_DIR" ${depthArgs} -type f \\ \$ -iname "*.fastq" -o -iname "*.fastq.gz" -o -iname "*.fq" -o -iname "*.fq.gz" \$ \\ ! -path "$OUTPUT_DIR/*" \\ -print0 | sort -z } if [ "$USE_FOLDER_NAMES" = true ] || [ -z "$CSV_MAP" ]; then echo "" echo "Using folder names for sample names..." echo "Only FASTQ files inside sample subfolders will be used." while IFS= read -r -d '' f; do parent_folder="$(basename "$(dirname "$f")")" sample_name="$parent_folder" add_sample "$f" "$sample_name" done < <(find_fastq_files) else echo "" echo "Using CSV barcode→phage map for sample names..." echo "Expected CSV format:" echo " barcode001,Olympi" echo " barcode002,Tadpole" echo "" while IFS=, read -r barcode phage extra; do barcode="$(clean_name "$barcode")" phage="$(clean_name "$phage")" if [ -z "$barcode" ] || [ -z "$phage" ]; then continue fi barcode_lc="$(echo "$barcode" | tr '[:upper:]' '[:lower:]')" phage_lc="$(echo "$phage" | tr '[:upper:]' '[:lower:]')" if [ "$barcode_lc" = "barcode" ] || [ "$phage_lc" = "phage" ] || [ "$phage_lc" = "phage name" ]; then continue fi csv_num="$(barcode_number "$barcode")" if [ -z "$csv_num" ]; then echo "⚠️ Could not find barcode number in CSV entry: $barcode" continue fi found_file="" while IFS= read -r -d '' f; do folder="$(basename "$(dirname "$f")")" filebase="$(basename "$f")" folder_num="$(barcode_number "$folder")" file_num="$(barcode_number "$filebase")" if [ "$folder_num" = "$csv_num" ] || [ "$file_num" = "$csv_num" ]; then found_file="$f" break fi done < <(find_fastq_files) if [ -n "$found_file" ]; then add_sample "$found_file" "$phage" else echo "⚠️ No FASTQ found for CSV barcode: $barcode" fi done < "$CSV_MAP" fi echo "" echo "Matched \${#fastq_files[@]} samples" if [ "\${#fastq_files[@]}" -lt 2 ]; then echo "❌ ERROR: Need at least 2 matched samples to run NanoComp" read exit 1 fi echo "" echo "Files to compare:" printf ' - %s\\n' "\${fastq_files[@]}" echo "" echo "Sample names:" printf ' - %s\\n' "\${sample_names[@]}" NanoComp --fastq "\${fastq_files[@]}" -n "\${sample_names[@]}" -o "$OUTPUT_DIR" -t "$THREADS" echo "" echo "✅ NanoComp complete" echo "📁 Results: $OUTPUT_DIR" echo "Press Enter to exit..." read `; scriptArea.value = script; } folderInput.oninput = updateScript; csvInput.oninput = updateScript; useFolderNamesInput.oninput = updateScript; recursiveInput.oninput = updateScript; threadsInput.oninput = updateScript; updateScript(); copyBtn.onclick = () => { navigator.clipboard.writeText(scriptArea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL Script"; }, 2000); }; return html`

🧬 Generate NanoComp WSL Script from FASTQ folder, optional CSV, or folder names

${folderInput} ${csvInput} ${csvHelp} ${useFolderNamesInput} ${recursiveInput} ${threadsInput}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

${scriptArea}

`; }38:T163b,viewof nanoplotScript = { const inputPath = Inputs.text({ label: "📂 Path to folder containing FASTQ files (Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\Phage_580\\Concatenated_Barcodes", style: "width: 600px" }); const threadsInput = Inputs.range([1, 32], { label: "Threads for NanoPlot", value: 4, step: 1 }); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 300px" }); const recursiveInput = Inputs.toggle({ label: "🔍 Search subfolders recursively", value: true }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy WSL Script"), 1500); }; function toWSLPath(path) { return path .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const winPath = inputPath.value.trim(); const threads = threadsInput.value || 1; const envName = envInput.value.trim() || "reads-assembly-core"; const recursive = recursiveInput.value; if (!winPath) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const basePath = toWSLPath(winPath); const findCommand = recursive ? `mapfile -d '' files < <(find "$base_dir" -type f \$ -iname "*.fastq" -o -iname "*.fq" -o -iname "*.fastq.gz" -o -iname "*.fq.gz" \$ -print0 | sort -z)` : `mapfile -d '' files < <(find "$base_dir" -maxdepth 1 -type f \$ -iname "*.fastq" -o -iname "*.fq" -o -iname "*.fastq.gz" -o -iname "*.fq.gz" \$ -print0 | sort -z)`; const scriptLines = [ "#!/bin/bash", "set -Eeuo pipefail", "", 'trap \'status=$?; echo; echo "❌ Script failed with exit code $status at line $LINENO."; echo "Press Enter to close..."; read -r _; exit $status\' ERR', "", 'export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH"', "", 'if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then', ' . "$HOME/miniconda/etc/profile.d/conda.sh"', 'elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then', ' . "$HOME/miniconda3/etc/profile.d/conda.sh"', "else", ' echo "❌ Could not find conda.sh under $HOME/miniconda or $HOME/miniconda3"', ' echo "Press Enter to close..."', " read -r _", " exit 1", "fi", "", `conda activate ${shQuote(envName)}`, "", "if ! command -v NanoPlot >/dev/null 2>&1; then", ` echo "❌ NanoPlot is not available in conda environment: ${envName}"`, ' echo "Press Enter to close..."', " read -r _", " exit 1", "fi", "", `base_dir=${shQuote(basePath)}`, 'echo "Using directory: $base_dir"', 'if [ ! -d "$base_dir" ]; then', ' echo "❌ Directory not found: $base_dir"', ' echo "Press Enter to close..."', " read -r _", " exit 1", "fi", "", 'mkdir -p "$base_dir/nanoplot_output"', "", findCommand, "", "if (( ${#files[@]} == 0 )); then", ' echo "❌ No FASTQ files found in: $base_dir"', ' echo "Press Enter to close..."', " read -r _", " exit 1", "fi", "", 'echo "Found ${#files[@]} FASTQ files:"', 'for f in "${files[@]}"; do', ' echo " - $f"', "done", "", 'for file in "${files[@]}"; do', " echo", ' echo "Processing: $file"', ' sample_noext="$(basename "$(dirname "$file")")"', ' outdir="$base_dir/nanoplot_output/$sample_noext"', ' mkdir -p "$outdir"', "", " NanoPlot \\", ' --fastq "$file" \\', ` --threads ${threads} \\`, " --loglength \\", " --N50 \\", ' --outdir "$outdir" \\', ' --prefix "${sample_noext}_"', "", ' echo "Finished NanoPlot for $file -> $outdir"', "done", "", "echo", 'echo "✅ All NanoPlot reports written under: $base_dir/nanoplot_output"', 'echo "Press Enter to close..."', "read -r _" ]; textarea.value = scriptLines.join("\n"); } inputPath.addEventListener("input", generateScript); threadsInput.addEventListener("input", generateScript); envInput.addEventListener("input", generateScript); recursiveInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ NanoPlot WSL script

${inputPath} ${threadsInput} ${envInput} ${recursiveInput}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

${textarea}

`; }39:T1a6f,viewof nanoplotSummaryScript = { const inputPath = Inputs.text({ label: "📂 Path to NanoPlot output folder (Windows path)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results\\Run1\\nanoplot_output", style: "width: 600px" }); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 320px" }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy"), 1500); }; function toWSLPath(winPath) { return (winPath || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const path = toWSLPath(inputPath.value); const envName = envInput.value.trim() || "reads-assembly-core"; if (!path) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; echo "Press Enter to close..."; read -r _; exit $status' ERR # Always keep core system tools available export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" # --- Load conda --- if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then source "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then source "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Miniconda not found" read -p "Press Enter to exit..." _ exit 1 fi conda activate ${shQuote(envName)} # Restore system PATH again in case the conda env broke it export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" echo "🐍 Python:" which python || true python -c "import pandas, xlsxwriter; print('✅ pandas + xlsxwriter ok')" BASE_DIR=${shQuote(path)} if [ ! -d "$BASE_DIR" ]; then echo "❌ Directory not found: $BASE_DIR" read -p "Press Enter to exit..." _ exit 1 fi cd "$BASE_DIR" echo "📊 Searching for NanoStats files..." mapfile -t stats_files < <(find . -type f \$ -name "NanoStats.txt" -o -name "*NanoStats.txt" \$ | sort) if (( \${#stats_files[@]} == 0 )); then echo "❌ No NanoStats files found under: $BASE_DIR" read -p "Press Enter to exit..." _ exit 1 fi echo "Found \${#stats_files[@]} NanoStats files" after_colon() { sed -E 's/^[^:]*:[[:space:]]*//; s/[[:space:]]*$//' } first_num_after_colon() { sed -E 's/^[^:]*:[[:space:]]*//; s/^[[:space:]]*//' | grep -oE '^[0-9]+(\\.[0-9]+)?' || true } echo -e "Barcode\\tReads\\tBases\\tMedianLength\\tN50\\tMeanQual\\tMedianQual\\t>Q10\\t>Q15\\t>Q20\\t>Q25\\t>Q30\\tLongest1\\tLongest2\\tLongest3\\tLongest4\\tLongest5\\tTopQual1\\tTopQual2" > barcode_summary_raw.tsv for stats in "\${stats_files[@]}"; do folder=\$(basename "\$(dirname "$stats")") barcode="$folder" reads=\$(grep -m1 -E '^Number of reads:' "$stats" | after_colon || true) bases=\$(grep -m1 -E '^Total bases:' "$stats" | after_colon || true) medianlength=\$(grep -m1 -E '^Median read length:' "$stats" | after_colon || true) n50=\$(grep -m1 -E '^Read length N50:' "$stats" | after_colon || true) meanqual=\$(grep -m1 -E '^Mean read quality:' "$stats" | after_colon || true) medianqual=\$(grep -m1 -E '^Median read quality:' "$stats" | after_colon || true) q10=\$(grep -m1 -E '^>Q10:' "$stats" | first_num_after_colon) q15=\$(grep -m1 -E '^>Q15:' "$stats" | first_num_after_colon) q20=\$(grep -m1 -E '^>Q20:' "$stats" | first_num_after_colon) q25=\$(grep -m1 -E '^>Q25:' "$stats" | first_num_after_colon) q30=\$(grep -m1 -E '^>Q30:' "$stats" | first_num_after_colon) high1=\$(awk 'BEGIN{flag=0} /^Top 5 highest mean basecall quality/{flag=1; next} flag && /^1:/{sub(/^1:[ \\t]*/,""); print; exit}' "$stats" || true) high2=\$(awk 'BEGIN{flag=0} /^Top 5 highest mean basecall quality/{flag=1; next} flag && /^2:/{sub(/^2:[ \\t]*/,""); print; exit}' "$stats" || true) long1=\$(awk 'BEGIN{flag=0} /^Top 5 longest reads/{flag=1; next} flag && /^1:/{sub(/^1:[ \\t]*/,""); print; exit}' "$stats" || true) long2=\$(awk 'BEGIN{flag=0} /^Top 5 longest reads/{flag=1; next} flag && /^2:/{sub(/^2:[ \\t]*/,""); print; exit}' "$stats" || true) long3=\$(awk 'BEGIN{flag=0} /^Top 5 longest reads/{flag=1; next} flag && /^3:/{sub(/^3:[ \\t]*/,""); print; exit}' "$stats" || true) long4=\$(awk 'BEGIN{flag=0} /^Top 5 longest reads/{flag=1; next} flag && /^4:/{sub(/^4:[ \\t]*/,""); print; exit}' "$stats" || true) long5=\$(awk 'BEGIN{flag=0} /^Top 5 longest reads/{flag=1; next} flag && /^5:/{sub(/^5:[ \\t]*/,""); print; exit}' "$stats" || true) echo -e "\${barcode}\\t\${reads}\\t\${bases}\\t\${medianlength}\\t\${n50}\\t\${meanqual}\\t\${medianqual}\\t\${q10}\\t\${q15}\\t\${q20}\\t\${q25}\\t\${q30}\\t\${long1}\\t\${long2}\\t\${long3}\\t\${long4}\\t\${long5}\\t\${high1}\\t\${high2}" >> barcode_summary_raw.tsv done echo "📊 Creating Excel summary..." python - <<'PY' import pandas as pd df = pd.read_csv("barcode_summary_raw.tsv", sep="\\t") with pd.ExcelWriter("barcode_summary.xlsx", engine="xlsxwriter") as writer: df.to_excel(writer, index=False, sheet_name="Summary") print("✅ Done! Summary saved to barcode_summary.xlsx") PY echo echo "📁 Output files:" ls -lh barcode_summary_raw.tsv barcode_summary.xlsx || true echo echo "✅ Done." echo "Press Enter to exit..." read -r _ `; } inputPath.addEventListener("input", generateScript); envInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ NanoPlot summary WSL script

${inputPath} ${envInput}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

${textarea}

`; }3a:T1d4e,viewof filtlongConfigurator = { const folderPathInput = Inputs.text({ label: "📁 Path to parent folder containing FASTQ files or subfolders (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results\\Run1\\Demux_Output", style: "width: 600px" }); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 320px" }); const iterateSubfoldersInput = Inputs.toggle({ label: "🔁 Iterate immediate subfolders of parent folder", value: true }); const dedupeInput = Inputs.toggle({ label: "🧹 Remove duplicate read names before Filtlong", value: true }); const minReadLengthInput = Inputs.number({ label: "📏 Min Read Length (bp)", value: 200, min: 50, max: 10000, step: 50, style: "width: 300px" }); const minQualityInput = Inputs.number({ label: "🔎 Min Mean Quality (--min_mean_q)", value: 10, min: 0, max: 40, step: 1, style: "width: 300px" }); const targetBasesInput = Inputs.text({ label: "🧮 Max Bases (--target_bases) — leave blank to omit", placeholder: "6000000", style: "width: 300px" }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied"; setTimeout(() => (copyBtn.textContent = "📋 Copy"), 2000); }; function toWSLPath(winPath) { return (winPath || "") .trim() .replace(/^([A-Za-z]):/, (_, drive) => `/mnt/${drive.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function formatMB(value) { const n = Number(value); return isNaN(n) ? null : `${Math.round(n / 1_000_000)}Mb`; } function generateCommand() { const minLength = minReadLengthInput.value || 200; const minQuality = minQualityInput.value || 10; const envName = envInput.value.trim() || "reads-assembly-core"; const winPath = folderPathInput.value.trim(); const iterateSubfolders = iterateSubfoldersInput.value; const dedupeReads = dedupeInput.value; const targetBasesRaw = targetBasesInput.value.trim(); const hasTargetBases = targetBasesRaw !== ""; const targetBases = Number(targetBasesRaw); const mbFormatted = formatMB(targetBases); if (!winPath) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const wslPath = toWSLPath(winPath); const baseFolderName = `FiltLong_MinRead-${minLength}bp_Q${minQuality}`; const outputFolderName = hasTargetBases && !isNaN(targetBases) && mbFormatted ? `${baseFolderName}_Bases-${mbFormatted}` : baseFolderName; const filtlongParams = [ `--min_length ${minLength}`, `--min_mean_q ${minQuality}` ]; if (hasTargetBases && !isNaN(targetBases)) { filtlongParams.unshift(`--target_bases ${targetBases}`); } textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; echo "Press Enter to close..."; read -r _; exit $status' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" BASE_DIR=${shQuote(wslPath)} ENV_NAME=${shQuote(envName)} ITERATE_SUBFOLDERS=${iterateSubfolders ? 1 : 0} DEDUPLICATE_READS=${dedupeReads ? 1 : 0} OUTPUT_FOLDER_NAME=${shQuote(outputFolderName)} if [ ! -d "$BASE_DIR" ]; then echo "❌ Directory not found: $BASE_DIR" read -r _ exit 1 fi if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Could not find conda.sh" read -r _ exit 1 fi conda activate "$ENV_NAME" export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" # Prevent Dorado's bundled libraries from breaking filtlong unset LD_LIBRARY_PATH if ! command -v filtlong >/dev/null 2>&1; then echo "❌ filtlong not found in environment: $ENV_NAME" read -r _ exit 1 fi PARENT_OUT_DIR="$BASE_DIR/$OUTPUT_FOLDER_NAME" mkdir -p "$PARENT_OUT_DIR" dedupe_fastq() { local input="$1" local output="$2" awk ' NR % 4 == 1 { header=$0 id=$1 sub(/^@/, "", id) getline seq getline plus getline qual if (!(id in seen)) { seen[id]=1 print header print seq print plus print qual } } ' "$input" > "$output" } process_folder() { local folder="$1" local folder_name local out_dir folder_name="$(basename "$folder")" out_dir="$PARENT_OUT_DIR/$folder_name" mkdir -p "$out_dir" shopt -s nullglob local files=( "$folder"/*.fastq ) shopt -u nullglob if (( \${#files[@]} == 0 )); then echo "⚠ No .fastq files found in: $folder" return 0 fi echo echo "📁 Processing folder: $folder" echo "📁 Output folder: $out_dir" for file in "\${files[@]}"; do local bn out_file input_for_filtlong tmp_dedup bn="$(basename "$file")" out_file="$out_dir/filtered_$bn" if [ "$DEDUPLICATE_READS" -eq 1 ]; then tmp_dedup="$out_dir/dedup_input_$bn" echo "🧹 Deduplicating read names in $bn" dedupe_fastq "$file" "$tmp_dedup" input_for_filtlong="$tmp_dedup" else input_for_filtlong="$file" fi echo "🔍 Filtlong processing $bn" filtlong \\ ${filtlongParams.join(" \\\n ")} \\ "$input_for_filtlong" > "$out_file" rm -f "\${tmp_dedup:-}" echo "✅ Wrote: $out_file" done } if [ "$ITERATE_SUBFOLDERS" -eq 1 ]; then echo "🔁 Iterating immediate subfolders of: $BASE_DIR" for folder in "$BASE_DIR"/*/; do [ -d "$folder" ] || continue folder_clean="\${folder%/}" if [ "$folder_clean" = "$PARENT_OUT_DIR" ]; then continue fi process_folder "$folder_clean" done else process_folder "$BASE_DIR" fi echo echo "✅ Done." echo "📁 Parent output folder: $PARENT_OUT_DIR" echo "Press Enter to close..." read -r _ `; } [ folderPathInput, envInput, iterateSubfoldersInput, dedupeInput, minReadLengthInput, minQualityInput, targetBasesInput ].forEach((el) => el.addEventListener("input", generateCommand)); generateCommand(); return html`

⬇ Filtlong filtering WSL script

${folderPathInput} ${envInput} ${iterateSubfoldersInput} ${dedupeInput} ${minReadLengthInput} ${minQualityInput} ${targetBasesInput}

Save as run_filtlong.sh, then run in WSL.

${copyBtn}

${textarea}

`; }3b:T1a09,viewof chopperScriptGenerator = { const fastqFolderInput = Inputs.text({ label: "📂 Path to folder with FASTQ files", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results\\Demux", style: "width: 600px" }); const filePatternInput = Inputs.text({ label: "FASTQ filename pattern", value: "*.fastq", style: "width: 420px" }); const recursiveInput = Inputs.toggle({ label: "🔍 Search subfolders recursively", value: true }); const recreateFoldersInput = Inputs.toggle({ label: "🔁 Recreate folder structure inside Chopper output", value: true }); const options = { quality: Inputs.range([0, 30], { label: "Min Q-score (-q)", step: 1, value: 10 }), minLength: Inputs.range([0, 10000], { label: "Min read length (-l)", step: 100, value: 300 }), maxLength: Inputs.range([0, 100000], { label: "Max read length (--maxlength)", step: 500, value: 50000 }), threads: Inputs.range([1, 64], { label: "Threads (-t)", step: 1, value: 4 }) }; const outputBox = html``; const copyButton = html``; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } const updateScript = () => { const fastqPath = toWSLPath(fastqFolderInput.value.trim()); const filePattern = filePatternInput.value.trim() || "*.fastq"; const recursive = recursiveInput.value ? 1 : 0; const recreateFolders = recreateFoldersInput.value ? 1 : 0; const filteredFolder = `Chopper_Filtered_Q${options.quality.value}_L${options.minLength.value}_Max${options.maxLength.value}`; const script = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; echo "Press Enter to close..."; read -r _; exit $status' ERR INPUT_DIR=${shQuote(fastqPath)} FILE_PATTERN=${shQuote(filePattern)} OUTPUT_ROOT="$INPUT_DIR/${filteredFolder}" RECURSIVE=${recursive} RECREATE_FOLDERS=${recreateFolders} if [ ! -d "$INPUT_DIR" ]; then echo "❌ Folder not found: $INPUT_DIR" echo "Press Enter to close..." read -r _ exit 1 fi if ! command -v chopper >/dev/null 2>&1; then echo "❌ chopper not found. Activate your conda environment first." echo "Press Enter to close..." read -r _ exit 1 fi mkdir -p "$OUTPUT_ROOT" echo "📂 Input folder: $INPUT_DIR" echo "📂 Output root: $OUTPUT_ROOT" echo "🔎 File pattern: $FILE_PATTERN" echo "🔁 Recreate folder structure: $RECREATE_FOLDERS" echo "🔍 Searching for FASTQ files..." if [ "$RECURSIVE" -eq 1 ]; then mapfile -d '' files < <( find "$INPUT_DIR" -type f \\ -iname "$FILE_PATTERN" \\ ! -path "$OUTPUT_ROOT/*" \\ ! -path "*/Chopper_Filtered_*/*" \\ ! -path "*/nanocomp_output/*" \\ -print0 | sort -z ) else mapfile -d '' files < <( find "$INPUT_DIR" -maxdepth 1 -type f \\ -iname "$FILE_PATTERN" \\ -print0 | sort -z ) fi echo "✅ Found \${#files[@]} FASTQ file(s)." if (( \${#files[@]} == 0 )); then echo echo "❌ No matching FASTQ files found." echo "Try pattern: *.fastq" echo "Or enable recursive search if files are inside barcode folders." echo "Press Enter to close..." read -r _ exit 1 fi for file in "\${files[@]}"; do base="$(basename "$file")" name="$base" name="\${name%.fastq.gz}" name="\${name%.fq.gz}" name="\${name%.fastq}" name="\${name%.fq}" barcode="$(echo "$name" | grep -oE 'barcode[0-9]+' | head -n1 || true)" if [ "$RECREATE_FOLDERS" -eq 1 ]; then rel_dir="$(dirname "\${file#"$INPUT_DIR"/}")" if [ "$rel_dir" = "." ]; then out_dir="$OUTPUT_ROOT" else out_dir="$OUTPUT_ROOT/$rel_dir" fi else out_dir="$OUTPUT_ROOT" fi mkdir -p "$out_dir" if [ -n "$barcode" ]; then out_file="$out_dir/filtered_\${barcode}.fastq" else out_file="$out_dir/filtered_\${name}.fastq" fi echo echo "🔍 Processing: $file" echo "➡ Output: $out_file" if [[ "$file" == *.gz ]]; then gzip -cd "$file" | chopper \\ -q ${options.quality.value} \\ -l ${options.minLength.value} \\ --maxlength ${options.maxLength.value} \\ -t ${options.threads.value} \\ > "$out_file" else chopper \\ -i "$file" \\ -q ${options.quality.value} \\ -l ${options.minLength.value} \\ --maxlength ${options.maxLength.value} \\ -t ${options.threads.value} \\ > "$out_file" fi if [ ! -s "$out_file" ]; then echo "⚠ Output file is empty: $out_file" else echo "✅ Finished: $out_file" fi done echo echo "✅ Done filtering with Chopper." echo "📁 Files saved in: $OUTPUT_ROOT" echo "Press Enter to close..." read -r _ `; outputBox.value = script; }; fastqFolderInput.addEventListener("input", updateScript); filePatternInput.addEventListener("input", updateScript); recursiveInput.addEventListener("input", updateScript); recreateFoldersInput.addEventListener("input", updateScript); options.quality.addEventListener("input", updateScript); options.minLength.addEventListener("input", updateScript); options.maxLength.addEventListener("input", updateScript); options.threads.addEventListener("input", updateScript); copyButton.onclick = () => { navigator.clipboard.writeText(outputBox.value); copyButton.textContent = "✅ Copied"; setTimeout(() => (copyButton.textContent = "📋 Copy"), 2000); }; updateScript(); return html`

⬇ Chopper filtering WSL script

${fastqFolderInput} ${filePatternInput} ${recursiveInput} ${recreateFoldersInput} ${options.quality} ${options.minLength} ${options.maxLength} ${options.threads} ${copyButton}

${outputBox}

`; }3c:Tab5,viewof flyeChecker = { const releasesUrl = "https://github.com/mikolmogorov/Flye/releases/latest"; const commitUrl = "https://api.github.com/repos/mikolmogorov/Flye/commits/flye"; const buttonStyle = ` padding: 0.5em 0.85em; border: 1px solid #bbb; border-radius: 6px; background: #f7f7f7; cursor: pointer; font-size: 0.9em; `; const latest = html`Not checked`; const checkBtn = html``; const openBtn = html``; const copyBtn = html``; const cmd = `conda activate reads-assembly-core flye --version`; const codeBox = html`

${cmd}

`; checkBtn.onclick = async () => { latest.textContent = "checking..."; try { const releaseResp = await fetch( "https://api.github.com/repos/mikolmogorov/Flye/releases/latest" ); const releaseData = await releaseResp.json(); const commitResp = await fetch(commitUrl); const commitData = await commitResp.json(); const tag = releaseData.tag_name || "unknown"; const sha = commitData.sha ? commitData.sha.slice(0, 7) : "unknown"; latest.textContent = `${tag}, commit ${sha}`; } catch (err) { latest.textContent = "check failed"; } }; openBtn.onclick = () => { window.open(releasesUrl, "_blank", "noopener,noreferrer"); }; copyBtn.onclick = () => { navigator.clipboard.writeText(cmd); copyBtn.textContent = "✅ Copied"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL code"; }, 1200); }; return html`

Flye version checker

Use the latest Flye version for improved Nanopore assembly performance.
Latest Flye GitHub release/commit: ${latest}

${checkBtn} ${openBtn}

To check your installed Flye version/build, copy and run the code below in WSL:

${copyBtn} ${codeBox}

`; }3d:Ta4d,viewof flyeLatestInstaller = { const envInput = Inputs.text({ label: "🧪 Conda environment name", value: "reads-assembly-core", style: "width: 320px" }); const textarea = html``; const copyBtn = html``; function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const envName = envInput.value.trim() || "reads-assembly-core"; const innerScript = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Flye install failed at line $LINENO (exit $status)"; echo "Press Enter to close..."; read -r _; exit $status' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:\${PATH:-}" CONDA_ENV=${shQuote(envName)} if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Could not find conda.sh" echo "Press Enter to close..." read -r _ exit 1 fi conda activate "$CONDA_ENV" echo "🧹 Removing pip-installed Flye, if present..." python -m pip uninstall -y flye || true echo "🧹 Removing existing conda Flye, if present..." conda remove -y flye || true echo "📦 Installing Flye from Bioconda..." conda install -y -c conda-forge -c bioconda flye echo echo "✅ Testing Flye:" which flye which flye-modules flye --version flye-modules polisher -h >/dev/null echo echo "✅ Flye install looks good." echo "Press Enter to close..." read -r _`; textarea.value = `bash <<'FLYE_INSTALLER' ${innerScript} FLYE_INSTALLER`; } copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy Flye Installer"), 1500); }; envInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ Install/update Flye from Bioconda

${envInput}

Copy this block, paste it directly into WSL, and press Enter.

${textarea} ${copyBtn}

`; }3e:T2066,viewof flyeScriptGenerator = { const inputPath = Inputs.text({ label: "📂 Path to folder containing FASTQ files (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results\\Run1\\Demux_Output", style: "width: 600px" }); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 320px" }); const readTypeInput = Inputs.select(["--nano-hq", "--nano-raw"], { label: "🧬 Flye read type", value: "--nano-hq", style: "width: 220px" }); const genomeSizeInput = Inputs.text({ label: "🧫 Expected genome size", value: "50k", placeholder: "50k", style: "width: 220px" }); const flyeIterationsInput = Inputs.number({ label: "🔁 Flye polishing iterations", value: 2, min: 0, step: 1, style: "width: 220px" }); const asmCoverageInput = Inputs.number({ label: "📉 Flye asm coverage (0 = off)", value: 0, step: 10, style: "width: 220px" }); const threads = Inputs.range([1, 32], { label: "🧵 Threads", value: 4, step: 1 }); const minOverlap = Inputs.number({ label: "🔗 Flye min overlap (0 = auto)", value: 0, step: 100 }); const recursiveInput = Inputs.toggle({ label: "🔍 Search subfolders", value: true }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy"), 1500); }; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const basePath = toWSLPath(inputPath.value.trim()); if (!basePath) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const envName = envInput.value.trim() || "reads-assembly-core"; const readType = readTypeInput.value || "--nano-hq"; const genomeSize = genomeSizeInput.value.trim() || "50k"; const flyeIterationsVal = Math.max( 0, Math.floor(Number(flyeIterationsInput.value) || 0) ); const asmCoverageVal = Number(asmCoverageInput.value) || 0; const nThreads = threads.value || 1; const overlapVal = minOverlap.value || 0; const findCommand = recursiveInput.value ? `mapfile -d '' files < <(find "$BASE_DIR" -type f \$ -iname "*.fastq" -o -iname "*.fq" -o -iname "*.fastq.gz" -o -iname "*.fq.gz" \$ -print0 | sort -z)` : `mapfile -d '' files < <(find "$BASE_DIR" -maxdepth 1 -type f \$ -iname "*.fastq" -o -iname "*.fq" -o -iname "*.fastq.gz" -o -iname "*.fq.gz" \$ -print0 | sort -z)`; textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO"; echo "Press Enter to close..."; read -r _; exit $status' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" if [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" else echo "❌ Conda not found" read -r _ exit 1 fi conda activate ${shQuote(envName)} for tool in flye awk; do command -v "$tool" >/dev/null 2>&1 || { echo "❌ Missing tool: $tool" read -r _ exit 1 } done THREADS=${nThreads} MIN_OVERLAP=${overlapVal} GENOME_SIZE=${shQuote(genomeSize)} FLYE_ITERATIONS=${flyeIterationsVal} ASM_COVERAGE=${asmCoverageVal} BASE_DIR=${shQuote(basePath)} FLYE_READ_TYPE=${shQuote(readType)} if [ ! -d "$BASE_DIR" ]; then echo "❌ Folder not found: $BASE_DIR" read -r _ exit 1 fi write_fasta_info() { local fasta="$1" local out="$2" local stage="$3" if [ ! -f "$fasta" ]; then echo "⚠️ FASTA not found for info summary: $fasta" return 0 fi awk -v stage="$stage" ' BEGIN { print "stage\\tcontig\\tlength_bp\\tgc_percent" name = "" seq = "" } /^>/ { if (name != "") { len = length(seq) tmp = seq gc = gsub(/[GgCc]/, "", tmp) printf "%s\\t%s\\t%d\\t%.2f\\n", stage, name, len, (len ? gc * 100 / len : 0) } name = $0 sub(/^>/, "", name) sub(/[ \\t].*/, "", name) seq = "" next } { gsub(/[ \\t\\r\\n]/, "", $0) seq = seq $0 } END { if (name != "") { len = length(seq) tmp = seq gc = gsub(/[GgCc]/, "", tmp) printf "%s\\t%s\\t%d\\t%.2f\\n", stage, name, len, (len ? gc * 100 / len : 0) } } ' "$fasta" > "$out" } mkdir -p "$BASE_DIR/assemblies" cd "$BASE_DIR" echo "▶ Running Flye assembly pipeline..." echo "Flye read type: $FLYE_READ_TYPE" echo "Expected genome size: $GENOME_SIZE" echo "Flye polishing iterations: $FLYE_ITERATIONS" if [[ "$ASM_COVERAGE" -gt 0 ]]; then echo "Flye asm coverage: $ASM_COVERAGE" else echo "Flye asm coverage: off" fi ${findCommand} if (( \${#files[@]} == 0 )); then echo "❌ No FASTQ files found" read -r _ exit 1 fi for file in "\${files[@]}"; do echo sample="$(basename "$file")" sample="\${sample%.fastq.gz}" sample="\${sample%.fq.gz}" sample="\${sample%.fastq}" sample="\${sample%.fq}" sample="\${sample//[^A-Za-z0-9._-]/_}" sample_dir="$BASE_DIR/assemblies/$sample" mkdir -p "$sample_dir" echo "🧬 Assembling $sample" echo "Input reads: $file" flye_args=( "$FLYE_READ_TYPE" "$file" --genome-size "$GENOME_SIZE" --iterations "$FLYE_ITERATIONS" -o "$sample_dir" --threads "$THREADS" ) if [[ "$MIN_OVERLAP" -ge 1000 ]]; then flye_args+=(--min-overlap "$MIN_OVERLAP") fi if [[ "$ASM_COVERAGE" -gt 0 ]]; then flye_args+=(--asm-coverage "$ASM_COVERAGE") fi flye "\${flye_args[@]}" || continue assembly="$sample_dir/assembly.fasta" final="$sample_dir/\${sample}_assembly.fasta" if [ ! -f "$assembly" ]; then echo "⚠️ Flye finished, but assembly.fasta was not found for $sample" continue fi mv "$assembly" "$final" if [ -f "$sample_dir/assembly_info.txt" ]; then cp "$sample_dir/assembly_info.txt" "$sample_dir/\${sample}_flye_assembly_info.txt" fi write_fasta_info "$final" "$sample_dir/\${sample}_flye_fasta_info.tsv" "flye" echo echo "📄 Flye FASTA summary:" cat "$sample_dir/\${sample}_flye_fasta_info.tsv" echo echo "✅ Done $sample" echo "Final Flye-polished FASTA:" echo "$final" done echo echo "✅ All assemblies complete → $BASE_DIR/assemblies" echo "Press Enter to close..." read -r _ `; } inputPath.addEventListener("input", generateScript); envInput.addEventListener("input", generateScript); readTypeInput.addEventListener("input", generateScript); genomeSizeInput.addEventListener("input", generateScript); flyeIterationsInput.addEventListener("input", generateScript); asmCoverageInput.addEventListener("input", generateScript); threads.addEventListener("input", generateScript); minOverlap.addEventListener("input", generateScript); recursiveInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ Flye assembly WSL script

${inputPath} ${envInput} ${readTypeInput} ${genomeSizeInput} ${flyeIterationsInput} ${asmCoverageInput} ${threads} ${minOverlap} ${recursiveInput}

Copy the generated script into a file such as run_flye.sh, then run it in WSL.

${textarea} ${copyBtn}

`; }3f:T4159,viewof flyeScriptGeneratorThreads = { const inputPath = Inputs.text({ label: "📂 Path to folder containing FASTQ files (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results\\Run1\\Demux_Output", style: "width: 600px" }); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: "width: 320px" }); const readTypeInput = Inputs.select(["--nano-hq", "--nano-raw"], { label: "🧬 Flye read type", value: "--nano-hq", style: "width: 220px" }); const genomeSizeInput = Inputs.text({ label: "🧫 Expected genome size", value: "50k", placeholder: "50k", style: "width: 220px" }); const flyeIterationsInput = Inputs.number({ label: "🔁 Flye polishing iterations", value: 2, min: 0, step: 1, style: "width: 220px" }); const asmCoverageInput = Inputs.number({ label: "📉 Flye asm coverage (0 = off)", value: 0, min: 0, step: 10, style: "width: 220px" }); const threadsPerAssemblyInput = Inputs.range([1, 32], { label: "🧵 Threads per Flye assembly", value: 4, step: 1 }); const parallelAssembliesInput = Inputs.range([1, 16], { label: "🚀 Simultaneous Flye assemblies", value: 4, step: 1 }); const minOverlap = Inputs.number({ label: "🔗 Flye min overlap (0 = auto)", value: 0, min: 0, step: 100 }); const recursiveInput = Inputs.toggle({ label: "🔍 Search subfolders", value: true }); const cpuSummary = html`

`; const textarea = html``; const copyBtn = html``; copyBtn.onclick = async () => { await navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL script"; }, 1500); }; function toWSLPath(path) { return (path || "") .trim() .replace( /^([A-Za-z]):/, (_, driveLetter) => `/mnt/${driveLetter.toLowerCase()}` ) .replace(/\\/g, "/"); } function shQuote(value) { return `'${String(value).replace(/'/g, `'\\''`)}'`; } function updateCpuSummary() { const threadsPerAssembly = Math.max( 1, Math.floor(Number(threadsPerAssemblyInput.value) || 1) ); const parallelAssemblies = Math.max( 1, Math.floor(Number(parallelAssembliesInput.value) || 1) ); const maximumThreads = threadsPerAssembly * parallelAssemblies; cpuSummary.innerHTML = ` Maximum Flye CPU use: ${parallelAssemblies} simultaneous assemblies × ${threadsPerAssembly} threads each = ${maximumThreads} CPU threads
Try to keep this value at or below the number of logical CPU threads available to WSL. `; } function generateScript() { const basePath = toWSLPath(inputPath.value.trim()); updateCpuSummary(); if (!basePath) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const envName = envInput.value.trim() || "reads-assembly-core"; const readType = readTypeInput.value || "--nano-hq"; const genomeSize = genomeSizeInput.value.trim() || "50k"; const flyeIterationsVal = Math.max( 0, Math.floor(Number(flyeIterationsInput.value) || 0) ); const asmCoverageVal = Math.max(0, Number(asmCoverageInput.value) || 0); const threadsPerAssemblyVal = Math.max( 1, Math.floor(Number(threadsPerAssemblyInput.value) || 1) ); const parallelAssembliesVal = Math.max( 1, Math.floor(Number(parallelAssembliesInput.value) || 1) ); const overlapVal = Math.max(0, Math.floor(Number(minOverlap.value) || 0)); const findCommand = recursiveInput.value ? `mapfile -d '' files < <( find "$BASE_DIR" \ -type f \ \$ \ -iname "*.fastq" \ -o -iname "*.fq" \ -o -iname "*.fastq.gz" \ -o -iname "*.fq.gz" \ \$ \ -print0 | sort -z )` : `mapfile -d '' files < <( find "$BASE_DIR" \ -maxdepth 1 \ -type f \ \$ \ -iname "*.fastq" \ -o -iname "*.fq" \ -o -iname "*.fastq.gz" \ -o -iname "*.fq.gz" \ \$ \ -print0 | sort -z )`; textarea.value = `#!/usr/bin/env bash # ============================================================ # Parallel Flye assembly pipeline for WSL # # Several independent FASTQ files can be assembled at the # same time. Each Flye process receives its own CPU threads. # ============================================================ set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; exit $status' ERR export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" pause_before_exit() { echo echo "Press Enter to close..." read -r _ /dev/null 2>&1; then eval "$(conda shell.bash hook)" else echo "❌ Conda was not found." pause_before_exit exit 1 fi conda activate ${shQuote(envName)} # ------------------------------------------------------------ # Verify required programs # ------------------------------------------------------------ for tool in flye awk find sort nproc tail; do command -v "$tool" >/dev/null 2>&1 || { echo "❌ Missing required tool: $tool" pause_before_exit exit 1 } done # ------------------------------------------------------------ # User settings # ------------------------------------------------------------ THREADS_PER_ASSEMBLY=${threadsPerAssemblyVal} MAX_PARALLEL_ASSEMBLIES=${parallelAssembliesVal} MIN_OVERLAP=${overlapVal} GENOME_SIZE=${shQuote(genomeSize)} FLYE_ITERATIONS=${flyeIterationsVal} ASM_COVERAGE=${asmCoverageVal} BASE_DIR=${shQuote(basePath)} FLYE_READ_TYPE=${shQuote(readType)} if [ ! -d "$BASE_DIR" ]; then echo "❌ Folder not found:" echo " $BASE_DIR" pause_before_exit exit 1 fi # ------------------------------------------------------------ # Create FASTA information table # ------------------------------------------------------------ write_fasta_info() { local fasta="$1" local out="$2" local stage="$3" if [ ! -f "$fasta" ]; then echo "⚠️ FASTA not found for information summary:" echo " $fasta" return 0 fi awk -v stage="$stage" ' BEGIN { print "stage\\tcontig\\tlength_bp\\tgc_percent" name = "" seq = "" } /^>/ { if (name != "") { len = length(seq) tmp = seq gc = gsub(/[GgCc]/, "", tmp) printf "%s\\t%s\\t%d\\t%.2f\\n", stage, name, len, (len ? gc * 100 / len : 0) } name = $0 sub(/^>/, "", name) sub(/[ \\t].*/, "", name) seq = "" next } { gsub(/[ \\t\\r\\n]/, "", $0) seq = seq $0 } END { if (name != "") { len = length(seq) tmp = seq gc = gsub(/[GgCc]/, "", tmp) printf "%s\\t%s\\t%d\\t%.2f\\n", stage, name, len, (len ? gc * 100 / len : 0) } } ' "$fasta" > "$out" } # ------------------------------------------------------------ # Generate a unique sample name # # The relative folder path is included in the sample name. # This prevents two files with the same filename in different # folders from overwriting one another. # # Example: # barcode01/filtered.fastq.gz # # becomes: # barcode01__filtered # ------------------------------------------------------------ make_sample_name() { local file="$1" local relative local sample if [[ "$file" == "$BASE_DIR/"* ]]; then relative="\${file#"$BASE_DIR"/}" else relative="$(basename "$file")" fi sample="$relative" sample="\${sample%.fastq.gz}" sample="\${sample%.fq.gz}" sample="\${sample%.fastq}" sample="\${sample%.fq}" sample="\${sample//\\//__}" sample="\${sample//[^A-Za-z0-9._-]/_}" printf '%s\\n' "$sample" } # ------------------------------------------------------------ # Assemble one FASTQ file # # Each call to this function is run as a separate background # job. Flye itself uses THREADS_PER_ASSEMBLY CPU threads. # ------------------------------------------------------------ process_file() { local file="$1" local sample local sample_dir local assembly local final local log_file local fasta_info local -a flye_args sample="$(make_sample_name "$file")" sample_dir="$BASE_DIR/assemblies/$sample" assembly="$sample_dir/assembly.fasta" final="$sample_dir/\${sample}_assembly.fasta" log_file="$sample_dir/\${sample}_flye_console.log" fasta_info="$sample_dir/\${sample}_flye_fasta_info.tsv" # Skip assemblies that were already completed successfully. if [ -s "$final" ]; then echo "⏭️ Skipping completed sample: $sample" return 0 fi mkdir -p "$sample_dir" echo echo "🚀 Starting: $sample" echo " Input: $file" echo " Threads: $THREADS_PER_ASSEMBLY" echo " Log: $log_file" flye_args=( "$FLYE_READ_TYPE" "$file" --genome-size "$GENOME_SIZE" --iterations "$FLYE_ITERATIONS" -o "$sample_dir" --threads "$THREADS_PER_ASSEMBLY" ) if [[ "$MIN_OVERLAP" -ge 1000 ]]; then flye_args+=( --min-overlap "$MIN_OVERLAP" ) fi if [[ "$ASM_COVERAGE" -gt 0 ]]; then flye_args+=( --asm-coverage "$ASM_COVERAGE" ) fi # Send the detailed Flye output to a separate sample log. # This prevents output from simultaneous assemblies from # becoming mixed together in the terminal. if ! flye "\${flye_args[@]}" > "$log_file" 2>&1; then echo echo "❌ Flye failed: $sample" echo " Log file:" echo " $log_file" echo echo " Last 20 log lines:" tail -n 20 "$log_file" || true return 1 fi if [ ! -f "$assembly" ]; then echo echo "❌ Flye finished, but assembly.fasta was not found:" echo " $sample" return 1 fi mv "$assembly" "$final" if [ -f "$sample_dir/assembly_info.txt" ]; then cp \ "$sample_dir/assembly_info.txt" \ "$sample_dir/\${sample}_flye_assembly_info.txt" fi write_fasta_info \ "$final" \ "$fasta_info" \ "flye" echo echo "✅ Finished: $sample" echo " FASTA:" echo " $final" echo echo " FASTA summary:" cat "$fasta_info" } # ------------------------------------------------------------ # Prepare output folder # ------------------------------------------------------------ mkdir -p "$BASE_DIR/assemblies" cd "$BASE_DIR" available_threads="$(nproc)" requested_threads=$( ( THREADS_PER_ASSEMBLY * MAX_PARALLEL_ASSEMBLIES ) ) echo echo "============================================================" echo " Parallel Flye assembly pipeline" echo "============================================================" echo echo "Base folder:" echo " $BASE_DIR" echo echo "Logical CPU threads available to WSL:" echo " $available_threads" echo echo "Simultaneous Flye assemblies:" echo " $MAX_PARALLEL_ASSEMBLIES" echo echo "Threads per Flye assembly:" echo " $THREADS_PER_ASSEMBLY" echo echo "Maximum requested Flye CPU threads:" echo " $requested_threads" echo echo "Flye read type:" echo " $FLYE_READ_TYPE" echo echo "Expected genome size:" echo " $GENOME_SIZE" echo echo "Flye polishing iterations:" echo " $FLYE_ITERATIONS" if [[ "$ASM_COVERAGE" -gt 0 ]]; then echo echo "Flye asm coverage:" echo " $ASM_COVERAGE" else echo echo "Flye asm coverage:" echo " off" fi if [[ "$MIN_OVERLAP" -ge 1000 ]]; then echo echo "Flye minimum overlap:" echo " $MIN_OVERLAP" else echo echo "Flye minimum overlap:" echo " automatic" fi if (( requested_threads > available_threads )); then echo echo "⚠️ WARNING:" echo "The requested Flye CPU threads exceed the logical CPU" echo "threads currently available to WSL." echo echo "Requested: $requested_threads" echo "Available: $available_threads" echo echo "The script can still run, but performance may decrease" echo "because the CPU is oversubscribed." fi # ------------------------------------------------------------ # Find FASTQ files # ------------------------------------------------------------ ${findCommand} if (( \${#files[@]} == 0 )); then echo echo "❌ No FASTQ or FASTQ.GZ files were found." pause_before_exit exit 1 fi echo echo "FASTQ files found:" echo " \${#files[@]}" echo echo "The script will run up to" echo "$MAX_PARALLEL_ASSEMBLIES assemblies simultaneously." echo # ------------------------------------------------------------ # Run the Flye assemblies in parallel # # The script launches new samples until the maximum number of # simultaneous assemblies is reached. It then waits for one # assembly to finish before launching the next sample. # ------------------------------------------------------------ running=0 completed_count=0 failed_count=0 for file in "\${files[@]}"; do while (( running >= MAX_PARALLEL_ASSEMBLIES )); do if wait -n; then completed_count=$((completed_count + 1)) else failed_count=$((failed_count + 1)) fi running=$((running - 1)) done process_file "$file" & running=$((running + 1)) done # Wait for the remaining background assemblies. while (( running > 0 )); do if wait -n; then completed_count=$((completed_count + 1)) else failed_count=$((failed_count + 1)) fi running=$((running - 1)) done echo echo "============================================================" echo "✅ Parallel Flye processing complete" echo "============================================================" echo echo "Successful or previously completed:" echo " $completed_count" echo echo "Failed:" echo " $failed_count" echo echo "Output folder:" echo " $BASE_DIR/assemblies" echo if (( failed_count > 0 )); then echo "⚠️ One or more samples failed." echo echo "Each assembly folder contains a Flye log named:" echo " *_flye_console.log" echo echo "Review these log files to identify the failed samples." fi pause_before_exit `; } const controls = [ inputPath, envInput, readTypeInput, genomeSizeInput, flyeIterationsInput, asmCoverageInput, threadsPerAssemblyInput, parallelAssembliesInput, minOverlap, recursiveInput ]; for (const control of controls) { control.addEventListener("input", generateScript); } generateScript(); return html`

⬇ Parallel Flye assembly WSL script

${inputPath} ${envInput} ${readTypeInput} ${genomeSizeInput} ${flyeIterationsInput} ${asmCoverageInput}

Parallel CPU settings

${threadsPerAssemblyInput} ${parallelAssembliesInput} ${cpuSummary} ${minOverlap} ${recursiveInput}

Copy the generated script into a file such as run_parallel_flye.sh, then run it in WSL. Detailed Flye output is saved in a separate log file inside each sample assembly folder.

${textarea} ${copyBtn}

`; }40:T637d,viewof combineAssemblyScript = { const fullWidthInputStyle = "width:100%; max-width:100%; box-sizing:border-box;"; const inputPath = Inputs.text({ label: "📂 Folder containing assembly subfolders (Windows path)", placeholder: "C:\\Users\\yourname\\Desktop\\Nanopore_Results\\Run1\\Sample1\\assemblies", style: fullWidthInputStyle }); const envInput = Inputs.text({ label: "🧪 Conda environment", value: "reads-assembly-core", style: fullWidthInputStyle }); const contigSelection = Inputs.checkbox(["Keep all contigs"], { label: "Contig selection", value: ["Keep all contigs"] }); const barcodeMapPath = Inputs.text({ label: "🧬 Optional barcode-to-phage Excel file (Windows path)", placeholder: "C:\\Users\\yourname\\Desktop\\barcode_to_phage.xlsx", value: "", style: fullWidthInputStyle }); const barcodeMapWorksheet = Inputs.text({ label: "Barcode-map worksheet name", placeholder: "Leave blank to use the first worksheet", value: "", style: fullWidthInputStyle }); const barcodeColumn = Inputs.text({ label: "Barcode column in mapping file", value: "Barcode", style: fullWidthInputStyle }); const phageNameColumn = Inputs.text({ label: "Phage-name column in mapping file", value: "Phage Name", style: fullWidthInputStyle }); const keepUnmappedSamples = Inputs.toggle({ label: "Keep barcodes not found in the barcode map", value: true }); const lengthColumn = Inputs.text({ label: "Length column in assembly_info.txt", value: "length", style: fullWidthInputStyle }); const outputBaseName = Inputs.text({ label: "Output filename base", value: "combined_assembly_info", style: fullWidthInputStyle }); const overwriteOutput = Inputs.toggle({ label: "♻️ Overwrite existing output files", value: true }); const openResults = Inputs.toggle({ label: "📂 Open the output folder when finished", value: false }); const modeMessage = html`

`; const barcodeHelp = html`

Optional barcode-map format
The mapping workbook may contain columns such as Barcode and Phage Name.

Barcode     Phage Name
barcode005  Olympi
barcode006  Tadpole
barcode007  Seldom

When a map is supplied, the output keeps the barcode in a separate Barcode column and writes the mapped phage name in Sample.

`; const textarea = html``; const copyBtn = html``; copyBtn.onclick = async () => { try { await navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; } catch (error) { textarea.select(); document.execCommand("copy"); copyBtn.textContent = "✅ Copied!"; } setTimeout(() => { copyBtn.textContent = "📋 Copy WSL script"; }, 1500); }; function removeOuterQuotes(value) { let text = String(value ?? "").trim(); while ( text.length >= 2 && ((text.startsWith('"') && text.endsWith('"')) || (text.startsWith("'") && text.endsWith("'"))) ) { text = text.slice(1, -1).trim(); } return text; } function toWSLPath(path) { const cleaned = removeOuterQuotes(path); return cleaned .replace(/^([A-Za-z]):/, (_, drive) => `/mnt/${drive.toLowerCase()}`) .replace(/\\/g, "/"); } function shellSingleQuote(value) { return `'${String(value ?? "").replace(/'/g, `'\\''`)}'`; } function pythonString(value) { return JSON.stringify(String(value ?? "")); } function pythonBoolean(value) { return value ? "True" : "False"; } function generateScript() { const defaultBasePath = "/mnt/c/Users/yourname/Desktop/Nanopore_Results/Run1/Sample1/assemblies"; const enteredBasePath = removeOuterQuotes(inputPath.value); const basePath = enteredBasePath ? toWSLPath(enteredBasePath) : defaultBasePath; const enteredMapPath = removeOuterQuotes(barcodeMapPath.value); const mapPath = enteredMapPath ? toWSLPath(enteredMapPath) : ""; const envName = envInput.value.trim() || "reads-assembly-core"; const mapWorksheet = barcodeMapWorksheet.value.trim(); const mapBarcodeHeader = barcodeColumn.value.trim() || "Barcode"; const mapPhageHeader = phageNameColumn.value.trim() || "Phage Name"; const requestedLengthColumn = lengthColumn.value.trim() || "length"; const requestedOutputBase = outputBaseName.value.trim() || "combined_assembly_info"; const keepAll = contigSelection.value.includes("Keep all contigs"); modeMessage.innerHTML = keepAll ? `Current selection: Keep every contig.
The output will contain one row for every contig and will also show the total number of contigs for each barcode.` : `Current selection: Keep only the longest contig for each barcode.
The output will contain one row per barcode and will report how many contigs were present before filtering.`; textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; echo; echo "Press Enter to close..."; read -r _ /dev/null 2>&1; then eval "$(conda shell.bash hook)" elif [ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda3/etc/profile.d/conda.sh" elif [ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]; then . "$HOME/miniconda/etc/profile.d/conda.sh" elif [ -f "$HOME/anaconda3/etc/profile.d/conda.sh" ]; then . "$HOME/anaconda3/etc/profile.d/conda.sh" else echo "❌ Conda could not be found." echo " Checked the current PATH and common Miniconda/Anaconda locations." exit 1 fi if ! conda env list | awk '{print $1}' | grep -Fxq "$ENV_NAME"; then echo "❌ Conda environment not found: $ENV_NAME" echo echo "Available environments:" conda env list exit 1 fi conda activate "$ENV_NAME" python - <<'PYTHON_SCRIPT' from __future__ import annotations import os import re import sys from pathlib import Path from typing import Optional base_dir = Path(${pythonString(basePath)}) barcode_map_path = ${pythonString(mapPath)} barcode_map_worksheet = ${pythonString(mapWorksheet)} barcode_column_requested = ${pythonString(mapBarcodeHeader)} phage_column_requested = ${pythonString(mapPhageHeader)} length_column_requested = ${pythonString(requestedLengthColumn)} output_base_requested = ${pythonString(requestedOutputBase)} keep_all_contigs = ${pythonBoolean(keepAll)} keep_unmapped_samples = ${pythonBoolean(keepUnmappedSamples.value)} overwrite_output = ${pythonBoolean(overwriteOutput.value)} open_results = ${pythonBoolean(openResults.value)} try: import pandas as pd except ImportError: print("❌ pandas is not installed in the selected Conda environment.") print(" Install it with: conda install -n ${envName} pandas openpyxl") sys.exit(1) try: import openpyxl from openpyxl.styles import Alignment, Font, PatternFill except ImportError: print("❌ openpyxl is not installed in the selected Conda environment.") print(" Install it with: conda install -n ${envName} pandas openpyxl") sys.exit(1) def clean_text(value) -> str: if value is None: return "" try: if pd.isna(value): return "" except Exception: pass return re.sub(r"\\s+", " ", str(value).replace("\\u00a0", " ")).strip() def normalize_barcode(value) -> str: text = clean_text(value) if not text: return "" match = re.search( r"barcode[\\s_-]*0*([0-9]+)", text, flags=re.IGNORECASE ) if not match: match = re.fullmatch(r"0*([0-9]+)", text) if not match: return "" return f"barcode{int(match.group(1)):03d}" def find_column(columns, requested_name: str) -> Optional[str]: requested = clean_text(requested_name).casefold() for column in columns: if clean_text(column).casefold() == requested: return column return None def sanitize_output_base(value: str) -> str: text = clean_text(value) if not text: text = "combined_assembly_info" for extension in ( ".xlsx", ".xlsm", ".xls", ".tsv", ".txt" ): if text.lower().endswith(extension): text = text[: -len(extension)] break text = re.sub(r'[<>:"/\\\\|?*]+', "_", text) text = text.strip(" ._") return text or "combined_assembly_info" def read_assembly_info(info_path: Path) -> pd.DataFrame: try: frame = pd.read_csv( info_path, sep="\\t", dtype=str, keep_default_na=False ) except Exception as error: raise RuntimeError( f"Could not read {info_path}: {error}" ) from error # Fall back to whitespace separation if the file was not tab-separated. if frame.shape[1] == 1: frame = pd.read_csv( info_path, sep=r"\\s+", dtype=str, keep_default_na=False ) frame.columns = [ clean_text(column).lstrip("#").strip() for column in frame.columns ] return frame def load_barcode_map(path_text: str) -> dict[str, str]: if not path_text: return {} map_path = Path(path_text) sheet_name = ( barcode_map_worksheet if barcode_map_worksheet else 0 ) try: map_frame = pd.read_excel( map_path, sheet_name=sheet_name, dtype=str, engine="openpyxl" ) except Exception as error: raise RuntimeError( f"Could not read the barcode-map workbook " f"{map_path}: {error}" ) from error barcode_column = find_column( map_frame.columns, barcode_column_requested ) phage_column = find_column( map_frame.columns, phage_column_requested ) available = ", ".join( str(column) for column in map_frame.columns ) if barcode_column is None: raise RuntimeError( f"Barcode-map column " f"'{barcode_column_requested}' was not found. " f"Available columns: {available}" ) if phage_column is None: raise RuntimeError( f"Phage-name column " f"'{phage_column_requested}' was not found. " f"Available columns: {available}" ) mapping: dict[str, str] = {} for _, row in map_frame.iterrows(): barcode = normalize_barcode( row[barcode_column] ) phage_name = clean_text( row[phage_column] ) if ( barcode and phage_name and barcode not in mapping ): mapping[barcode] = phage_name return mapping def choose_output_paths( stem: str ) -> tuple[Path, Path]: preferred_tsv = base_dir / f"{stem}.tsv" preferred_xlsx = base_dir / f"{stem}.xlsx" if overwrite_output: for path in ( preferred_tsv, preferred_xlsx ): if path.exists(): try: path.unlink() except PermissionError as error: raise RuntimeError( "Close the existing output file " f"before overwriting it: {path}" ) from error return preferred_tsv, preferred_xlsx if ( not preferred_tsv.exists() and not preferred_xlsx.exists() ): return preferred_tsv, preferred_xlsx number = 2 while True: candidate_tsv = ( base_dir / f"{stem}_{number}.tsv" ) candidate_xlsx = ( base_dir / f"{stem}_{number}.xlsx" ) if ( not candidate_tsv.exists() and not candidate_xlsx.exists() ): return candidate_tsv, candidate_xlsx number += 1 def format_excel_workbook(path: Path) -> None: workbook = openpyxl.load_workbook(path) worksheet = workbook["Assemblies"] worksheet.freeze_panes = "A2" worksheet.auto_filter.ref = worksheet.dimensions header_fill = PatternFill( fill_type="solid", fgColor="D9EAD3" ) for cell in worksheet[1]: cell.font = Font(bold=True) cell.fill = header_fill cell.alignment = Alignment( horizontal="center", vertical="center" ) for column_cells in worksheet.columns: maximum_length = 0 for cell in column_cells: value = ( "" if cell.value is None else str(cell.value) ) maximum_length = max( maximum_length, len(value) ) column_letter = ( column_cells[0].column_letter ) worksheet.column_dimensions[ column_letter ].width = min( max(maximum_length + 2, 11), 60 ) workbook.save(path) workbook.close() if not base_dir.is_dir(): raise RuntimeError( f"Assembly folder does not exist: {base_dir}" ) barcode_to_phage = load_barcode_map( barcode_map_path ) subfolders = sorted( ( path for path in base_dir.iterdir() if path.is_dir() ), key=lambda path: path.name.casefold() ) if not subfolders: raise RuntimeError( f"No immediate subfolders were found in: " f"{base_dir}" ) frames = [] missing_info_files = [] excluded_unmapped_folders = [] source_order = 0 for folder in subfolders: info_path = folder / "assembly_info.txt" if not info_path.is_file(): missing_info_files.append( str(info_path) ) continue frame = read_assembly_info(info_path) length_column = find_column( frame.columns, length_column_requested ) if length_column is None: available = ", ".join( str(column) for column in frame.columns ) raise RuntimeError( f"Length column " f"'{length_column_requested}' was not " f"found in {info_path}. " f"Available columns: {available}" ) if length_column != "length": frame = frame.rename( columns={ length_column: "length" } ) folder_name = folder.name standardized_barcode = normalize_barcode( folder_name ) barcode_value = ( standardized_barcode or folder_name ) if barcode_to_phage: mapped_name = barcode_to_phage.get( standardized_barcode, "" ) if mapped_name: sample_value = mapped_name elif keep_unmapped_samples: sample_value = folder_name else: excluded_unmapped_folders.append( folder_name ) continue else: sample_value = folder_name group_key = ( f"barcode:{standardized_barcode}" if standardized_barcode else f"folder:{folder_name.casefold()}" ) frame.insert( 0, "Sample", sample_value ) frame.insert( 0, "Barcode", barcode_value ) frame.insert( 0, "Folder", folder_name ) frame["_GroupKey"] = group_key frame["_SourceOrder"] = range( source_order, source_order + len(frame) ) source_order += len(frame) frames.append(frame) print(f"✅ Added: {info_path}") if not frames: raise RuntimeError( "No usable assembly_info.txt files were " "found in the immediate subfolders." ) combined = pd.concat( frames, ignore_index=True, sort=False ) combined["_NumericLength"] = pd.to_numeric( combined["length"] .astype(str) .str.replace(",", "", regex=False) .str.replace("_", "", regex=False) .str.strip(), errors="coerce" ) combined["Number of Contigs"] = ( combined.groupby( "_GroupKey", sort=False )["_GroupKey"] .transform("size") .astype(int) ) invalid_length_rows = int( combined["_NumericLength"].isna().sum() ) if keep_all_contigs: result = combined.copy() mode_suffix = "all_contigs" else: valid_rows = combined[ combined["_NumericLength"].notna() ].copy() if valid_rows.empty: raise RuntimeError( "No contig rows contained a numeric " "length, so the longest contig could " "not be selected." ) groups_with_valid_length = set( valid_rows["_GroupKey"] ) all_groups = set( combined["_GroupKey"] ) groups_without_valid_length = sorted( all_groups - groups_with_valid_length ) winner_indices = ( valid_rows.groupby( "_GroupKey", sort=False )["_NumericLength"] .idxmax() .tolist() ) result = combined.loc[ winner_indices ].copy() result = result.sort_values( "_SourceOrder", kind="stable" ) mode_suffix = ( "longest_contig_per_barcode" ) if groups_without_valid_length: print() print( "⚠️ Groups skipped because none of " "their contigs had a numeric length:" ) for group in groups_without_valid_length: print(f" {group}") output_base = sanitize_output_base( output_base_requested ) output_stem = ( f"{output_base}_{mode_suffix}" ) output_tsv, output_xlsx = ( choose_output_paths(output_stem) ) preferred_columns = [ "Folder", "Barcode", "Sample", "Number of Contigs" ] remaining_columns = [ column for column in result.columns if column not in preferred_columns and not str(column).startswith("_") ] result = result[ preferred_columns + remaining_columns ] result.to_csv( output_tsv, sep="\\t", index=False, lineterminator="\\n" ) with pd.ExcelWriter( output_xlsx, engine="openpyxl" ) as writer: result.to_excel( writer, index=False, sheet_name="Assemblies" ) format_excel_workbook(output_xlsx) print() print( "============================================================" ) print( "✅ Assembly information combined successfully" ) print( "============================================================" ) print( f"Assembly folders examined: " f"{len(subfolders)}" ) print( f"assembly_info.txt files used: " f"{len(frames)}" ) print( f"Input contig rows: {len(combined)}" ) print( f"Output rows: {len(result)}" ) print( f"Unique barcodes or folders: " f"{combined['_GroupKey'].nunique()}" ) print( f"Rows with a non-numeric length: " f"{invalid_length_rows}" ) print( f"Barcode mappings loaded: " f"{len(barcode_to_phage)}" ) print( "Mode: " + ( "keep all contigs" if keep_all_contigs else "longest contig per barcode" ) ) if missing_info_files: print() print( "⚠️ Missing assembly_info.txt files:" ) for missing_path in missing_info_files: print(f" {missing_path}") if excluded_unmapped_folders: print() print( "⚠️ Folders excluded because they were " "absent from the barcode map:" ) for folder_name in excluded_unmapped_folders: print(f" {folder_name}") print() print("Created:") print(f" {output_tsv}") print(f" {output_xlsx}") if open_results: try: os.system( f'explorer.exe "{str(base_dir)}" ' '>/dev/null 2>&1 &' ) except Exception: pass PYTHON_SCRIPT echo echo "🏁 Finished." echo "Press Enter to close..." read -r _

Combine assembly information and select contigs

Assembly folders and environment

${inputPath} ${envInput}

Contig selection

${contigSelection} ${modeMessage}

Optional barcode-to-phage workbook

${barcodeMapPath} ${barcodeMapWorksheet} ${barcodeColumn} ${phageNameColumn} ${keepUnmappedSamples} ${barcodeHelp}

Output settings

${lengthColumn} ${outputBaseName} ${overwriteOutput} ${openResults}

Copy the generated script, paste it into WSL, and press Enter. The script searches the immediate subfolders for assembly_info.txt, creates both TSV and Excel files, and adds a Number of Contigs column.

${copyBtn}

${textarea}

`; }41:T62e3,viewof fastaFastqToBamGeneratorBest = { const mode = Inputs.radio(["One Barcode", "Multiple Barcodes (Batch)"], { label: "1) Processing mode", value: "One Barcode" }); const condaEnvName = Inputs.text({ label: "Conda environment name", value: "reads-assembly-core", style: "width: 320px" }); const threads = Inputs.range([1, 64], { label: "Threads", value: 4, step: 1 }); const endWindowBp = Inputs.number({ label: "Genome-end window size (bp)", value: 1000, min: 1, step: 100, style: "width: 220px" }); const singleBarcode = Inputs.text({ label: "Barcode/sample name", value: "barcode001", style: "width: 260px" }); const singleFastqPattern = Inputs.text({ label: "FASTQ path or glob (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Demux_Output\\*barcode001*.fastq*", style: "width: 760px" }); const singleFastaPath = Inputs.text({ label: "Reference FASTA path (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Demux_Output\\assemblies\\barcode001\\barcode001_racon2.fasta", style: "width: 760px" }); const singleOutPrefixPath = Inputs.text({ label: "Output prefix path, without extension (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Demux_Output\\assemblies\\barcode001\\barcode001_racon2", style: "width: 760px" }); const parentFolderPath = Inputs.text({ label: "Parent/Demux_Output folder (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\Nanopore_Results\\Run1\\Demux_Output", style: "width: 760px" }); const barcodeFolderGlob = Inputs.text({ label: "Barcode-folder glob, relative to parent", value: "assemblies/barcode*/", style: "width: 760px" }); const readsGlob = Inputs.text({ label: "FASTQ glob template, relative to parent; include {BARCODE}", value: "*{BARCODE}*.fastq*", style: "width: 760px" }); const refRel = Inputs.text({ label: "Reference FASTA template, relative to parent; include {BARCODE}", value: "assemblies/{BARCODE}/{BARCODE}_racon2.fasta", style: "width: 760px" }); const outPrefixRel = Inputs.text({ label: "Output prefix template, relative to parent; include {BARCODE}", value: "assemblies/{BARCODE}/{BARCODE}_racon2", style: "width: 760px" }); const summaryRel = Inputs.text({ label: "Combined coverage-summary path, relative to parent", value: "assemblies/coverage_summary.tsv", style: "width: 760px" }); const formArea = html`

`; const singleSection = html`

`; const batchSection = html`

`; const textarea = html``; const copyBtn = html``; const downloadBtn = html``; const status = html``; copyBtn.onclick = async () => { try { await navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy WSL script"), 1400); } catch (error) { copyBtn.textContent = "❌ Copy failed"; setTimeout(() => (copyBtn.textContent = "📋 Copy WSL script"), 1800); } }; downloadBtn.onclick = () => { const blob = new Blob([textarea.value], { type: "text/x-shellscript;charset=utf-8" }); const url = URL.createObjectURL(blob); const a = document.createElement("a"); a.href = url; a.download = "fastq_fasta_to_bam_and_coverage.sh"; document.body.appendChild(a); a.click(); a.remove(); URL.revokeObjectURL(url); }; function stripOuterQuotes(value) { const s = String(value ?? "").trim(); if (s.length >= 2) { const first = s[0]; const last = s[s.length - 1]; if ((first === '"' && last === '"') || (first === "'" && last === "'")) { return s.slice(1, -1).trim(); } } return s; } function toWSLPath(path) { return stripOuterQuotes(path) .replace(/^([A-Za-z]):/, (_, drive) => `/mnt/${drive.toLowerCase()}`) .replace(/\\/g, "/"); } function bashDoubleQuoteEscape(value) { return String(value ?? "") .replace(/\\/g, "\\\\") .replace(/"/g, '\\"') .replace(/\$/g, "\\$") .replace(/`/g, "\\`"); } function buildForm() { formArea.innerHTML = ""; singleSection.innerHTML = ""; batchSection.innerHTML = ""; singleSection.append( html`

One Barcode
The FASTQ entry may be one file or a glob matching several FASTQ/FASTQ.GZ files. All matches are aligned together.

`, singleBarcode, singleFastqPattern, singleFastaPath, singleOutPrefixPath ); batchSection.append( html`

Multiple Barcodes
Each barcode folder is processed separately. All FASTQ files matching that barcode’s glob are combined during alignment.

`, parentFolderPath, barcodeFolderGlob, readsGlob, refRel, outPrefixRel, summaryRel ); formArea.append( mode, condaEnvName, html`

${threads} ${endWindowBp}

`, singleSection, batchSection ); updateModeVisibility(); } function updateModeVisibility() { const isBatch = mode.value === "Multiple Barcodes (Batch)"; singleSection.style.display = isBatch ? "none" : "grid"; batchSection.style.display = isBatch ? "grid" : "none"; } function updateScript() { const isBatch = mode.value === "Multiple Barcodes (Batch)"; const envName = stripOuterQuotes(condaEnvName.value) || "reads-assembly-core"; const threadCount = Math.max(1, Math.round(Number(threads.value) || 1)); const endWindow = Math.max( 1, Math.round(Number(endWindowBp.value) || 1000) ); const oneBarcode = stripOuterQuotes(singleBarcode.value); const oneFastqPattern = toWSLPath(singleFastqPattern.value); const oneFasta = toWSLPath(singleFastaPath.value); const oneOutPrefix = toWSLPath(singleOutPrefixPath.value); const parentDir = toWSLPath(parentFolderPath.value); const barcodeGlobValue = stripOuterQuotes(barcodeFolderGlob.value); const readsGlobValue = stripOuterQuotes(readsGlob.value); const refTemplateValue = stripOuterQuotes(refRel.value); const outTemplateValue = stripOuterQuotes(outPrefixRel.value); const summaryRelValue = stripOuterQuotes(summaryRel.value) || "assemblies/coverage_summary.tsv"; const q = bashDoubleQuoteEscape; const script = `#!/usr/bin/env bash # Generated by Observable # Nanopore FASTQ + reference FASTA -> sorted/indexed BAM # Also creates mapping reports, per-base depth, per-contig coverage statistics, # strand-aware read-start counts, and a clean batch summary. set -u set -o pipefail export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" CONDA_ENV_NAME="${q(envName)}" THREADS="${threadCount}" END_WIN_BP="${endWindow}" MODE="${isBatch ? "batch" : "single"}" ONE_BARCODE="${q(oneBarcode)}" ONE_FASTQ_PATTERN="${q(oneFastqPattern)}" ONE_FASTA="${q(oneFasta)}" ONE_OUT_PREFIX="${q(oneOutPrefix)}" PARENT_DIR="${q(parentDir)}" BARCODE_FOLDER_GLOB="${q(barcodeGlobValue)}" READS_GLOB_TEMPLATE="${q(readsGlobValue)}" REF_TEMPLATE="${q(refTemplateValue)}" OUT_TEMPLATE="${q(outTemplateValue)}" SUMMARY_REL="${q(summaryRelValue)}" pause_and_exit() { local status_code="\${1:-0}" echo echo "Press Enter to close..." read -r _ /dev/null || true exit "$status_code" } print_error() { echo "❌ $*" >&2 } print_warning() { echo "⚠️ $*" >&2 } print_section() { echo echo "============================================================" echo "$*" echo "============================================================" } apply_template() { local template_str="$1" local barcode_name="$2" printf '%s\\n' "\${template_str//\\{BARCODE\\}/$barcode_name}" } ensure_fasta_index() { local reference_fasta="$1" if [ ! -f "$reference_fasta.fai" ] || [ "$reference_fasta" -nt "$reference_fasta.fai" ]; then rm -f "$reference_fasta.fai" if ! samtools faidx "$reference_fasta"; then print_error "Could not index reference FASTA: $reference_fasta" return 1 fi fi return 0 } # Populates the global READ_FILES array. resolve_read_files() { local reads_pattern="$1" READ_FILES=() if [ -f "$reads_pattern" ]; then READ_FILES+=("$reads_pattern") return 0 fi mapfile -t READ_FILES < <(compgen -G "$reads_pattern" | LC_ALL=C sort) if [ "\${#READ_FILES[@]}" -eq 0 ]; then return 1 fi local read_file for read_file in "\${READ_FILES[@]}"; do if [ ! -f "$read_file" ]; then print_warning "Ignoring non-file FASTQ match: $read_file" fi done local valid_files=() for read_file in "\${READ_FILES[@]}"; do [ -f "$read_file" ] && valid_files+=("$read_file") done READ_FILES=("\${valid_files[@]}") [ "\${#READ_FILES[@]}" -gt 0 ] } compute_coverage_stats() { local reference_fasta="$1" local bam_file="$2" local stats_tsv="$3" local depth_tsv="$4" local end_window="$5" if ! ensure_fasta_index "$reference_fasta"; then return 1 fi # -aa: include every reference position, including zero-depth positions. # -d 0: remove the maximum-depth cap. # -G 3844: exclude unmapped, secondary, QC-fail, duplicate, and supplementary alignments. if ! samtools depth -aa -d 0 -G 3844 "$bam_file" > "$depth_tsv"; then print_error "samtools depth failed: $bam_file" return 1 fi if ! awk -v FAI="$reference_fasta.fai" -v WIN="$end_window" -v OUT="$stats_tsv" ' BEGIN { OFS="\\t"; contig_count=0; while ((getline line < FAI) > 0) { split(line, fields, "\\t"); contig=fields[1]; contig_count++; contig_order[contig_count]=contig; contig_length[contig]=fields[2] + 0; } close(FAI); } { contig=$1; position=$2 + 0; depth=$3 + 0; contig_len=contig_length[contig]; window=WIN + 0; if (window < 1) window=1; if (window > contig_len) window=contig_len; sum[contig] += depth; sumsq[contig] += depth * depth; count[contig]++; if (!(contig in minimum) || depth < minimum[contig]) minimum[contig]=depth; if (!(contig in maximum) || depth > maximum[contig]) maximum[contig]=depth; if (depth == 0) zero_count[contig]++; if (position <= window) { left_sum[contig] += depth; left_count[contig]++; } if (position > contig_len - window) { right_sum[contig] += depth; right_count[contig]++; } if (position > window && position <= contig_len - window) { interior_sum[contig] += depth; interior_count[contig]++; } } END { print "contig", "length_bp", "end_window_bp", "mean_depth", "stdev_depth", "min_depth", "max_depth", "zero_coverage_bp", "covered_fraction", "left_mean", "right_mean", "interior_mean", "ends_mean", "end_ratio" > OUT; for (i=1; i<=contig_count; i++) { contig=contig_order[i]; contig_len=contig_length[contig]; window=WIN + 0; if (window < 1) window=1; if (window > contig_len) window=contig_len; n=count[contig] + 0; mean=(n > 0) ? sum[contig] / n : 0; variance=(n > 0) ? (sumsq[contig] / n) - (mean * mean) : 0; if (variance < 0) variance=0; stdev=variance ^ 0.5; min_depth=(contig in minimum) ? minimum[contig] : 0; max_depth=(contig in maximum) ? maximum[contig] : 0; zero_bp=zero_count[contig] + 0; covered_fraction=(contig_len > 0) ? (contig_len - zero_bp) / contig_len : 0; left_mean=(left_count[contig] > 0) ? left_sum[contig] / left_count[contig] : 0; right_mean=(right_count[contig] > 0) ? right_sum[contig] / right_count[contig] : 0; ends_mean=(left_mean + right_mean) / 2.0; if (interior_count[contig] > 0) { interior_mean=interior_sum[contig] / interior_count[contig]; interior_text=sprintf("%.6f", interior_mean); if (interior_mean > 0) { ratio_text=sprintf("%.6f", ends_mean / interior_mean); } else { ratio_text="NA"; } } else { interior_text="NA"; ratio_text="NA"; } printf "%s\\t%d\\t%d\\t%.6f\\t%.6f\\t%d\\t%d\\t%d\\t%.6f\\t%.6f\\t%.6f\\t%s\\t%.6f\\t%s\\n", \ contig, contig_len, window, mean, stdev, min_depth, max_depth, zero_bp, covered_fraction, \ left_mean, right_mean, interior_text, ends_mean, ratio_text >> OUT; } } ' "$depth_tsv"; then print_error "Coverage-statistics calculation failed: $bam_file" return 1 fi return 0 } extract_read_starts() { local reference_fasta="$1" local bam_file="$2" local output_tsv="$3" if ! ensure_fasta_index "$reference_fasta"; then return 1 fi # Exclude unmapped, secondary, QC-fail, duplicate, and supplementary alignments. # Reverse-strand reads are assigned to the rightmost aligned reference coordinate. if ! samtools view -F 3844 "$bam_file" | awk -v FAI="$reference_fasta.fai" ' function cigar_reference_length(cigar, number, operation, total) { total=0; while (match(cigar, /[0-9]+[MIDNSHP=X]/)) { number=substr(cigar, RSTART, RLENGTH - 1) + 0; operation=substr(cigar, RSTART + RLENGTH - 1, 1); if (operation == "M" || operation == "D" || operation == "N" || operation == "=" || operation == "X") { total += number; } cigar=substr(cigar, RSTART + RLENGTH); } return total; } BEGIN { OFS="\\t"; contig_count=0; while ((getline line < FAI) > 0) { split(line, fields, "\\t"); contig=fields[1]; contig_count++; contig_order[contig_count]=contig; contig_length[contig]=fields[2] + 0; } close(FAI); } { flag=$2 + 0; contig=$3; position=$4 + 0; cigar=$6; if (!(contig in contig_length) || cigar == "*") next; reference_span=cigar_reference_length(cigar); if (reference_span < 1) next; is_reverse=int(flag / 16) % 2; read_start=is_reverse ? position + reference_span - 1 : position; if (read_start < 1) read_start=1; if (read_start > contig_length[contig]) read_start=contig_length[contig]; key=contig SUBSEP read_start; if (is_reverse) reverse_count[key]++; else forward_count[key]++; } END { print "contig", "pos", "forward_start_count", "reverse_start_count", "total_start_count"; for (i=1; i<=contig_count; i++) { contig=contig_order[i]; contig_len=contig_length[contig]; for (position=1; position<=contig_len; position++) { key=contig SUBSEP position; forward=(key in forward_count) ? forward_count[key] : 0; reverse=(key in reverse_count) ? reverse_count[key] : 0; print contig, position, forward, reverse, forward + reverse; } } } ' > "$output_tsv"; then print_error "Read-start extraction failed: $bam_file" return 1 fi return 0 } process_one() { local barcode_name="$1" local reference_fasta="$2" local reads_pattern="$3" local output_prefix="$4" print_section "Processing $barcode_name" echo "Reference FASTA: $reference_fasta" echo "FASTQ pattern: $reads_pattern" echo "Output prefix: $output_prefix" if [ -z "$barcode_name" ]; then print_error "Barcode/sample name is empty." return 1 fi if [ -z "$reference_fasta" ] || [ ! -f "$reference_fasta" ]; then print_error "Reference FASTA not found: $reference_fasta" return 1 fi if [ -z "$reads_pattern" ]; then print_error "FASTQ path/glob is empty." return 1 fi if ! resolve_read_files "$reads_pattern"; then print_error "No FASTQ files matched: $reads_pattern" return 1 fi if [ -z "$output_prefix" ]; then print_error "Output prefix is empty." return 1 fi if ! mkdir -p "$(dirname "$output_prefix")"; then print_error "Could not create output directory: $(dirname "$output_prefix")" return 1 fi if ! ensure_fasta_index "$reference_fasta"; then return 1 fi echo echo "FASTQ files to align: \${#READ_FILES[@]}" printf ' %s\\n' "\${READ_FILES[@]}" echo local bam_file="$output_prefix.sorted.bam" local bai_file="$bam_file.bai" local flagstat_file="$output_prefix.flagstat.txt" local idxstats_file="$output_prefix.idxstats.txt" local coverage_file="$output_prefix.coverage_stats.tsv" local depth_file="$output_prefix.depth.tsv" local starts_file="$output_prefix.read_starts.tsv" # Remove outputs from an earlier run so failed reruns cannot leave mixed old/new results. rm -f "$bam_file" "$bai_file" "$flagstat_file" "$idxstats_file" \ "$coverage_file" "$depth_file" "$starts_file" echo "Running minimap2 and samtools sort..." if ! minimap2 -a -x map-ont -t "$THREADS" "$reference_fasta" "\${READ_FILES[@]}" \ | samtools sort -@ "$THREADS" -o "$bam_file" -; then print_error "Alignment or BAM sorting failed for $barcode_name" rm -f "$bam_file" "$bai_file" return 1 fi if ! samtools quickcheck -v "$bam_file"; then print_error "The BAM file failed samtools quickcheck: $bam_file" rm -f "$bam_file" "$bai_file" return 1 fi echo "Indexing BAM..." if ! samtools index -@ "$THREADS" "$bam_file"; then print_error "BAM indexing failed for $barcode_name" return 1 fi echo "Creating mapping reports..." if ! samtools flagstat -@ "$THREADS" "$bam_file" > "$flagstat_file"; then print_warning "Could not create flagstat report." fi if ! samtools idxstats "$bam_file" > "$idxstats_file"; then print_warning "Could not create idxstats report." fi echo "Calculating depth and coverage statistics..." if ! compute_coverage_stats "$reference_fasta" "$bam_file" "$coverage_file" "$depth_file" "$END_WIN_BP"; then return 1 fi echo "Calculating strand-aware read-start counts..." if ! extract_read_starts "$reference_fasta" "$bam_file" "$starts_file"; then return 1 fi echo echo "✅ Finished $barcode_name" echo "BAM: $bam_file" echo "BAI: $bai_file" echo "Flagstat: $flagstat_file" echo "Idxstats: $idxstats_file" echo "Coverage stats: $coverage_file" echo "Per-base depth: $depth_file" echo "Read starts: $starts_file" LAST_COVERAGE_FILE="$coverage_file" return 0 } print_section "FASTQ + FASTA to BAM and coverage analysis" echo "Conda environment: $CONDA_ENV_NAME" echo "Threads: $THREADS" echo "End window: $END_WIN_BP bp" echo "Mode: $MODE" if ! command -v conda >/dev/null 2>&1; then print_error "conda was not found in this shell." pause_and_exit 1 fi if ! eval "$(conda shell.bash hook)"; then print_error "Conda could not be initialized in this Bash shell." pause_and_exit 1 fi if ! conda activate "$CONDA_ENV_NAME"; then print_error "Could not activate Conda environment: $CONDA_ENV_NAME" pause_and_exit 1 fi hash -r missing_program=0 for program in minimap2 samtools awk compgen sort; do if ! command -v "$program" >/dev/null 2>&1; then print_error "$program was not found after activating $CONDA_ENV_NAME" missing_program=1 fi done if [ "$missing_program" -ne 0 ]; then pause_and_exit 1 fi if [ "$MODE" = "single" ]; then if ! process_one "$ONE_BARCODE" "$ONE_FASTA" "$ONE_FASTQ_PATTERN" "$ONE_OUT_PREFIX"; then pause_and_exit 1 fi print_section "All requested processing completed successfully" pause_and_exit 0 fi if [ -z "$PARENT_DIR" ] || [ ! -d "$PARENT_DIR" ]; then print_error "Parent folder not found: $PARENT_DIR" pause_and_exit 1 fi for required_template in "$READS_GLOB_TEMPLATE" "$REF_TEMPLATE" "$OUT_TEMPLATE"; do if [[ "$required_template" != *"{BARCODE}"* ]]; then print_error "Every batch FASTQ/reference/output template must contain {BARCODE}: $required_template" pause_and_exit 1 fi done barcode_folder_pattern="$PARENT_DIR/$BARCODE_FOLDER_GLOB" mapfile -t BARCODE_DIRS < <(compgen -G "$barcode_folder_pattern" | LC_ALL=C sort) if [ "\${#BARCODE_DIRS[@]}" -eq 0 ]; then print_error "No barcode folders matched: $barcode_folder_pattern" pause_and_exit 1 fi summary_tsv="$PARENT_DIR/$SUMMARY_REL" mkdir -p "$(dirname "$summary_tsv")" printf 'barcode\\tcontig\\tlength_bp\\tend_window_bp\\tmean_depth\\tstdev_depth\\tmin_depth\\tmax_depth\\tzero_coverage_bp\\tcovered_fraction\\tleft_mean\\tright_mean\\tinterior_mean\\tends_mean\\tend_ratio\\n' > "$summary_tsv" processed_count=0 success_count=0 failed_count=0 for barcode_dir in "\${BARCODE_DIRS[@]}"; do [ -d "$barcode_dir" ] || continue processed_count=$((processed_count + 1)) barcode_name="$(basename "\${barcode_dir%/}")" reference_rel="$(apply_template "$REF_TEMPLATE" "$barcode_name")" reads_rel="$(apply_template "$READS_GLOB_TEMPLATE" "$barcode_name")" output_rel="$(apply_template "$OUT_TEMPLATE" "$barcode_name")" reference_fasta="$PARENT_DIR/$reference_rel" reads_pattern="$PARENT_DIR/$reads_rel" output_prefix="$PARENT_DIR/$output_rel" LAST_COVERAGE_FILE="" if process_one "$barcode_name" "$reference_fasta" "$reads_pattern" "$output_prefix"; then success_count=$((success_count + 1)) if [ -n "$LAST_COVERAGE_FILE" ] && [ -f "$LAST_COVERAGE_FILE" ]; then awk -v BARCODE="$barcode_name" 'BEGIN{OFS="\\t"} NR>1{print BARCODE, $0}' "$LAST_COVERAGE_FILE" >> "$summary_tsv" else print_warning "Coverage file was not available for summary: $barcode_name" fi else failed_count=$((failed_count + 1)) print_warning "Skipping $barcode_name after an error." fi done if [ "$processed_count" -eq 0 ]; then print_error "The barcode glob matched entries, but none were directories: $barcode_folder_pattern" pause_and_exit 1 fi print_section "Batch processing complete" echo "Barcode folders found: $processed_count" echo "Successful: $success_count" echo "Failed: $failed_count" echo "Coverage summary: $summary_tsv" if [ "$failed_count" -gt 0 ]; then print_warning "$failed_count barcode(s) failed. Review the messages above." pause_and_exit 1 fi pause_and_exit 0 `; textarea.value = script; status.textContent = isBatch ? "Batch mode creates BAM/BAI, reports, depth, read-start files, and a newly rebuilt combined summary." : "Single mode accepts one FASTQ or a glob matching multiple FASTQ/FASTQ.GZ files."; } buildForm(); const scriptInputs = [ mode, condaEnvName, threads, endWindowBp, singleBarcode, singleFastqPattern, singleFastaPath, singleOutPrefixPath, parentFolderPath, barcodeFolderGlob, readsGlob, refRel, outPrefixRel, summaryRel ]; scriptInputs.forEach((input) => { input.addEventListener("input", () => { if (input === mode) updateModeVisibility(); updateScript(); }); }); updateScript(); return html`

⬇ FASTQ + FASTA → BAM, coverage, and read-start analysis (WSL)

${formArea}

The generated script automatically activates the Conda environment, validates the BAM, combines all matching FASTQ files, and excludes secondary/supplementary alignments from coverage and read-start statistics while retaining them in the BAM for inspection.

${copyBtn} ${downloadBtn}

${status}

${textarea}

`; }42:T2c0e,viewof fastaFastqToBamGenerator3 = { const mode = Inputs.radio(["One Barcode", "Multiple Barcodes (Batch)"], { label: "1) Create BAM file for One Barcode or Multiple Barcodes (Batch)", value: "One Barcode" }); const condaEnvName = Inputs.text({ label: "Conda environment name", value: "reads-assembly-core", style: "width: 320px" }); const parentFolderPath = Inputs.text({ label: "Parent folder path (Windows-style)", style: "width: 760px" }); const barcode = Inputs.text({ label: "Barcode name", value: "barcode001", style: "width: 220px" }); const singleFastqPath = Inputs.text({ label: "FASTQ file path (Windows-style)", style: "width: 760px" }); const singleFastaPath = Inputs.text({ label: "Reference FASTA file path (Windows-style)", style: "width: 760px" }); const singleOutPrefixPath = Inputs.text({ label: "Output prefix path (no extension, Windows-style)", style: "width: 760px" }); const readsGlob = Inputs.text({ label: "FASTQ glob template (batch, include {BARCODE})", value: "*{BARCODE}*.fastq", style: "width: 760px" }); const refRel = Inputs.text({ label: "FASTA template (batch, include {BARCODE})", value: "assemblies/{BARCODE}/{BARCODE}_assembly.fasta", style: "width: 760px" }); const outPrefixRel = Inputs.text({ label: "Output prefix template (batch, include {BARCODE})", value: "assemblies/{BARCODE}/{BARCODE}", style: "width: 760px" }); const threads = Inputs.range([1, 64], { label: "Threads", value: 4, step: 1 }); const formArea = html`

`; const singleSection = html`

`; const batchSection = html`

`; const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy"), 1200); }; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function bashEscape(s) { return String(s || "").replace(/"/g, '\\"'); } function buildForm() { formArea.innerHTML = ""; singleSection.innerHTML = ""; batchSection.innerHTML = ""; singleSection.append( html`

One Barcode
Enter one reference FASTA, one FASTQ, and one output prefix.

`, barcode, singleFastqPath, singleFastaPath, singleOutPrefixPath ); batchSection.append( html`

Multiple Barcodes (Batch)
Enter a parent folder and use {BARCODE} in the batch templates.

`, parentFolderPath, readsGlob, refRel, outPrefixRel ); formArea.append( mode, condaEnvName, html`

${threads}

⬇ FASTQ + FASTA to sorted BAM WSL script

${formArea}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

${textarea}

`; }43:Tfdf,viewof reverseComplementFastaGenerator = { const fastaInput = Inputs.text({ label: "🧬 Input FASTA file path (Windows-style)", placeholder: "C:\\Users\\YourName\\Desktop\\genome.fasta", style: "width: 600px" }); const outputNameInput = Inputs.text({ label: "📄 Output FASTA filename", value: "reverse_complement.fasta", style: "width: 360px" }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy WSL Script"), 1500); }; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function shQuote(s) { return `'${String(s).replace(/'/g, `'\\''`)}'`; } function generateScript() { const fastaWin = fastaInput.value.trim(); const outputName = outputNameInput.value.trim() || "reverse_complement.fasta"; if (!fastaWin) { textarea.value = ""; copyBtn.disabled = true; return; } copyBtn.disabled = false; const fastaWSL = toWSLPath(fastaWin); textarea.value = `#!/usr/bin/env bash set -Eeuo pipefail trap 'status=$?; echo; echo "❌ Script failed at line $LINENO (exit $status)"; echo "Press Enter to close..."; read -r _; exit $status' ERR INPUT_FASTA=${shQuote(fastaWSL)} OUTPUT_NAME=${shQuote(outputName)} if [ ! -f "$INPUT_FASTA" ]; then echo "❌ Input FASTA not found: $INPUT_FASTA" echo "Press Enter to close..." read -r _ exit 1 fi INPUT_DIR="$(dirname "$INPUT_FASTA")" OUTPUT_FASTA="$INPUT_DIR/$OUTPUT_NAME" export INPUT_FASTA export OUTPUT_FASTA python3 - <<'PY' import os from pathlib import Path input_fasta = Path(os.environ["INPUT_FASTA"]) output_fasta = Path(os.environ["OUTPUT_FASTA"]) comp = str.maketrans( "ACGTRYKMSWBDHVNacgtrykmswbdhvn", "TGCAYRMKSWVHDBNtgcayrmkswvhdbn" ) records = [] header = None seq_parts = [] with input_fasta.open() as handle: for line in handle: line = line.rstrip("\\n") if line.startswith(">"): if header is not None: records.append((header, "".join(seq_parts))) header = line seq_parts = [] else: seq_parts.append(line.strip()) if header is not None: records.append((header, "".join(seq_parts))) if not records: raise SystemExit("No FASTA records found.") with output_fasta.open("w") as out: for header, seq in records: rc = seq.translate(comp)[::-1] out.write(header + " reverse_complement\\n") for i in range(0, len(rc), 80): out.write(rc[i:i+80] + "\\n") print(f"✅ Reverse-complement FASTA written to: {output_fasta}") PY echo echo "✅ Done." echo "📁 Output FASTA: $OUTPUT_FASTA" echo "Press Enter to close..." read -r _ `; } fastaInput.addEventListener("input", generateScript); outputNameInput.addEventListener("input", generateScript); generateScript(); return html`

⬇ Reverse-complement FASTA WSL script

${fastaInput} ${outputNameInput}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

${textarea}

`; }44:T20bf,viewof quastRunnerGenerator = { const condaEnvName = Inputs.text({ label: "Conda environment name", value: "reads-assembly-core", style: "width: 320px" }); const assemblyPaths = Inputs.textarea({ label: "Assembly FASTA file path(s), one per line (Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\project\\assembly1.fasta\nC:\\Users\\kiera\\Desktop\\project\\assembly2.fasta", rows: 6, style: "width: 760px; font-family: monospace;" }); const outputDir = Inputs.text({ label: "QUAST results folder (Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\project\\quast_results", style: "width: 760px" }); const referencePath = Inputs.text({ label: "Reference FASTA file path (optional, Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\project\\reference.fasta", style: "width: 760px" }); const reads1Path = Inputs.text({ label: "Reads 1 FASTQ file path (optional, Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\project\\reads_R1.fastq.gz", style: "width: 760px" }); const reads2Path = Inputs.text({ label: "Reads 2 FASTQ file path (optional, Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\project\\reads_R2.fastq.gz", style: "width: 760px" }); const nanoporeReadsPath = Inputs.text({ label: "Nanopore / long reads FASTQ file path (optional, Windows-style)", placeholder: "C:\\Users\\kiera\\Desktop\\project\\reads.fastq.gz", style: "width: 760px" }); const threads = Inputs.range([1, 64], { label: "Threads", value: 4, step: 1 }); const minContig = Inputs.number({ label: "Minimum contig length", value: 500, step: 100, min: 0 }); const isLargeGenome = Inputs.toggle({ label: "Large genome mode", value: false }); const useGeneFinding = Inputs.toggle({ label: "Run gene finding", value: false }); const isFragmented = Inputs.toggle({ label: "Fragmented reference", value: false }); const extraArgs = Inputs.text({ label: "Extra QUAST arguments (optional)", placeholder: "--labels asm1,asm2", style: "width: 760px" }); const textarea = html``; const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(textarea.value); copyBtn.textContent = "✅ Copied!"; setTimeout(() => (copyBtn.textContent = "📋 Copy"), 1200); }; function toWSLPath(path) { return (path || "") .trim() .replace(/^([A-Za-z]):/, (_, d) => `/mnt/${d.toLowerCase()}`) .replace(/\\/g, "/"); } function bashEscape(s) { return String(s || "").replace(/"/g, '\\"'); } function buildAssemblyArray(lines) { return lines .map((s) => s.trim()) .filter(Boolean) .map(toWSLPath); } function updateScript() { const assemblies = buildAssemblyArray( (assemblyPaths.value || "").split("\n") ); const outDirWSL = toWSLPath(outputDir.value || ""); const refWSL = toWSLPath(referencePath.value || ""); const r1WSL = toWSLPath(reads1Path.value || ""); const r2WSL = toWSLPath(reads2Path.value || ""); const longReadsWSL = toWSLPath(nanoporeReadsPath.value || ""); const envName = (condaEnvName.value || "").trim(); const extra = (extraArgs.value || "").trim(); const assembliesEscaped = assemblies.length > 0 ? assemblies.map((p) => `"${bashEscape(p)}"`) : [`"PUT_ASSEMBLY_PATH_HERE.fasta"`]; const quastArgs = []; quastArgs.push(`--threads ${threads.value}`); quastArgs.push(`--min-contig ${minContig.value}`); if (refWSL) quastArgs.push(`--reference "${bashEscape(refWSL)}"`); if (r1WSL) quastArgs.push(`--pe1 "${bashEscape(r1WSL)}"`); if (r2WSL) quastArgs.push(`--pe2 "${bashEscape(r2WSL)}"`); if (longReadsWSL) quastArgs.push(`--nanopore "${bashEscape(longReadsWSL)}"`); if (isLargeGenome.value) quastArgs.push(`--large`); if (useGeneFinding.value) quastArgs.push(`--gene-finding`); if (isFragmented.value) quastArgs.push(`--fragmented`); if (extra) quastArgs.push(extra); const lines = [ "#!/usr/bin/env bash", "# Generated by Observable — run QUAST in WSL", "", "set -u", "set -o pipefail", "", 'export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH"', "", `CONDA_ENV_NAME="${bashEscape(envName)}"`, `OUT_DIR="${bashEscape(outDirWSL)}"`, 'LOG_FILE="$OUT_DIR/quast_run.log"', "", 'if [ -z "$OUT_DIR" ]; then', ' echo "Error: output directory is required"', ' read -p "Press Enter to exit..."', " exit 1", "fi", "", 'mkdir -p "$OUT_DIR" || {', ' echo "Error: could not create output directory: $OUT_DIR"', ' read -p "Press Enter to exit..."', " exit 1", "}", "", "if ! command -v conda >/dev/null 2>&1; then", ' echo "Error: conda not found in this shell."', ' read -p "Press Enter to exit..."', " exit 1", "fi", "", 'eval "$(conda shell.bash hook)" || {', ' echo "Error: failed to initialize conda in this shell."', ' read -p "Press Enter to exit..."', " exit 1", "}", "", 'conda activate "$CONDA_ENV_NAME" || {', ' echo "Error: failed to activate conda environment: $CONDA_ENV_NAME"', ' read -p "Press Enter to exit..."', " exit 1", "}", "", "hash -r", "", "if ! command -v quast.py >/dev/null 2>&1; then", ' echo "Error: quast.py not found after activating $CONDA_ENV_NAME"', ' read -p "Press Enter to exit..."', " exit 1", "fi", "", 'echo "Logging to: $LOG_FILE"', 'echo "QUAST run started at $(date)" > "$LOG_FILE"', 'echo "Output directory: $OUT_DIR" >> "$LOG_FILE"', 'echo "Conda environment: $CONDA_ENV_NAME" >> "$LOG_FILE"', "", 'echo "Running QUAST..."', "quast.py \\", ` ${assembliesEscaped.join(" \\\n ")} \\`, ' -o "$OUT_DIR" \\', ` ${quastArgs.join(" \\\n ")} \\`, ' 2>&1 | tee -a "$LOG_FILE"', "", 'STATUS="${PIPESTATUS[0]}"', "", 'if [ "$STATUS" -ne 0 ]; then', " echo", ' echo "Error: QUAST failed with exit code $STATUS"', ' echo "See log: $LOG_FILE"', ' read -p "Press Enter to exit..."', ' exit "$STATUS"', "fi", "", "echo", 'echo "QUAST completed successfully."', 'echo "Output folder: $OUT_DIR"', 'echo "Log file: $LOG_FILE"', 'read -p "Press Enter to exit..."' ]; textarea.value = lines.join("\n"); } [ condaEnvName, assemblyPaths, outputDir, referencePath, reads1Path, reads2Path, nanoporeReadsPath, threads, minContig, isLargeGenome, useGeneFinding, isFragmented, extraArgs ].forEach((input) => input.addEventListener("input", updateScript)); updateScript(); return html`

⬇ QUAST WSL runner script

Run QUAST in WSL
Enter one or more assembly FASTA files, choose a QUAST results folder, and optionally provide a reference and reads.

${condaEnvName} ${assemblyPaths} ${outputDir} ${referencePath} ${reads1Path} ${reads2Path} ${nanoporeReadsPath}

${threads} ${minContig} ${isLargeGenome} ${useGeneFinding} ${isFragmented}

${extraArgs}

Copy the generated script, paste it into WSL, and press Enter.

${textarea} ${copyBtn}

`; }45:T108a,viewof installQuastWSL = { const installScript = `#!/bin/bash # 🧬 Install QUAST in WSL (reads-assembly-core) export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" hash -r LOG_FILE="$HOME/install_quast_reads_assembly_core.log" echo "📝 Logging to $LOG_FILE" echo "QUAST installation started at $(date)" > "$LOG_FILE" echo "PATH=$PATH" | tee -a "$LOG_FILE" conda deactivate >/dev/null 2>&1 || true if ! command -v conda >/dev/null 2>&1; then echo "❌ ERROR: conda not found in this shell." | tee -a "$LOG_FILE" echo "Open WSL and ensure conda works first." | tee -a "$LOG_FILE" echo "Press Enter to exit..." read exit 1 fi eval "$(conda shell.bash hook)" 2>>"$LOG_FILE" if ! conda env list | grep -E '^reads-assembly-core[[:space:]]' >/dev/null 2>&1; then echo "❌ ERROR: 'reads-assembly-core' environment not found." | tee -a "$LOG_FILE" echo "Create it first before installing QUAST." | tee -a "$LOG_FILE" echo "Press Enter to exit..." read exit 1 fi conda activate reads-assembly-core 2>&1 | tee -a "$LOG_FILE" || { echo "❌ ERROR: Failed to activate environment." | tee -a "$LOG_FILE" echo "Press Enter to exit..." read exit 1 } echo "📜 Accepting Anaconda Terms of Service..." | tee -a "$LOG_FILE" conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/main 2>&1 | tee -a "$LOG_FILE" conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/r 2>&1 | tee -a "$LOG_FILE" conda config --add channels defaults 2>&1 | tee -a "$LOG_FILE" || true conda config --add channels bioconda 2>&1 | tee -a "$LOG_FILE" || true conda config --add channels conda-forge 2>&1 | tee -a "$LOG_FILE" || true conda config --set channel_priority strict 2>&1 | tee -a "$LOG_FILE" PY_VER=$(python -c 'import sys; print(f"{sys.version_info.major}.{sys.version_info.minor}")' 2>/dev/null) echo "🐍 Detected Python version in reads-assembly-core: $PY_VER" | tee -a "$LOG_FILE" if [ "$PY_VER" = "3.13" ]; then echo "⚠️ Python 3.13 detected. Downgrading to Python 3.10 for QUAST compatibility..." | tee -a "$LOG_FILE" conda install -y -c conda-forge -c bioconda python=3.10 2>&1 | tee -a "$LOG_FILE" || { echo "❌ ERROR: Failed to switch environment to Python 3.10." | tee -a "$LOG_FILE" echo "Press Enter to exit..." read exit 1 } hash -r fi echo "📦 Installing QUAST via conda..." | tee -a "$LOG_FILE" conda install -y -c conda-forge -c bioconda quast 2>&1 | tee -a "$LOG_FILE" hash -r if ! command -v quast.py >/dev/null 2>&1 && [ ! -f "$CONDA_PREFIX/bin/quast.py" ]; then echo "⚠️ QUAST not found after conda install. Trying pip fallback..." | tee -a "$LOG_FILE" if command -v python3 >/dev/null 2>&1; then python3 -m pip install --upgrade pip 2>&1 | tee -a "$LOG_FILE" python3 -m pip install quast 2>&1 | tee -a "$LOG_FILE" hash -r else echo "❌ ERROR: python3 not found for QUAST fallback." | tee -a "$LOG_FILE" fi fi if command -v quast.py >/dev/null 2>&1 || [ -f "$CONDA_PREFIX/bin/quast.py" ]; then echo "✅ QUAST installed successfully!" | tee -a "$LOG_FILE" quast.py --version 2>&1 | tee -a "$LOG_FILE" else echo "❌ ERROR: QUAST installation failed." | tee -a "$LOG_FILE" fi echo "" echo "👉 To use QUAST:" echo "conda activate reads-assembly-core" echo "quast.py --help" echo "" echo "📄 Log saved to: $LOG_FILE" echo "Press Enter to exit..." read `; const textarea = Inputs.textarea({ value: installScript, rows: 46 }); const copyBtn = html``; copyBtn.onclick = () => { navigator.clipboard.writeText(installScript); copyBtn.textContent = "✅ Copied!"; setTimeout(() => { copyBtn.textContent = "📋 Copy QUAST Install Script"; }, 2000); }; return html`

🧬 Install QUAST in reads-assembly-core (WSL)

${textarea}

${copyBtn}

`; }46:T47b3,viewof runMedakaWSL = { // ============================================================ // Input controls // ============================================================ const fastqInput = Inputs.text({ label: "Path to Nanopore FASTQ or FASTQ.GZ reads", placeholder: "C:\\Users\\kiera\\Desktop\\Nanopore_Results\\filtered_barcode01.fastq", description: "Use the filtered reads that were used to make the Flye assembly.", style: "width: 100%;" }); const draftFasta = Inputs.text({ label: "Path to draft FASTA file", placeholder: "C:\\Users\\kiera\\Desktop\\Nanopore_Results\\assembly.fasta", description: "The Flye assembly.fasta is recommended. A Racon-polished FASTA can also be used.", style: "width: 100%;" }); const outputDir = Inputs.text({ label: "Path to Medaka output folder", placeholder: "C:\\Users\\kiera\\Desktop\\Nanopore_Results\\medaka_output", description: "Use a separate output folder for each genome or polishing attempt.", style: "width: 100%;" }); const condaEnv = Inputs.text({ label: "Conda environment", value: "medaka-polish", style: "width: 100%;" }); const model = Inputs.text({ label: "Medaka consensus model", value: "r1041_e82_400bps_sup_v5.2.0", description: "The model must match the Dorado/basecaller model used for the reads.", style: "width: 100%;" }); const threads = Inputs.range([1, 32], { label: "CPU threads", value: 8, step: 1 }); const runVersionReport = Inputs.toggle({ label: "Run Medaka dependency report before polishing", value: true }); // ============================================================ // Path and Bash-quoting helpers // ============================================================ function removeOuterQuotes(value) { let cleaned = String(value ?? "").trim(); const surroundedByDoubleQuotes = cleaned.startsWith('"') && cleaned.endsWith('"'); const surroundedBySingleQuotes = cleaned.startsWith("'") && cleaned.endsWith("'"); if ( cleaned.length >= 2 && (surroundedByDoubleQuotes || surroundedBySingleQuotes) ) { cleaned = cleaned.slice(1, -1); } return cleaned.trim(); } function toWSLPath(path) { let cleaned = removeOuterQuotes(path); // Convert Windows backslashes to forward slashes. cleaned = cleaned.replace(/\\/g, "/"); // Convert C:/... to /mnt/c/... cleaned = cleaned.replace( /^([A-Za-z]):/, (_, drive) => `/mnt/${drive.toLowerCase()}` ); return cleaned; } /* * Safely quote a value for Bash. * * For example: * John's reads.fastq * * becomes: * 'John'"'"'s reads.fastq' */ function shellQuote(value) { return "'" + String(value ?? "").replace(/'/g, "'\"'\"'") + "'"; } function getValues() { return { fastq: toWSLPath(fastqInput.value), draft: toWSLPath(draftFasta.value), output: toWSLPath(outputDir.value), environment: String(condaEnv.value ?? "").trim(), model: String(model.value ?? "").trim(), threads: Math.max(1, Math.floor(Number(threads.value) || 1)), runVersionReport: Boolean(runVersionReport.value) }; } function getMissingInputs() { const values = getValues(); const missing = []; if (!values.fastq) { missing.push("FASTQ path"); } if (!values.draft) { missing.push("draft FASTA path"); } if (!values.output) { missing.push("output-folder path"); } if (!values.environment) { missing.push("Conda environment"); } return missing; } // ============================================================ // Generate WSL script // ============================================================ function buildScript() { const values = getValues(); const versionReportValue = values.runVersionReport ? "true" : "false"; return `#!/usr/bin/env bash # ============================================================ # Medaka polishing script generated by Observable # # Recommended workflow: # # Filtered Nanopore reads # ↓ # Flye assembly.fasta # ↓ # Medaka consensus.fasta # # The script runs inside a Bash subshell so that an error does # not close the interactive WSL terminal when pasted directly. # ============================================================ ( set -Eeuo pipefail # ------------------------------------------------------------ # User-selected settings # ------------------------------------------------------------ CONDA_ENV=${shellQuote(values.environment)} FASTQ=${shellQuote(values.fastq)} DRAFT=${shellQuote(values.draft)} OUTDIR=${shellQuote(values.output)} MODEL=${shellQuote(values.model)} THREADS=${values.threads} RUN_VERSION_REPORT=${versionReportValue} # ------------------------------------------------------------ # Error helper # ------------------------------------------------------------ die() { echo echo "❌ ERROR: $*" >&2 echo exit 1 } echo echo "============================================================" echo " Medaka genome polishing" echo "============================================================" echo # ------------------------------------------------------------ # Validate generated values # ------------------------------------------------------------ [[ -n "$CONDA_ENV" ]] || die "The Conda environment name is empty." [[ -n "$FASTQ" ]] || die "The Nanopore FASTQ path is empty." [[ -n "$DRAFT" ]] || die "The draft FASTA path is empty." [[ -n "$OUTDIR" ]] || die "The Medaka output-directory path is empty." [[ "$OUTDIR" != "/" ]] || die "The root directory cannot be used as the output folder." # ------------------------------------------------------------ # Locate and initialize Conda # ------------------------------------------------------------ echo "🔎 Locating Conda..." if command -v conda >/dev/null 2>&1; then eval "$(conda shell.bash hook)" elif [[ -f "$HOME/miniconda/etc/profile.d/conda.sh" ]]; then source "$HOME/miniconda/etc/profile.d/conda.sh" elif [[ -f "$HOME/miniconda3/etc/profile.d/conda.sh" ]]; then source "$HOME/miniconda3/etc/profile.d/conda.sh" elif [[ -f "$HOME/anaconda3/etc/profile.d/conda.sh" ]]; then source "$HOME/anaconda3/etc/profile.d/conda.sh" else die "Conda could not be located." fi echo "✅ Conda located." echo # ------------------------------------------------------------ # Activate Medaka environment # ------------------------------------------------------------ echo "🧪 Activating Conda environment: $CONDA_ENV" if ! conda activate "$CONDA_ENV"; then die "Could not activate Conda environment: $CONDA_ENV" fi hash -r echo "✅ Conda environment activated." echo # ------------------------------------------------------------ # Verify required programs # ------------------------------------------------------------ echo "🔎 Checking required programs..." REQUIRED_PROGRAMS=( python medaka medaka_consensus medaka_version_report minimap2 samtools bcftools bgzip tabix ) MISSING_PROGRAMS=() for PROGRAM in "\${REQUIRED_PROGRAMS[@]}"; do if command -v "$PROGRAM" >/dev/null 2>&1; then echo " ✅ $PROGRAM" else echo " ❌ $PROGRAM" MISSING_PROGRAMS+=("$PROGRAM") fi done if [[ "\${#MISSING_PROGRAMS[@]}" -gt 0 ]]; then echo echo "The following programs are missing:" printf " - %s\\n" "\${MISSING_PROGRAMS[@]}" die "The Medaka environment is incomplete." fi echo echo "✅ All required programs were found." echo # ------------------------------------------------------------ # Verify that Medaka starts # # Medaka 2.2.2 accepts -h rather than --help. # ------------------------------------------------------------ echo "🔎 Testing Medaka startup..." MEDAKA_STARTUP_LOG="$(mktemp)" trap 'rm -f "$MEDAKA_STARTUP_LOG"' EXIT if ! medaka_consensus -h >"$MEDAKA_STARTUP_LOG" 2>&1; then echo echo "Medaka startup output:" cat "$MEDAKA_STARTUP_LOG" die "medaka_consensus was found but could not start." fi rm -f "$MEDAKA_STARTUP_LOG" trap - EXIT echo "✅ medaka_consensus startup test passed." echo echo "Medaka version:" medaka --version || true echo # ------------------------------------------------------------ # Run Medaka dependency-version report # ------------------------------------------------------------ if [[ "$RUN_VERSION_REPORT" == "true" ]]; then echo "🔎 Running Medaka dependency report..." echo if ! medaka_version_report; then die "One or more Medaka dependencies failed the version check." fi echo echo "✅ Medaka dependency report passed." echo fi # ------------------------------------------------------------ # Validate input files # ------------------------------------------------------------ echo "🔎 Checking input files..." if [[ ! -e "$FASTQ" ]]; then die "Nanopore read file was not found: $FASTQ" fi if [[ ! -f "$FASTQ" ]]; then die "Nanopore read path is not a regular file: $FASTQ" fi if [[ ! -s "$FASTQ" ]]; then die "Nanopore read file is empty: $FASTQ" fi if [[ ! -e "$DRAFT" ]]; then die "Draft FASTA was not found: $DRAFT" fi if [[ ! -f "$DRAFT" ]]; then die "Draft FASTA path is not a regular file: $DRAFT" fi if [[ ! -s "$DRAFT" ]]; then die "Draft FASTA is empty: $DRAFT" fi echo "✅ Input files were found and are not empty." echo # ------------------------------------------------------------ # Display selected files and settings # ------------------------------------------------------------ echo "📂 Nanopore reads:" echo " $FASTQ" echo echo "🧬 Draft assembly:" echo " $DRAFT" echo echo "📁 Medaka output folder:" echo " $OUTDIR" echo echo "🧵 CPU threads:" echo " $THREADS" echo if [[ -n "$MODEL" ]]; then echo "🧠 Medaka model:" echo " $MODEL" else echo "🧠 Medaka model:" echo " Automatic model selection" fi echo # ------------------------------------------------------------ # Prepare output location # ------------------------------------------------------------ CONSENSUS="$OUTDIR/consensus.fasta" if [[ -s "$CONSENSUS" ]]; then die "A non-empty consensus.fasta already exists in the output folder. Choose a new output folder or remove the previous Medaka output." fi mkdir -p "$OUTDIR" if [[ ! -d "$OUTDIR" ]]; then die "The output directory could not be created: $OUTDIR" fi # ------------------------------------------------------------ # Construct Medaka command as a Bash array # ------------------------------------------------------------ MEDAKA_COMMAND=( medaka_consensus -i "$FASTQ" -d "$DRAFT" -o "$OUTDIR" -t "$THREADS" ) if [[ -n "$MODEL" ]]; then MEDAKA_COMMAND+=( -m "$MODEL" ) fi echo "============================================================" echo "🔧 Running Medaka" echo "============================================================" echo echo "Command:" printf " %q" "\${MEDAKA_COMMAND[@]}" printf "\\n\\n" # ------------------------------------------------------------ # Run Medaka # ------------------------------------------------------------ "\${MEDAKA_COMMAND[@]}" # ------------------------------------------------------------ # Confirm output # ------------------------------------------------------------ if [[ ! -e "$CONSENSUS" ]]; then die "Medaka finished without creating consensus.fasta." fi if [[ ! -f "$CONSENSUS" ]]; then die "The expected consensus output is not a regular file." fi if [[ ! -s "$CONSENSUS" ]]; then die "Medaka created consensus.fasta, but the file is empty." fi echo echo "============================================================" echo "✅ Medaka polishing completed successfully" echo "============================================================" echo echo "📄 Polished consensus FASTA:" echo " $CONSENSUS" echo ls -lh "$CONSENSUS" echo echo "The original draft FASTA was not modified." echo ) SCRIPT_STATUS=$? echo if [[ "$SCRIPT_STATUS" -ne 0 ]]; then echo "============================================================" echo "❌ Medaka did not complete successfully" echo "============================================================" echo echo "Review the error message above." echo "Your WSL terminal will remain open." echo # Preserve the error status when this is run as a saved script, # without closing an interactive terminal when pasted directly. if [[ $- != *i* ]]; then exit "$SCRIPT_STATUS" fi else echo "============================================================" echo "✅ Medaka script finished" echo "============================================================" fi `; } // ============================================================ // Interface elements // ============================================================ const textarea = html` `; const copyBtn = html` `; const controlsColumn = html`

Medaka polishing
Enter the Nanopore reads, draft assembly, and output-folder paths. The generated WSL script will update automatically.

${fastqInput} ${draftFasta} ${outputDir} ${condaEnv} ${model}

${threads} ${runVersionReport}

Copy the generated script, paste it into WSL, and press Enter.

${copyBtn}

`; const scriptColumn = html`

Generated WSL script Updates automatically

${textarea}

`; const panel = html`

🧬 Generate Medaka polishing WSL script

${controlsColumn} ${scriptColumn}

`; // ============================================================ // Keep generated script synchronized with the controls // ============================================================ function refreshScript() { const script = buildScript(); textarea.value = script; panel.value = script; const missing = getMissingInputs(); const inputsComplete = missing.length === 0; copyBtn.disabled = !inputsComplete; copyBtn.style.opacity = inputsComplete ? "1" : "0.6"; copyBtn.style.cursor = inputsComplete ? "pointer" : "not-allowed"; panel.dispatchEvent( new Event("input", { bubbles: true }) ); } const controls = [ fastqInput, draftFasta, outputDir, condaEnv, model, threads, runVersionReport ]; for (const control of controls) { control.addEventListener("input", refreshScript); control.addEventListener("change", refreshScript); } // ============================================================ // Copy button // ============================================================ copyBtn.onclick = async () => { refreshScript(); const missing = getMissingInputs(); if (missing.length > 0) { copyBtn.textContent = "⚠️ Complete the paths first"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL Script"; }, 2500); return; } try { if ( navigator.clipboard && typeof navigator.clipboard.writeText === "function" ) { await navigator.clipboard.writeText(textarea.value); } else { textarea.focus(); textarea.select(); const copied = document.execCommand("copy"); if (!copied) { throw new Error("Browser copy command failed."); } } copyBtn.textContent = "✅ Copied!"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL Script"; }, 2000); } catch (error) { console.error(error); copyBtn.textContent = "❌ Copy failed"; setTimeout(() => { copyBtn.textContent = "📋 Copy WSL Script"; }, 3000); } }; refreshScript(); return panel; }47:T314d,viewof pharokkaInstallerWSL = { const envInput = Inputs.text({ label: "🧪 Conda environment", value: "pharokka", placeholder: "pharokka", style: "width:100%" }); const versionInput = Inputs.text({ label: "🏷️ Pharokka version", value: "", placeholder: "Leave blank for latest available from Bioconda", style: "width:100%" }); const databaseDirInput = Inputs.text({ label: "📚 Database directory (WSL/Linux path)", value: "~/pharokka_databases", placeholder: "~/pharokka_databases", style: "width:100%" }); const installDatabasesInput = Inputs.toggle({ label: "Download or verify the Pharokka databases", value: true }); const recreateEnvironmentInput = Inputs.toggle({ label: "Recreate the Conda environment if it already exists", value: false }); const textarea = html``; const copyButton = html``; const copyStatus = html``; /* * Left side: installer controls */ const controlsPanel = html`

`; const controlsHeading = html`

Installer controls

`; const controlsBody = html`

`; controlsBody.append( envInput, versionInput, databaseDirInput, installDatabasesInput, recreateEnvironmentInput, html`

The default installs Pharokka in its own environment and stores the databases in your WSL home directory. Recreating the environment removes the existing environment with the same name before installation.

` ); controlsPanel.append(controlsHeading, controlsBody); /* * Right side: generated WSL code */ const outputPanel = html`

`; const outputHeader = html`

Generated WSL code

${copyStatus} ${copyButton}

`; outputPanel.append(outputHeader, textarea); /* * Controls and generated code appear side by side, * but are contained in one collapsible section. */ const contentGrid = html`

`; contentGrid.append(controlsPanel, outputPanel); /* * One collapsible section containing everything */ const arrow = html`▶`; const summary = html`

${arrow}🧬 Pharokka installer generator — WSL

`; const detailsBody = html`

`; detailsBody.append(contentGrid); const details = html`

`; details.append(summary, detailsBody); details.addEventListener("toggle", () => { arrow.textContent = details.open ? "▼" : "▶"; }); const root = html`

`; root.append(details); function shQuote(value) { return `'${String(value).replace(/'/g, `'\\''`)}'`; } function generateScript() { const envName = envInput.value.trim() || "pharokka"; const requestedVersion = versionInput.value.trim(); const databaseDir = databaseDirInput.value.trim() || "~/pharokka_databases"; const installDatabases = installDatabasesInput.value ? "yes" : "no"; const recreateEnvironment = recreateEnvironmentInput.value ? "yes" : "no"; const innerScript = `#!/usr/bin/env bash # 🧬 Install Pharokka in WSL using a dedicated Conda environment set -Eeuo pipefail ENV_NAME=${shQuote(envName)} PHAROKKA_VERSION=${shQuote(requestedVersion)} DB_DIR_INPUT=${shQuote(databaseDir)} INSTALL_DATABASES=${shQuote(installDatabases)} RECREATE_ENV=${shQuote(recreateEnvironment)} if [[ ! "$ENV_NAME" =~ ^[A-Za-z0-9._-]+$ ]]; then echo "❌ Invalid Conda environment name: $ENV_NAME" echo " Use only letters, numbers, periods, underscores, and hyphens." exit 1 fi if [ -n "$PHAROKKA_VERSION" ] && [[ ! "$PHAROKKA_VERSION" =~ ^[0-9A-Za-z._-]+$ ]]; then echo "❌ Invalid Pharokka version: $PHAROKKA_VERSION" exit 1 fi expand_user_path() { local raw="$1" case "$raw" in "~") printf '%s\\n' "$HOME" ;; "~/"*) printf '%s/%s\\n' "$HOME" "\${raw#~/}" ;; *) printf '%s\\n' "$raw" ;; esac } DB_DIR="$(expand_user_path "$DB_DIR_INPUT")" if [ -z "$DB_DIR" ]; then echo "❌ Database directory cannot be empty." exit 1 fi SAFE_ENV_NAME="$( printf '%s' "$ENV_NAME" | tr -c 'A-Za-z0-9._-' '_' )" LOG_FILE="$HOME/install_pharokka_\${SAFE_ENV_NAME}.log" : > "$LOG_FILE" exec > >(tee -a "$LOG_FILE") 2>&1 finish() { local status="$1" trap - ERR echo echo "📄 Installation log:" echo " $LOG_FILE" echo echo "Press Enter to close..." read -r _ /dev/null; then echo "✅ WSL detected." else echo "⚠️ WSL was not detected from /proc/version." echo " The script can still run on a standard Linux system." fi echo echo "💽 Available space on the home filesystem:" df -h "$HOME" | awk 'NR == 1 || NR == 2 {print}' echo echo "🔎 Locating Conda..." export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" hash -r if ! command -v conda >/dev/null 2>&1; then for conda_sh in \\ "$HOME/miniforge3/etc/profile.d/conda.sh" \\ "$HOME/mambaforge/etc/profile.d/conda.sh" \\ "$HOME/miniconda3/etc/profile.d/conda.sh" \\ "$HOME/anaconda3/etc/profile.d/conda.sh" \\ "/opt/conda/etc/profile.d/conda.sh"; do if [ -f "$conda_sh" ]; then # shellcheck disable=SC1090 source "$conda_sh" break fi done fi if ! command -v conda >/dev/null 2>&1; then echo "❌ Conda was not found." echo " Install Miniforge first, reopen WSL, and run this installer again." finish 1 fi echo "✅ Conda found: $(command -v conda)" conda --version conda deactivate >/dev/null 2>&1 || true PACKAGE_SPEC="pharokka" if [ -n "$PHAROKKA_VERSION" ]; then PACKAGE_SPEC="pharokka=$PHAROKKA_VERSION" fi CONDA_COMMON=( --yes --override-channels --strict-channel-priority -c conda-forge -c bioconda ) if conda create --help 2>&1 | grep -q -- '--solver'; then CONDA_COMMON+=(--solver libmamba) echo "✅ Using Conda's libmamba solver." else echo "ℹ️ This Conda version does not expose --solver." echo " Conda's default solver will be used." fi env_exists() { conda env list | awk 'NF && $1 !~ /^#/ {print $1}' | grep -Fxq "$ENV_NAME" } if [ "$RECREATE_ENV" = "yes" ] && env_exists; then echo echo "♻️ Removing the existing environment: $ENV_NAME" conda env remove \ -n "$ENV_NAME" \ --yes fi echo if env_exists; then echo "🧪 Environment '$ENV_NAME' already exists." echo " Installing or updating $PACKAGE_SPEC in that environment..." conda install \ -n "$ENV_NAME" \ "\${CONDA_COMMON[@]}" \ "$PACKAGE_SPEC" else echo "🧪 Creating environment '$ENV_NAME' with $PACKAGE_SPEC..." conda create \ -n "$ENV_NAME" \ "\${CONDA_COMMON[@]}" \ "$PACKAGE_SPEC" fi CONDA_RUN=( conda run -n "$ENV_NAME" ) if conda run --help 2>&1 | grep -q -- '--no-capture-output'; then CONDA_RUN=( conda run --no-capture-output -n "$ENV_NAME" ) fi echo echo "🔍 Verifying Pharokka and core command-line dependencies..." "\${CONDA_RUN[@]}" \ pharokka --version "\${CONDA_RUN[@]}" \ pharokka --help \ >/dev/null "\${CONDA_RUN[@]}" \ pharokka run --help \ >/dev/null "\${CONDA_RUN[@]}" \ pharokka install --help \ >/dev/null "\${CONDA_RUN[@]}" bash -c ' set -e for command_name in pharokka mmseqs mash minced aragorn; do command -v "$command_name" >/dev/null printf "✅ %-12s %s\\n" \\ "$command_name" \\ "$(command -v "$command_name")" done ' if [ "$INSTALL_DATABASES" = "yes" ]; then echo echo "📚 Installing or checking the Pharokka databases..." echo " Destination: $DB_DIR" mkdir -p "$DB_DIR" "\${CONDA_RUN[@]}" \ pharokka install \ -o "$DB_DIR" DB_FILE_COUNT="$( find "$DB_DIR" \ -maxdepth 1 \ -type f | wc -l | tr -d ' ' )" DB_SIZE="$( du -sh "$DB_DIR" | awk '{print $1}' )" echo echo "✅ Database check completed." echo " Files: $DB_FILE_COUNT" echo " Size: $DB_SIZE" else echo echo "⏭️ Database installation was skipped." echo " Install later with:" echo echo " conda run -n $ENV_NAME pharokka install -o '$DB_DIR'" fi echo echo "============================================================" echo "✅ Pharokka installation complete" echo "============================================================" echo echo "Environment:" echo " $ENV_NAME" echo echo "Activate it later with:" echo " conda activate $ENV_NAME" echo echo "Check the version with:" echo " pharokka --version" echo echo "Run an annotation with:" echo " pharokka run -i genome.fasta -o pharokka_output -d '$DB_DIR' -t 8" finish 0`; textarea.value = `bash <<'PHAROKKA_INSTALL_SCRIPT' ${innerScript} PHAROKKA_INSTALL_SCRIPT`; root.value = textarea.value; copyStatus.textContent = ""; } for (const control of [ envInput, versionInput, databaseDirInput, installDatabasesInput, recreateEnvironmentInput ]) { control.addEventListener("input", generateScript); control.addEventListener("change", generateScript); } copyButton.addEventListener("click", async () => { try { await navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; copyStatus.textContent = ""; setTimeout(() => { copyButton.textContent = "📋 Copy WSL installer"; }, 1500); } catch (error) { textarea.focus(); textarea.select(); const copied = document.execCommand("copy"); copyStatus.textContent = copied ? "Copied" : "Select and copy manually"; } }); generateScript(); return root; }48:T5f16,viewof installPholdWSL = { const environmentInput = Inputs.text({ label: "Conda environment name", value: "phold", placeholder: "phold", style: "width:100%" }); const condaBaseInput = Inputs.text({ label: "Conda base directory — optional WSL/Linux path", value: "", placeholder: "Leave blank to detect Conda automatically", style: "width:100%" }); const versionInput = Inputs.text({ label: "Phold version — optional", value: "", placeholder: "Leave blank for the latest compatible version", style: "width:100%" }); const pythonVersionInput = Inputs.select( new Map([ ["Python 3.11 — recommended", "3.11"], ["Python 3.12", "3.12"], ["Python 3.13", "3.13"], ["Automatic — newest compatible", "auto"] ]), { label: "Python version", value: "3.11" } ); const pytorchModeInput = Inputs.select( new Map([ ["CPU-only PyTorch — smaller installation", "cpu"], ["NVIDIA CUDA PyTorch — use an NVIDIA GPU", "cuda"] ]), { label: "PyTorch installation", value: "cpu" } ); const databaseInput = Inputs.text({ label: "Phold database directory — WSL/Linux path", value: "~/phold_databases", placeholder: "~/phold_databases", style: "width:100%" }); const threadsInput = Inputs.range([1, 32], { value: 8, step: 1, label: "Database download and preparation threads" }); const installDatabaseInput = Inputs.toggle({ label: "Install or verify the Phold database and ProstT5 model", value: true }); const foldseekGpuDatabaseInput = Inputs.toggle({ label: "Prepare the database for Foldseek GPU acceleration", value: false }); const extendedDatabaseInput = Inputs.toggle({ label: "Install the extended 3.16-million-protein database", value: false }); const recreateEnvironmentInput = Inputs.toggle({ label: "Remove and recreate an existing environment", value: false }); const lowResourceInput = Inputs.toggle({ label: "Use low-resource Conda installation settings", value: true }); const cleanCacheInput = Inputs.toggle({ label: "Clean unused Conda package caches after installation", value: false }); const informationBox = html`

`; const gpuInformationBox = html`

`; const textarea = html``; const copyButton = html``; const downloadButton = html``; const copyStatus = html``; const controlsPanel = html`

`; const controlsHeading = html`

Phold installation settings

`; const controlsBody = html`

`; controlsBody.append( environmentInput, condaBaseInput, versionInput, pythonVersionInput, pytorchModeInput, gpuInformationBox, databaseInput, threadsInput, installDatabaseInput, foldseekGpuDatabaseInput, extendedDatabaseInput, recreateEnvironmentInput, lowResourceInput, cleanCacheInput, informationBox ); controlsPanel.append(controlsHeading, controlsBody); const outputPanel = html`

`; const outputHeader = html`

Generated WSL code

${copyStatus} ${copyButton} ${downloadButton}

`; outputPanel.append(outputHeader, textarea); const contentGrid = html`

`; contentGrid.append(controlsPanel, outputPanel); const arrow = html`▶`; const summary = html`

${arrow}🧬 Install Phold — WSL

`; const detailsBody = html`

`; detailsBody.append(contentGrid); const details = html`

`; details.append(summary, detailsBody); details.addEventListener("toggle", () => { arrow.textContent = details.open ? "▼" : "▶"; }); const root = html`

`; root.append(details); function shQuote(value) { return `'${String(value).replace(/'/g, `'\\''`)}'`; } function updateDisplay() { const cudaMode = pytorchModeInput.value === "cuda"; foldseekGpuDatabaseInput.style.display = cudaMode ? "" : "none"; if (!cudaMode && foldseekGpuDatabaseInput.value) { foldseekGpuDatabaseInput.value = false; } if (cudaMode) { gpuInformationBox.innerHTML = ` CUDA mode follows Phold's documented Conda installation by adding pytorch=*=cuda*. The generated script requires NVIDIA GPU access through WSL and verifies that PyTorch can see the GPU before declaring the installation successful. `; } else { gpuInformationBox.innerHTML = ` CPU mode explicitly requests a CPU PyTorch build. It avoids the large CUDA libraries and is the most portable installation choice. Phold can still process individual phage genomes on CPU. `; } informationBox.innerHTML = ` The script binds all commands to one detected Conda base and runs the environment by its full prefix. This prevents another Conda installation from being selected later. Paths beginning with ~/ are expanded correctly to the user's WSL home directory. `; } function generateScript() { const environmentName = environmentInput.value.trim() || "phold"; const condaBase = condaBaseInput.value.trim(); const pholdVersion = versionInput.value.trim(); const pythonVersion = pythonVersionInput.value || "3.11"; const pytorchMode = pytorchModeInput.value || "cpu"; const databaseDirectory = databaseInput.value.trim() || "~/phold_databases"; const threads = Math.max(1, Math.min(32, Number(threadsInput.value) || 8)); const installDatabases = installDatabaseInput.value ? "yes" : "no"; const foldseekGpuDatabase = pytorchMode === "cuda" && foldseekGpuDatabaseInput.value ? "yes" : "no"; const extendedDatabase = extendedDatabaseInput.value ? "yes" : "no"; const recreateEnvironment = recreateEnvironmentInput.value ? "yes" : "no"; const lowResourceMode = lowResourceInput.value ? "yes" : "no"; const cleanCache = cleanCacheInput.value ? "yes" : "no"; if (!environmentName) { textarea.value = ""; copyButton.disabled = true; downloadButton.disabled = true; return; } copyButton.disabled = false; downloadButton.disabled = false; const innerScript = `#!/usr/bin/env bash # 🧬 Install Phold in WSL using one explicitly selected Conda installation set -Eeuo pipefail ENV_NAME=${shQuote(environmentName)} CONDA_BASE_INPUT=${shQuote(condaBase)} PHOLD_VERSION=${shQuote(pholdVersion)} PYTHON_VERSION=${shQuote(pythonVersion)} PYTORCH_MODE=${shQuote(pytorchMode)} DB_DIR_INPUT=${shQuote(databaseDirectory)} THREADS=${threads} INSTALL_DATABASES=${shQuote(installDatabases)} FOLDSEEK_GPU_DB=${shQuote(foldseekGpuDatabase)} EXTENDED_DATABASE=${shQuote(extendedDatabase)} RECREATE_ENV=${shQuote(recreateEnvironment)} LOW_RESOURCE_MODE=${shQuote(lowResourceMode)} CLEAN_CONDA_CACHE=${shQuote(cleanCache)} LOG_FILE="$HOME/install_phold_\${ENV_NAME}.log" CONDA_BASE="" CONDA_EXE="" ENV_PREFIX="" DB_DIR="" finish() { local status="$1" trap - ERR echo echo "📄 Installation log:" echo " $LOG_FILE" echo echo "Press Enter to close..." read -r _ "$LOG_FILE" exec > >(tee -a "$LOG_FILE") 2>&1 expand_user_path() { local raw="$1" case "$raw" in "~") printf '%s\\n' "$HOME" ;; "~/"*) printf '%s/%s\\n' "$HOME" "\${raw:2}" ;; *) printf '%s\\n' "$raw" ;; esac } is_valid_conda_base() { local candidate="$1" [ -n "$candidate" ] && [ -x "$candidate/bin/conda" ] } find_conda_base() { local candidate="" local detected="" if [ -n "$CONDA_BASE_INPUT" ]; then candidate="$(expand_user_path "$CONDA_BASE_INPUT")" if ! is_valid_conda_base "$candidate"; then echo "❌ The selected Conda base is not valid:" echo " $candidate" echo echo "Expected executable:" echo " $candidate/bin/conda" return 1 fi printf '%s\\n' "$candidate" return 0 fi if command -v conda >/dev/null 2>&1; then detected="$(conda info --base 2>/dev/null || true)" if is_valid_conda_base "$detected"; then printf '%s\\n' "$detected" return 0 fi fi for candidate in \\ "$HOME/miniforge3" \\ "$HOME/mambaforge" \\ "$HOME/miniconda3" \\ "$HOME/miniconda" \\ "$HOME/anaconda3" \\ "/opt/conda"; do if is_valid_conda_base "$candidate"; then printf '%s\\n' "$candidate" return 0 fi done return 1 } if [[ ! "$ENV_NAME" =~ ^[A-Za-z0-9._-]+$ ]]; then echo "❌ Invalid Conda environment name:" echo " $ENV_NAME" echo echo "Use only letters, numbers, periods, underscores and hyphens." finish 1 fi if [ -n "$PHOLD_VERSION" ] && [[ ! "$PHOLD_VERSION" =~ ^[0-9A-Za-z._+-]+$ ]]; then echo "❌ Invalid Phold version:" echo " $PHOLD_VERSION" finish 1 fi if [ "$PYTHON_VERSION" != "auto" ] && [[ ! "$PYTHON_VERSION" =~ ^3\\.[0-9]+$ ]]; then echo "❌ Invalid Python version:" echo " $PYTHON_VERSION" finish 1 fi if [ "$PYTORCH_MODE" != "cpu" ] && [ "$PYTORCH_MODE" != "cuda" ]; then echo "❌ Unknown PyTorch mode:" echo " $PYTORCH_MODE" finish 1 fi if ! [[ "$THREADS" =~ ^[1-9][0-9]*$ ]]; then echo "❌ Threads must be a positive integer." finish 1 fi DB_DIR="$(expand_user_path "$DB_DIR_INPUT")" if [ -z "$DB_DIR" ]; then echo "❌ The database directory cannot be empty." finish 1 fi if [ "$LOW_RESOURCE_MODE" = "yes" ]; then export CONDA_VERIFY_THREADS=1 export CONDA_EXECUTE_THREADS=1 export CONDA_DOWNLOAD_THREADS=1 export OMP_NUM_THREADS=1 export OPENBLAS_NUM_THREADS=1 export MKL_NUM_THREADS=1 export NUMEXPR_NUM_THREADS=1 export TOKENIZERS_PARALLELISM=false fi echo "============================================================" echo " Phold installer for WSL" echo "============================================================" echo echo "Environment name: $ENV_NAME" echo "Phold request: \${PHOLD_VERSION:-latest compatible}" echo "Python version: $PYTHON_VERSION" echo "PyTorch mode: $PYTORCH_MODE" echo "Database directory: $DB_DIR" echo "Install database: $INSTALL_DATABASES" echo "Database threads: $THREADS" echo "Foldseek GPU database: $FOLDSEEK_GPU_DB" echo "Extended database: $EXTENDED_DATABASE" echo "Recreate environment: $RECREATE_ENV" echo "Low-resource mode: $LOW_RESOURCE_MODE" echo if ! grep -qi microsoft /proc/version 2>/dev/null; then echo "⚠️ WSL was not detected." echo " This script is intended for Ubuntu running under WSL." else echo "✅ WSL detected." fi echo echo "💽 Available space on the home filesystem:" df -h "$HOME" | tail -n 2 echo echo "🔎 Locating Conda..." export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" hash -r CONDA_BASE="$(find_conda_base)" || { echo "❌ Conda was not found." echo echo "Install Miniforge or Miniconda and reopen WSL." finish 1 } CONDA_EXE="$CONDA_BASE/bin/conda" ENV_PREFIX="$CONDA_BASE/envs/$ENV_NAME" echo "✅ Conda base selected:" echo " $CONDA_BASE" echo echo "Conda executable:" echo " $CONDA_EXE" "$CONDA_EXE" --version if "$CONDA_EXE" create --help 2>&1 | grep -q -- '--solver'; then SOLVER_ARGUMENTS=(--solver libmamba) echo "✅ Using Conda's libmamba solver." else SOLVER_ARGUMENTS=() echo "ℹ️ Using Conda's configured solver." fi if [ -d "$ENV_PREFIX" ]; then if [ "$RECREATE_ENV" = "yes" ]; then echo echo "🗑️ Removing the existing environment:" echo " $ENV_PREFIX" if [ -f "$ENV_PREFIX/conda-meta/history" ]; then "$CONDA_EXE" env remove \\ --yes \\ --prefix "$ENV_PREFIX" else echo "⚠️ The directory is not a complete Conda environment." echo " Removing the incomplete directory." rm -rf -- "$ENV_PREFIX" fi elif [ ! -f "$ENV_PREFIX/conda-meta/history" ] || [ ! -x "$ENV_PREFIX/bin/python" ]; then echo echo "⚠️ An incomplete environment directory was found:" echo " $ENV_PREFIX" echo " Removing it before installation." rm -rf -- "$ENV_PREFIX" fi fi if [ ! -d "$ENV_PREFIX" ]; then echo echo "🧪 Creating the Phold environment..." echo " Prefix: $ENV_PREFIX" CREATE_ARGUMENTS=( create --yes --prefix "$ENV_PREFIX" --override-channels --strict-channel-priority -c conda-forge -c bioconda ) CREATE_ARGUMENTS+=("\${SOLVER_ARGUMENTS[@]}") if [ "$PYTHON_VERSION" != "auto" ]; then CREATE_ARGUMENTS+=("python=$PYTHON_VERSION") fi if [ -n "$PHOLD_VERSION" ]; then CREATE_ARGUMENTS+=("phold=$PHOLD_VERSION") else CREATE_ARGUMENTS+=(phold) fi if [ "$PYTORCH_MODE" = "cuda" ]; then CREATE_ARGUMENTS+=("pytorch=*=cuda*") else CREATE_ARGUMENTS+=("pytorch=*=cpu*") fi if ! "$CONDA_EXE" "\${CREATE_ARGUMENTS[@]}"; then echo echo "❌ Conda could not create the Phold environment." if [ -d "$ENV_PREFIX" ]; then echo "Removing the incomplete environment directory..." rm -rf -- "$ENV_PREFIX" fi finish 1 fi else echo echo "✅ Existing Conda environment found:" echo " $ENV_PREFIX" echo echo "The existing environment is being retained." echo "Enable 'Remove and recreate' in Observable to change its" echo "Python version or CPU/GPU PyTorch build." fi CONDA_RUN=( "$CONDA_EXE" run --prefix "$ENV_PREFIX" ) if "$CONDA_EXE" run --help 2>&1 | grep -q -- '--no-capture-output'; then CONDA_RUN=( "$CONDA_EXE" run --no-capture-output --prefix "$ENV_PREFIX" ) fi echo echo "🔍 Verifying Phold and its dependencies..." if [ ! -x "$ENV_PREFIX/bin/phold" ]; then echo "❌ The phold executable was not created:" echo " $ENV_PREFIX/bin/phold" finish 1 fi if [ ! -x "$ENV_PREFIX/bin/foldseek" ]; then echo "❌ The foldseek executable was not created:" echo " $ENV_PREFIX/bin/foldseek" finish 1 fi CLI_VERSION="$("\${CONDA_RUN[@]}" phold --version 2>&1 | head -n 1)" PACKAGE_VERSION="$("\${CONDA_RUN[@]}" python -c \ 'import importlib.metadata as m; print(m.version("phold"))')" PYTHON_ACTUAL="$("\${CONDA_RUN[@]}" python -c \ 'import platform; print(platform.python_version())')" FOLDSEEK_VERSION="$("\${CONDA_RUN[@]}" foldseek version 2>&1 | head -n 1)" echo "✅ Phold command:" echo " $ENV_PREFIX/bin/phold" echo echo "Phold CLI version:" echo " $CLI_VERSION" echo echo "Phold package metadata version:" echo " $PACKAGE_VERSION" echo echo "Python:" echo " $PYTHON_ACTUAL" echo echo "Foldseek:" echo " $FOLDSEEK_VERSION" if [[ "$CLI_VERSION" != *"$PACKAGE_VERSION"* ]]; then echo echo "⚠️ The CLI and package metadata report different versions." echo " This can result from an outdated internal version string." echo " The installation will continue because both commands work." fi echo echo "🔍 Verifying PyTorch hardware access..." PYTORCH_REPORT="$("\${CONDA_RUN[@]}" python - <<'PY' import torch print(f"PyTorch version: {torch.__version__}") print(f"CUDA available: {torch.cuda.is_available()}") if torch.cuda.is_available(): print(f"CUDA device: {torch.cuda.get_device_name(0)}") PY )" printf '%s\\n' "$PYTORCH_REPORT" CUDA_AVAILABLE="$(printf '%s\\n' "$PYTORCH_REPORT" | awk -F': ' '/^CUDA available:/ {print tolower($2)}')" if [ "$PYTORCH_MODE" = "cuda" ]; then if ! command -v nvidia-smi >/dev/null 2>&1; then echo echo "❌ NVIDIA CUDA mode was selected, but nvidia-smi was not found." echo " Confirm that the NVIDIA Windows driver and WSL GPU support" echo " are installed, or recreate the environment in CPU mode." finish 1 fi if [ "$CUDA_AVAILABLE" != "true" ]; then echo echo "❌ CUDA PyTorch was requested, but PyTorch cannot access a GPU." echo " Recreate the environment in CPU mode or repair WSL GPU access." finish 1 fi echo echo "✅ CUDA is available to PyTorch." else if [ "$CUDA_AVAILABLE" = "true" ]; then echo echo "⚠️ CPU mode was requested, but CUDA is available in PyTorch." echo " Phold will still work. Recreate the environment if a strictly" echo " CPU-only installation is required." else echo echo "✅ CPU-only PyTorch verified." fi fi if [ "$INSTALL_DATABASES" = "yes" ]; then echo echo "📚 Installing or verifying the Phold database and ProstT5 model..." echo " Destination: $DB_DIR" mkdir -p "$DB_DIR" INSTALL_ARGUMENTS=( phold install -d "$DB_DIR" -t "$THREADS" ) if [ "$FOLDSEEK_GPU_DB" = "yes" ]; then INSTALL_ARGUMENTS+=(--foldseek_gpu) fi if [ "$EXTENDED_DATABASE" = "yes" ]; then INSTALL_ARGUMENTS+=(--extended_db) fi "\${CONDA_RUN[@]}" "\${INSTALL_ARGUMENTS[@]}" else echo echo "ℹ️ Database installation was skipped." if [ ! -d "$DB_DIR" ] || [ -z "$(find "$DB_DIR" -mindepth 1 -print -quit 2>/dev/null)" ]; then echo "⚠️ The selected database directory is missing or empty:" echo " $DB_DIR" fi fi if [ "$INSTALL_DATABASES" = "yes" ]; then if [ ! -d "$DB_DIR" ] || [ -z "$(find "$DB_DIR" -mindepth 1 -print -quit 2>/dev/null)" ]; then echo echo "❌ Phold finished without creating database files:" echo " $DB_DIR" finish 1 fi if [ ! -s "$DB_DIR/acrs_plddt_over_70_metadata.tsv" ]; then echo echo "❌ A required Phold database metadata file is missing:" echo " $DB_DIR/acrs_plddt_over_70_metadata.tsv" finish 1 fi DATABASE_FILE_COUNT="$(find "$DB_DIR" -type f | wc -l | tr -d ' ')" DATABASE_SIZE="$(du -sh "$DB_DIR" | awk '{print $1}')" echo echo "✅ Phold database verification completed." echo " Files: $DATABASE_FILE_COUNT" echo " Size: $DATABASE_SIZE" fi if [ "$CLEAN_CONDA_CACHE" = "yes" ]; then echo echo "🧹 Cleaning unused Conda package caches..." "$CONDA_EXE" clean \\ --all \\ --yes || true fi echo echo "============================================================" echo "✅ Phold installation complete" echo "============================================================" echo echo "Conda base:" echo " $CONDA_BASE" echo echo "Environment prefix:" echo " $ENV_PREFIX" echo echo "Database directory:" echo " $DB_DIR" echo echo "Run Phold later with:" echo " '$CONDA_EXE' run --no-capture-output --prefix '$ENV_PREFIX' phold --version" echo echo "Example CPU run:" echo " '$CONDA_EXE' run --no-capture-output --prefix '$ENV_PREFIX' phold run -i pharokka.gbk -o phold_output -d '$DB_DIR' -t $THREADS --cpu" if [ "$PYTORCH_MODE" = "cuda" ]; then echo echo "Example GPU-enabled ProstT5 run:" echo " '$CONDA_EXE' run --no-capture-output --prefix '$ENV_PREFIX' phold run -i pharokka.gbk -o phold_output -d '$DB_DIR' -t $THREADS" if [ "$FOLDSEEK_GPU_DB" = "yes" ]; then echo echo "Example ProstT5 plus Foldseek GPU run:" echo " '$CONDA_EXE' run --no-capture-output --prefix '$ENV_PREFIX' phold run -i pharokka.gbk -o phold_output -d '$DB_DIR' -t $THREADS --foldseek_gpu" fi fi finish 0`; textarea.value = `bash <<'PHOLD_INSTALL_SCRIPT' ${innerScript} PHOLD_INSTALL_SCRIPT`; root.value = textarea.value; copyStatus.textContent = ""; } for (const control of [ environmentInput, condaBaseInput, versionInput, pythonVersionInput, pytorchModeInput, databaseInput, threadsInput, installDatabaseInput, foldseekGpuDatabaseInput, extendedDatabaseInput, recreateEnvironmentInput, lowResourceInput, cleanCacheInput ]) { control.addEventListener("input", () => { updateDisplay(); generateScript(); }); control.addEventListener("change", () => { updateDisplay(); generateScript(); }); } copyButton.addEventListener("click", async () => { try { await navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; copyStatus.textContent = ""; setTimeout(() => { copyButton.textContent = "📋 Copy Phold installer"; }, 1500); } catch (error) { textarea.focus(); textarea.select(); const copied = document.execCommand("copy"); copyStatus.textContent = copied ? "Copied" : "Select the code and copy it manually"; } }); downloadButton.addEventListener("click", () => { const blob = new Blob([textarea.value], { type: "text/x-shellscript;charset=utf-8" }); const objectUrl = URL.createObjectURL(blob); const link = Object.assign(document.createElement("a"), { href: objectUrl, download: "install_phold.sh" }); document.body.appendChild(link); link.click(); link.remove(); setTimeout(() => { URL.revokeObjectURL(objectUrl); }, 1000); }); updateDisplay(); generateScript(); return root; }49:T5420,viewof phyntenyInstaller = { let pi_condaBase = Inputs.text({ label: "Conda base directory — optional WSL/Linux path", value: "", placeholder: "Leave blank to search ~/miniconda3, ~/miniconda, Miniforge, and active Conda", style: "width:100%" }); let pi_envName = Inputs.text({ label: "Conda environment name", value: "phynteny_transformer", placeholder: "phynteny_transformer", style: "width:100%" }); let pi_computeMode = Inputs.select( new Map([ ["Auto-detect NVIDIA GPU — recommended", "auto"], ["Require NVIDIA GPU", "gpu"], ["CPU-compatible installation", "cpu"] ]), { label: "PyTorch installation mode", value: "auto" } ); let pi_modelRoot = Inputs.text({ label: "Phynteny model download folder — WSL/Linux path", value: "~/phynteny_transformer_models", placeholder: "~/phynteny_transformer_models", style: "width:100%" }); let pi_repairExisting = Inputs.toggle({ label: "Repair and update an existing environment", value: true }); let pi_recreate = Inputs.toggle({ label: "Delete and recreate the environment from scratch", value: false }); let pi_installModels = Inputs.toggle({ label: "Install or verify the Phynteny model files", value: true }); let pi_openModels = Inputs.toggle({ label: "Open the model folder in Windows Explorer when finished", value: true }); let infoBox = html`

`; let textarea = html``; let copyButton = html``; let downloadButton = html``; let copyStatus = html``; let controlsPanel = html`

`; let controlsHeading = html`

Phynteny Transformer installation settings

`; let controlsBody = html`

`; controlsBody.append( pi_condaBase, pi_envName, pi_computeMode, pi_modelRoot, pi_repairExisting, pi_recreate, pi_installModels, pi_openModels, infoBox ); controlsPanel.append(controlsHeading, controlsBody); let outputPanel = html`

`; let outputHeader = html`

Generated WSL installer code

${copyStatus} ${copyButton} ${downloadButton}

`; outputPanel.append(outputHeader, textarea); let contentGrid = html`

`; contentGrid.append(controlsPanel, outputPanel); let arrow = html`▶`; let summary = html`

${arrow}🧬 Install or repair Phynteny Transformer — WSL

`; let detailsBody = html`

`; detailsBody.append(contentGrid); let details = html`

`; details.append(summary, detailsBody); details.addEventListener("toggle", () => { arrow.textContent = details.open ? "▼" : "▶"; }); let root = html`

`; root.append(details); function shQuote(value) { return `'${String(value).replace(/'/g, `'\\''`)}'`; } function updateInformation() { let computeMode = pi_computeMode.value || "auto"; let recreate = Boolean(pi_recreate.value); let computeText = ""; if (computeMode === "gpu") { computeText = `The installer requires a CUDA-enabled PyTorch build and stops if CUDA is unavailable.`; } else if (computeMode === "cpu") { computeText = `The installer creates a CPU-compatible environment. The Phynteny runner can still be changed later if a GPU build is installed.`; } else { computeText = `The installer requests CUDA-enabled PyTorch when an NVIDIA driver is detected, with a standard PyTorch fallback if CUDA solving fails.`; } infoBox.innerHTML = ` Installer safeguards
Python is pinned to 3.11. The missing requests dependency is installed explicitly, and setuptools<81 keeps the pkg_resources API available for Phynteny v0.1.3. ${computeText}

${ recreate ? `Warning: the selected environment will be deleted before installation.` : `An existing environment is retained unless the recreate option is enabled.` } `; } function generateScript() { let condaBaseInput = pi_condaBase.value.trim(); let envName = pi_envName.value.trim() || "phynteny_transformer"; let computeMode = pi_computeMode.value || "auto"; let modelRoot = pi_modelRoot.value.trim() || "~/phynteny_transformer_models"; let repairExisting = pi_repairExisting.value ? "yes" : "no"; let recreateEnvironment = pi_recreate.value ? "yes" : "no"; let installModels = pi_installModels.value ? "yes" : "no"; let openModels = pi_openModels.value ? "yes" : "no"; if (!envName || !modelRoot) { textarea.value = ""; copyButton.disabled = true; downloadButton.disabled = true; copyStatus.textContent = ""; return; } copyButton.disabled = false; downloadButton.disabled = false; let innerScript = `#!/usr/bin/env bash # 🧬 Install or repair Phynteny Transformer in WSL set -Eeuo pipefail CONDA_BASE_INPUT=${shQuote(condaBaseInput)} ENV_NAME=${shQuote(envName)} COMPUTE_MODE=${shQuote(computeMode)} MODEL_ROOT_INPUT=${shQuote(modelRoot)} REPAIR_EXISTING=${shQuote(repairExisting)} RECREATE_ENVIRONMENT=${shQuote(recreateEnvironment)} INSTALL_MODELS=${shQuote(installModels)} OPEN_MODELS=${shQuote(openModels)} CONDA_BASE="" CONDA_EXE="" ENV_PREFIX="" MODEL_ROOT="" MODELS_DIR="" finish() { local status="$1" trap - ERR echo echo "Press Enter to close..." read -r _ /dev/null 2>&1; then active_base="$(conda info --base 2>/dev/null || true)" add_conda_candidate "$active_base" fi } choose_conda_base() { local candidate="" if [ -n "$CONDA_BASE_INPUT" ]; then CONDA_BASE="$(expand_user_path "$CONDA_BASE_INPUT")" elif is_valid_conda_base "$HOME/miniconda3"; then CONDA_BASE="$HOME/miniconda3" elif [ "\${#CONDA_CANDIDATES[@]}" -gt 0 ]; then CONDA_BASE="\${CONDA_CANDIDATES[0]}" else echo "❌ No usable Conda installation was found." return 1 fi CONDA_EXE="$CONDA_BASE/bin/conda" ENV_PREFIX="$CONDA_BASE/envs/$ENV_NAME" for candidate in "\${CONDA_CANDIDATES[@]}"; do if [ -f "$candidate/envs/$ENV_NAME/conda-meta/history" ]; then CONDA_BASE="$candidate" CONDA_EXE="$candidate/bin/conda" ENV_PREFIX="$candidate/envs/$ENV_NAME" break fi done } conda_supports_libmamba() { "$CONDA_EXE" create --help 2>&1 | grep -q -- '--solver' } build_create_command() { CREATE_COMMAND=( "$CONDA_EXE" create --yes --prefix "$ENV_PREFIX" --override-channels --strict-channel-priority -c conda-forge -c bioconda ) if conda_supports_libmamba; then CREATE_COMMAND+=(--solver libmamba) fi } build_install_command() { INSTALL_COMMAND=( "$CONDA_EXE" install --yes --prefix "$ENV_PREFIX" --override-channels --strict-channel-priority -c conda-forge -c bioconda ) if "$CONDA_EXE" install --help 2>&1 | grep -q -- '--solver'; then INSTALL_COMMAND+=(--solver libmamba) fi } base_packages() { printf '%s\\n' \\ 'python=3.11' \\ 'phynteny_transformer=0.1.3' \\ 'requests' \\ 'setuptools<81' \\ 'pip' } create_environment() { local -a packages=() local cuda_requested="no" mapfile -t packages < <(base_packages) if [ "$COMPUTE_MODE" != "cpu" ] && command -v nvidia-smi >/dev/null 2>&1; then cuda_requested="yes" fi build_create_command echo echo "📦 Creating the Phynteny Transformer environment..." echo " Conda base: $CONDA_BASE" echo " Environment: $ENV_PREFIX" if [ "$cuda_requested" = "yes" ]; then echo " Requesting CUDA-enabled PyTorch." if ! "\${CREATE_COMMAND[@]}" \\ "\${packages[@]}" \\ 'pytorch=*=cuda*'; then if [ "$COMPUTE_MODE" = "gpu" ]; then echo echo "❌ CUDA environment creation failed and GPU mode is required." return 1 fi echo echo "⚠️ CUDA environment creation failed." echo " Retrying with the standard PyTorch package..." rm -rf -- "$ENV_PREFIX" "\${CREATE_COMMAND[@]}" \\ "\${packages[@]}" \\ pytorch fi else "\${CREATE_COMMAND[@]}" \\ "\${packages[@]}" \\ pytorch fi } repair_environment() { local -a packages=() mapfile -t packages < <(base_packages) build_install_command echo echo "🔧 Repairing and updating the existing environment..." if [ "$COMPUTE_MODE" != "cpu" ] && command -v nvidia-smi >/dev/null 2>&1; then if ! "\${INSTALL_COMMAND[@]}" \\ "\${packages[@]}" \\ 'pytorch=*=cuda*'; then if [ "$COMPUTE_MODE" = "gpu" ]; then echo echo "❌ CUDA repair failed and GPU mode is required." return 1 fi echo echo "⚠️ CUDA repair failed." echo " Retrying with the standard PyTorch package..." "\${INSTALL_COMMAND[@]}" \\ "\${packages[@]}" \\ pytorch fi else "\${INSTALL_COMMAND[@]}" \\ "\${packages[@]}" \\ pytorch fi } resolve_models_directory() { local candidate="" if [ -d "$MODEL_ROOT" ] && find "$MODEL_ROOT" \\ -maxdepth 1 \\ -type f \\ -name '*.model' \\ -print -quit | grep -q .; then MODELS_DIR="$MODEL_ROOT" return 0 fi candidate="$MODEL_ROOT/models" if [ -d "$candidate" ] && find "$candidate" \\ -maxdepth 1 \\ -type f \\ -name '*.model' \\ -print -quit | grep -q .; then MODELS_DIR="$candidate" return 0 fi MODELS_DIR="" return 1 } if [[ ! "$ENV_NAME" =~ ^[A-Za-z0-9._-]+$ ]]; then echo "❌ Invalid Conda environment name:" echo " $ENV_NAME" finish 1 fi if [ "$COMPUTE_MODE" != "auto" ] && [ "$COMPUTE_MODE" != "gpu" ] && [ "$COMPUTE_MODE" != "cpu" ]; then echo "❌ Unknown compute mode: $COMPUTE_MODE" finish 1 fi MODEL_ROOT="$(expand_user_path "$MODEL_ROOT_INPUT")" MODEL_ROOT="$(normalize_directory_path "$MODEL_ROOT")" export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" hash -r echo "============================================================" echo " Phynteny Transformer installer and repair tool" echo "============================================================" echo echo "Conda environment:" echo " $ENV_NAME" echo echo "Compute mode:" echo " $COMPUTE_MODE" echo echo "Model folder:" echo " $MODEL_ROOT" echo echo "Python target:" echo " 3.11" echo collect_conda_candidates || finish 1 choose_conda_base || finish 1 echo "✅ Conda base:" echo " $CONDA_BASE" echo echo "Environment prefix:" echo " $ENV_PREFIX" echo "$CONDA_EXE" --version if [ "$RECREATE_ENVIRONMENT" = "yes" ] && [ -f "$ENV_PREFIX/conda-meta/history" ]; then echo echo "🗑️ Removing the existing environment..." "$CONDA_EXE" env remove \\ --yes \\ --prefix "$ENV_PREFIX" fi if [ ! -f "$ENV_PREFIX/conda-meta/history" ]; then create_environment || finish 1 elif [ "$REPAIR_EXISTING" = "yes" ]; then repair_environment || finish 1 else echo echo "ℹ️ Existing environment retained without package repair." fi if [ ! -x "$ENV_PREFIX/bin/python" ] || [ ! -x "$ENV_PREFIX/bin/phynteny_transformer" ] || [ ! -x "$ENV_PREFIX/bin/install_models" ]; then echo echo "❌ The environment does not contain all required commands." finish 1 fi CONDA_RUN=( "$CONDA_EXE" run --prefix "$ENV_PREFIX" ) if "$CONDA_EXE" run --help 2>&1 | grep -q -- '--no-capture-output'; then CONDA_RUN=( "$CONDA_EXE" run --no-capture-output --prefix "$ENV_PREFIX" ) fi echo echo "🔍 Verifying Phynteny and required Python packages..." "\${CONDA_RUN[@]}" phynteny_transformer --version "\${CONDA_RUN[@]}" install_models --help >/dev/null "\${CONDA_RUN[@]}" python - <<'PY' import sys import requests import torch import Bio import transformers import sklearn import pandas import numpy import pkg_resources print(f"Python: {sys.version.split()[0]}") print(f"requests: {requests.__version__}") print(f"PyTorch: {torch.__version__}") print(f"Biopython: {Bio.__version__}") print(f"Transformers: {transformers.__version__}") print(f"scikit-learn: {sklearn.__version__}") print(f"pandas: {pandas.__version__}") print(f"NumPy: {numpy.__version__}") print("pkg_resources: available") print(f"CUDA available: {torch.cuda.is_available()}") if torch.cuda.is_available(): print(f"GPU: {torch.cuda.get_device_name(0)}") PY CUDA_AVAILABLE="$("\${CONDA_RUN[@]}" python - <<'PY' import torch print("yes" if torch.cuda.is_available() else "no") PY )" if [ "$COMPUTE_MODE" = "gpu" ] && [ "$CUDA_AVAILABLE" != "yes" ]; then echo echo "❌ GPU mode was selected, but PyTorch cannot access CUDA." finish 1 fi if [ "$COMPUTE_MODE" = "auto" ]; then if [ "$CUDA_AVAILABLE" = "yes" ]; then echo "✅ Auto mode successfully installed GPU-enabled PyTorch." else echo "ℹ️ Auto mode installed a CPU-compatible PyTorch build." fi elif [ "$COMPUTE_MODE" = "gpu" ]; then echo "✅ CUDA is available as required." else echo "ℹ️ CPU-compatible installation selected." fi if [ "$INSTALL_MODELS" = "yes" ]; then if resolve_models_directory; then echo echo "✅ Existing Phynteny model files found." else echo echo "📚 Installing Phynteny model files..." echo " Destination: $MODEL_ROOT" mkdir -p "$MODEL_ROOT" "\${CONDA_RUN[@]}" \\ install_models \\ -o "$MODEL_ROOT" if ! resolve_models_directory; then echo echo "❌ Model installation finished, but no .model files were found." echo " Checked:" echo " $MODEL_ROOT" echo " $MODEL_ROOT/models" finish 1 fi fi MODEL_COUNT="$(find "$MODELS_DIR" -maxdepth 1 -type f -name '*.model' | wc -l)" echo " Models directory:" echo " $MODELS_DIR" echo " Model files: $MODEL_COUNT" if [ "$MODEL_COUNT" -lt 10 ]; then echo echo "⚠️ Fewer than 10 .model files were found." echo " The installation may be incomplete." finish 1 fi else echo echo "ℹ️ Model installation was skipped." fi echo echo "============================================================" echo " Installation summary" echo "============================================================" echo echo "Environment:" echo " $ENV_PREFIX" echo echo "Run command:" echo " $CONDA_EXE run --prefix $ENV_PREFIX phynteny_transformer --version" if [ -n "$MODELS_DIR" ]; then echo echo "Models directory:" echo " $MODELS_DIR" fi if [ "$OPEN_MODELS" = "yes" ] && [ -d "$MODEL_ROOT" ]; then WINDOWS_MODEL_ROOT="$(wslpath -w "$MODEL_ROOT" 2>/dev/null || true)" if command -v explorer.exe >/dev/null 2>&1; then echo echo "📂 Opening the model folder in Windows Explorer..." explorer.exe "$WINDOWS_MODEL_ROOT" >/dev/null 2>&1 || true elif [ -x /mnt/c/Windows/explorer.exe ]; then echo echo "📂 Opening the model folder in Windows Explorer..." /mnt/c/Windows/explorer.exe "$WINDOWS_MODEL_ROOT" \\ >/dev/null 2>&1 || true else echo echo "⚠️ Windows Explorer could not be opened automatically." fi fi echo echo "✅ Phynteny Transformer is installed and verified." finish 0`; textarea.value = `bash <<'PHYNTENY_INSTALL_SCRIPT' ${innerScript} PHYNTENY_INSTALL_SCRIPT`; root.value = textarea.value; copyStatus.textContent = ""; } for (let control of [ pi_condaBase, pi_envName, pi_computeMode, pi_modelRoot, pi_repairExisting, pi_recreate, pi_installModels, pi_openModels ]) { control.addEventListener("input", () => { updateInformation(); generateScript(); }); control.addEventListener("change", () => { updateInformation(); generateScript(); }); } copyButton.addEventListener("click", async () => { try { await navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; copyStatus.textContent = ""; setTimeout(() => { copyButton.textContent = "📋 Copy installer script"; }, 1500); } catch (error) { textarea.focus(); textarea.select(); let copied = document.execCommand("copy"); copyStatus.textContent = copied ? "Copied" : "Select the code and copy it manually"; } }); downloadButton.addEventListener("click", () => { let blob = new Blob([textarea.value], { type: "text/x-shellscript;charset=utf-8" }); let objectUrl = URL.createObjectURL(blob); let link = Object.assign(document.createElement("a"), { href: objectUrl, download: "install_phynteny.sh" }); document.body.appendChild(link); link.click(); link.remove(); setTimeout(() => { URL.revokeObjectURL(objectUrl); }, 1000); }); updateInformation(); generateScript(); return root; }4a:T5e9a,viewof pharokkaRunner = { const ph_mode = Inputs.select( new Map([ ["Run one FASTA file", "single"], ["Run FASTA files from a parent folder", "folder"] ]), { label: "Input mode", value: "single" } ); const ph_singlePath = Inputs.text({ label: "Windows FASTA file path", value: "C:\\Users\\kiera\\Downloads\\Seldom.fasta", placeholder: "C:\\Users\\kiera\\Downloads\\Seldom.fasta", style: "width:100%" }); const ph_parentPath = Inputs.text({ label: "Windows parent folder path", value: "C:\\Users\\kiera\\Downloads\\Phage_Genomes", placeholder: "C:\\Users\\kiera\\Downloads\\Phage_Genomes", style: "width:100%" }); const ph_recursive = Inputs.toggle({ label: "Search all nested subfolders recursively", value: true }); const ph_includeParent = Inputs.toggle({ label: "Also include FASTA files directly inside the parent folder", value: false }); const ph_outputMode = Inputs.select( new Map([ ["Use a user-specified Windows output folder", "custom"], ["Create a Pharokka folder beside each FASTA file", "beside"] ]), { label: "Output location", value: "custom" } ); const ph_outputPath = Inputs.text({ label: "Windows output parent folder", value: "C:\\Users\\kiera\\Downloads\\Pharokka_Results", placeholder: "C:\\Users\\kiera\\Downloads\\Pharokka_Results", style: "width:100%" }); const ph_env = Inputs.text({ label: "Conda environment", value: "pharokka", placeholder: "pharokka", style: "width:100%" }); const ph_database = Inputs.text({ label: "Pharokka database directory (WSL/Linux path)", value: "~/pharokka_databases", placeholder: "~/pharokka_databases", style: "width:100%" }); const ph_geneCaller = Inputs.select( new Map([ ["PHANOTATE — recommended for complete phage genomes", "phanotate"], ["Prodigal — conservative prokaryotic gene prediction", "prodigal"], [ "Prodigal-gv — viruses and possible alternative genetic codes", "prodigal-gv" ] ]), { label: "Gene predictor", value: "phanotate" } ); const ph_fast = Inputs.toggle({ label: "Use fast mode — skips slower PHROG HMM searches", value: false }); const ph_open = Inputs.toggle({ label: "Open results in Windows Explorer", value: true }); const ph_threads = Inputs.range([1, 16], { value: 8, step: 1, label: "Threads used for each Pharokka analysis" }); const singlePathPanel = html`

`; singlePathPanel.append(ph_singlePath); const folderPathPanel = html`

`; folderPathPanel.append(ph_parentPath, ph_recursive, ph_includeParent); const modeInformation = html`

`; const outputPathPanel = html`

`; outputPathPanel.append(ph_outputPath); const outputInformation = html`

`; const textarea = html``; const copyBtn = html``; const dlBtn = html``; const copyStatus = html``; const controlsPanel = html`

`; const controlsHeading = html`

Pharokka run settings

`; const controlsBody = html`

`; controlsBody.append( ph_mode, singlePathPanel, folderPathPanel, modeInformation, ph_outputMode, outputPathPanel, outputInformation, ph_env, ph_database, ph_geneCaller, ph_fast, ph_threads, ph_open, html`

Full mode is recommended when the Pharokka output will be passed to Phold and Phynteny. Fast mode finishes more quickly but omits the more-sensitive PHROG HMM search.

` ); controlsPanel.append(controlsHeading, controlsBody); const outputPanel = html`

`; const outputHeader = html`

Generated WSL code

${copyStatus} ${copyBtn} ${dlBtn}

`; outputPanel.append(outputHeader, textarea); const contentGrid = html`

`; contentGrid.append(controlsPanel, outputPanel); const arrow = html`▶`; const summary = html`

${arrow}🧬 Pharokka annotation runner — WSL

`; const detailsBody = html`

`; detailsBody.append(contentGrid); const details = html`

`; details.append(summary, detailsBody); details.addEventListener("toggle", () => { arrow.textContent = details.open ? "▼" : "▶"; }); const root = html`

`; singlePathPanel.append(phold_singlePath); const folderPathPanel = html`

`; folderPathPanel.append(phold_parentPath, phold_recursive, phold_filePattern); const modeInformation = html`

`; const computeInformation = html`

`; const databaseInformation = html`

`; const textarea = html``; const copyButton = html``; const downloadButton = html``; const copyStatus = html``; const controlsPanel = html`

`; const controlsHeading = html`

Phold run settings

`; const controlsBody = html`

`; controlsBody.append( phold_mode, singlePathPanel, folderPathPanel, modeInformation, phold_outputPath, phold_condaBase, phold_env, phold_database, databaseInformation, phold_computeMode, phold_foldseekGpu, phold_autotune, phold_batchSize, phold_threads, computeInformation, phold_hypotheticalsOnly, phold_checkDatabase, phold_force, phold_open ); controlsPanel.append(controlsHeading, controlsBody); const outputPanel = html`

`; const outputHeader = html`

Generated WSL code

${copyStatus} ${copyButton} ${downloadButton}

`; outputPanel.append(outputHeader, textarea); const contentGrid = html`

`; contentGrid.append(controlsPanel, outputPanel); const arrow = html`▶`; const summary = html`

${arrow}🧬 Phold annotation runner — WSL

`; const detailsBody = html`

`; detailsBody.append(contentGrid); const details = html`

`; details.append(summary, detailsBody); details.addEventListener("toggle", () => { arrow.textContent = details.open ? "▼" : "▶"; }); const root = html`

`; singlePathPanel.append(pt_singlePath); let folderPathPanel = html`

`; folderPathPanel.append(pt_parentPath, pt_recursive, pt_filePattern); let modeInformation = html`

`; let environmentInformation = html`

`; let modelInformation = html`

`; let computeInformation = html`

`; let textarea = html``; let copyButton = html``; let downloadButton = html``; let copyStatus = html``; let controlsPanel = html`

`; let controlsHeading = html`

Phynteny Transformer run settings

`; let controlsBody = html`

`; controlsBody.append( pt_mode, singlePathPanel, folderPathPanel, modeInformation, pt_outputPath, pt_condaBase, pt_env, pt_installEnvironment, pt_repairEnvironment, environmentInformation, pt_modelRoot, pt_checkModels, modelInformation, pt_computeMode, pt_esmModel, pt_batchSize, pt_confidence, pt_cpuThreads, computeInformation, pt_force, pt_debug, pt_attention, pt_open ); controlsPanel.append(controlsHeading, controlsBody); let outputPanel = html`

`; let outputHeader = html`

Generated WSL code

${copyStatus} ${copyButton} ${downloadButton}

`; outputPanel.append(outputHeader, textarea); let contentGrid = html`

`; contentGrid.append(controlsPanel, outputPanel); let arrow = html`▶`; let summary = html`

${arrow}🧬 Phynteny Transformer runner — WSL

`; let detailsBody = html`

`; detailsBody.append(contentGrid); let details = html`

`; details.append(summary, detailsBody); details.addEventListener("toggle", () => { arrow.textContent = details.open ? "▼" : "▶"; }); let root = html`

`; root.append(details); function toWslPath(pathValue) { return String(pathValue || "") .trim() .replace( /^([A-Za-z]):[\\/]/, (_, drive) => `/mnt/${drive.toLowerCase()}/` ) .replaceAll("\\", "/"); } function shQuote(value) { return `'${String(value).replace(/'/g, `'\\''`)}'`; } function updateDisplay() { let folderMode = pt_mode.value === "folder"; let computeMode = pt_computeMode.value || "auto"; singlePathPanel.style.display = folderMode ? "none" : "block"; folderPathPanel.style.display = folderMode ? "flex" : "none"; if (folderMode) { modeInformation.innerHTML = ` The script searches for phold.gbk files and processes every genome separately. Relative source folders are preserved under the timestamped Phynteny result folder. `; } else { modeInformation.innerHTML = ` Select the phold.gbk produced by Phold. The final output folder should contain phynteny.gbk and phynteny.tsv. `; } environmentInformation.innerHTML = ` The generated script searches all common Conda installations and binds to the full environment prefix. A new environment uses Python 3.11, Phynteny 0.1.3, requests, and setuptools<81. GPU installations also include cuda-cudart-dev so Triton can compile against cuda.h. Before analysis, the runner performs a real CUDA/Triton compilation test and falls back to CPU only in Auto mode when GPU compilation remains unavailable. `; modelInformation.innerHTML = ` Enter the model download root, not the final models subfolder. For example, ~/phynteny_transformer_models resolves to $HOME/phynteny_transformer_models, and the runner supplies its models subfolder to Phynteny. `; if (computeMode === "gpu") { computeInformation.innerHTML = ` GPU mode stops with a clear error unless CUDA is available through PyTorch. A starting batch size of 32 is safer than 128 for the 650-million-parameter ESM2 model. `; } else if (computeMode === "cpu") { computeInformation.innerHTML = ` CPU mode hides CUDA devices and limits common numerical libraries to the selected CPU-thread count. It will be slower than GPU inference. `; } else { computeInformation.innerHTML = ` Auto mode uses CUDA when PyTorch can access it and otherwise continues on CPU. The selected CPU-thread value is applied in either case. `; } } function generateScript() { let inputMode = pt_mode.value || "single"; let singleWindowsPath = pt_singlePath.value.trim(); let parentWindowsPath = pt_parentPath.value.trim(); let outputWindowsPath = pt_outputPath.value.trim(); let singleWslPath = toWslPath(singleWindowsPath); let parentWslPath = toWslPath(parentWindowsPath); let outputWslPath = toWslPath(outputWindowsPath); let condaBaseInput = pt_condaBase.value.trim(); let envName = pt_env.value.trim() || "phynteny_transformer"; let modelRoot = pt_modelRoot.value.trim() || "~/phynteny_transformer_models"; let computeMode = pt_computeMode.value || "auto"; let esmModel = pt_esmModel.value || "facebook/esm2_t33_650M_UR50D"; let filePattern = pt_filePattern.value || "phold"; let safeBatchSize = Math.max( 1, Math.min(128, Number(pt_batchSize.value) || 32) ); let safeConfidence = Math.max( 0, Math.min(1, Number(pt_confidence.value) || 0.8) ); let safeCpuThreads = Math.max( 1, Math.min(16, Number(pt_cpuThreads.value) || 8) ); let recursiveSearch = pt_recursive.value ? "yes" : "no"; let installEnvironment = pt_installEnvironment.value ? "yes" : "no"; let repairEnvironment = pt_repairEnvironment.value ? "yes" : "no"; let checkModels = pt_checkModels.value ? "yes" : "no"; let forceOutput = pt_force.value ? "yes" : "no"; let debugMode = pt_debug.value ? "yes" : "no"; let saveAttention = pt_attention.value ? "yes" : "no"; let openExplorer = pt_open.value ? "yes" : "no"; let selectedPath = inputMode === "folder" ? parentWindowsPath : singleWindowsPath; if (!selectedPath || !outputWindowsPath) { textarea.value = ""; copyButton.disabled = true; downloadButton.disabled = true; copyStatus.textContent = ""; return; } copyButton.disabled = false; downloadButton.disabled = false; let innerScript = `#!/usr/bin/env bash # 🧬 Run Phynteny Transformer on one GenBank file or a folder of GenBank files set -Eeuo pipefail INPUT_MODE=${shQuote(inputMode)} SINGLE_GENBANK=${shQuote(singleWslPath)} PARENT_DIR=${shQuote(parentWslPath)} OUTPUT_ROOT_INPUT=${shQuote(outputWslPath)} CONDA_BASE_INPUT=${shQuote(condaBaseInput)} ENV_NAME=${shQuote(envName)} INSTALL_ENVIRONMENT=${shQuote(installEnvironment)} REPAIR_ENVIRONMENT=${shQuote(repairEnvironment)} MODEL_ROOT_INPUT=${shQuote(modelRoot)} CHECK_MODELS=${shQuote(checkModels)} COMPUTE_MODE=${shQuote(computeMode)} EFFECTIVE_COMPUTE_MODE=${shQuote(computeMode)} ESM_MODEL=${shQuote(esmModel)} FILE_PATTERN=${shQuote(filePattern)} BATCH_SIZE=${safeBatchSize} CONFIDENCE_THRESHOLD=${safeConfidence} CPU_THREADS=${safeCpuThreads} RECURSIVE_SEARCH=${shQuote(recursiveSearch)} FORCE_OUTPUT=${shQuote(forceOutput)} DEBUG_MODE=${shQuote(debugMode)} SAVE_ATTENTION=${shQuote(saveAttention)} OPEN_EXPLORER=${shQuote(openExplorer)} CONDA_BASE="" CONDA_EXE="" ENV_PREFIX="" MODEL_ROOT="" MODELS_DIR="" CUDA_HEADER_DIR="" OUTPUT_ROOT="" RUN_ROOT="" RESULTS_ROOT="" LOGS_ROOT="" SUMMARY_FILE="" TOTAL_COUNT=0 SUCCESS_COUNT=0 FAILED_COUNT=0 finish() { local status="$1" trap - ERR echo echo "Press Enter to close..." read -r _ /dev/null 2>&1; then active_base="$(conda info --base 2>/dev/null || true)" add_conda_candidate "$active_base" fi } select_existing_environment() { local candidate="" local candidate_prefix="" local selected_count=0 CONDA_BASE="" CONDA_EXE="" ENV_PREFIX="" for candidate in "\${CONDA_CANDIDATES[@]}"; do candidate_prefix="$candidate/envs/$ENV_NAME" if is_valid_environment "$candidate_prefix"; then selected_count=$((selected_count + 1)) if [ -z "$ENV_PREFIX" ]; then CONDA_BASE="$candidate" CONDA_EXE="$candidate/bin/conda" ENV_PREFIX="$candidate_prefix" fi fi done if [ "$selected_count" -gt 1 ]; then echo "⚠️ More than one '$ENV_NAME' environment was found." echo " Using the first matching environment:" echo " $ENV_PREFIX" echo echo "Enter a Conda base path in Observable to choose another one." fi [ -n "$ENV_PREFIX" ] } choose_install_base() { local candidate="" if [ -n "$CONDA_BASE_INPUT" ]; then CONDA_BASE="$(expand_user_path "$CONDA_BASE_INPUT")" elif is_valid_conda_base "$HOME/miniconda3"; then CONDA_BASE="$HOME/miniconda3" elif [ "\${#CONDA_CANDIDATES[@]}" -gt 0 ]; then CONDA_BASE="\${CONDA_CANDIDATES[0]}" else echo "❌ No usable Conda installation was found." return 1 fi CONDA_EXE="$CONDA_BASE/bin/conda" ENV_PREFIX="$CONDA_BASE/envs/$ENV_NAME" } create_phynteny_environment() { local use_cuda_spec="no" local create_arguments=() local base_packages=( 'python=3.11' 'phynteny_transformer=0.1.3' 'requests' 'setuptools<81' 'pip' ) choose_install_base || return 1 echo echo "📦 Creating the Phynteny Transformer environment..." echo " Conda base: $CONDA_BASE" echo " Environment: $ENV_PREFIX" create_arguments=( "$CONDA_EXE" create --yes --prefix "$ENV_PREFIX" --override-channels --strict-channel-priority -c conda-forge -c bioconda ) if "$CONDA_EXE" create --help 2>&1 | grep -q -- '--solver'; then create_arguments+=(--solver libmamba) fi if [ "$COMPUTE_MODE" != "cpu" ] && command -v nvidia-smi >/dev/null 2>&1; then use_cuda_spec="yes" fi if [ "$use_cuda_spec" = "yes" ]; then echo " Requesting a CUDA-enabled PyTorch build." if ! "\${create_arguments[@]}" \ "\${base_packages[@]}" \ 'pytorch=*=cuda*' \ 'cuda-cudart-dev'; then if [ "$COMPUTE_MODE" = "gpu" ]; then echo echo "❌ CUDA environment creation failed and GPU mode is required." return 1 fi echo echo "⚠️ CUDA environment creation failed." echo " Retrying with the standard PyTorch package..." rm -rf -- "$ENV_PREFIX" "\${create_arguments[@]}" \ "\${base_packages[@]}" \ pytorch fi else "\${create_arguments[@]}" \ "\${base_packages[@]}" \ pytorch fi if ! is_valid_environment "$ENV_PREFIX"; then echo "❌ The environment was created, but phynteny_transformer was not found." return 1 fi } build_repair_command() { REPAIR_COMMAND=( "$CONDA_EXE" install --yes --prefix "$ENV_PREFIX" --override-channels --strict-channel-priority -c conda-forge -c bioconda ) if "$CONDA_EXE" install --help 2>&1 | grep -q -- '--solver'; then REPAIR_COMMAND+=(--solver libmamba) fi } repair_phynteny_environment() { local use_cuda_spec="no" local base_packages=( 'python=3.11' 'phynteny_transformer=0.1.3' 'requests' 'setuptools<81' 'pip' ) build_repair_command if [ "$COMPUTE_MODE" != "cpu" ] && command -v nvidia-smi >/dev/null 2>&1; then use_cuda_spec="yes" fi echo echo "🔧 Repairing the Phynteny Transformer environment..." echo " Environment: $ENV_PREFIX" if [ "$use_cuda_spec" = "yes" ]; then if ! "\${REPAIR_COMMAND[@]}" \ "\${base_packages[@]}" \ 'pytorch=*=cuda*' \ 'cuda-cudart-dev'; then if [ "$COMPUTE_MODE" = "gpu" ]; then echo echo "❌ CUDA environment repair failed and GPU mode is required." return 1 fi echo echo "⚠️ CUDA repair failed." echo " Retrying with the standard PyTorch package..." "\${REPAIR_COMMAND[@]}" \ "\${base_packages[@]}" \ pytorch fi else "\${REPAIR_COMMAND[@]}" \ "\${base_packages[@]}" \ pytorch fi } phynteny_dependencies_available() { "\${CONDA_RUN[@]}" python - <<'PY' >/dev/null 2>&1 import sys import requests import torch import Bio import transformers import sklearn import pandas import numpy import pkg_resources if sys.version_info[:2] != (3, 11): raise SystemExit(1) PY } find_cuda_header() { local header="" header="$(find "$ENV_PREFIX" \ -type f \ -name 'cuda.h' \ -print -quit 2>/dev/null || true)" if [ -n "$header" ]; then CUDA_HEADER_DIR="$(dirname "$header")" return 0 fi CUDA_HEADER_DIR="" return 1 } configure_cuda_header_path() { if find_cuda_header; then export CPATH="$CUDA_HEADER_DIR\${CPATH:+:$CPATH}" export CPLUS_INCLUDE_PATH="$CUDA_HEADER_DIR\${CPLUS_INCLUDE_PATH:+:$CPLUS_INCLUDE_PATH}" echo "✅ CUDA header found:" echo " $CUDA_HEADER_DIR/cuda.h" return 0 fi return 1 } install_cuda_development_headers() { build_repair_command echo echo "🔧 Installing CUDA development headers for Triton..." "\${REPAIR_COMMAND[@]}" \ 'cuda-cudart-dev' configure_cuda_header_path } gpu_compile_test() { "\${CONDA_RUN[@]}" python - <<'PY' import torch if not torch.cuda.is_available(): raise SystemExit("CUDA is unavailable in PyTorch") def tiny_function(x): return torch.relu(x + 1.0) compiled_function = torch.compile( tiny_function, backend="inductor", ) tensor = torch.randn(32, device="cuda") result = compiled_function(tensor) torch.cuda.synchronize() if result.shape != tensor.shape: raise SystemExit("Unexpected CUDA compilation-test output") print("CUDA/Triton compilation test: passed") PY } resolve_models_directory() { local candidate="" if [ -d "$MODEL_ROOT" ] && find "$MODEL_ROOT" \ -maxdepth 1 \ -type f \ -name '*.model' \ -print -quit | grep -q .; then MODELS_DIR="$MODEL_ROOT" return 0 fi candidate="$MODEL_ROOT/models" if [ -d "$candidate" ] && find "$candidate" \ -maxdepth 1 \ -type f \ -name '*.model' \ -print -quit | grep -q .; then MODELS_DIR="$candidate" return 0 fi MODELS_DIR="$candidate" return 1 } if [[ ! "$ENV_NAME" =~ ^[A-Za-z0-9._-]+$ ]]; then echo "❌ Invalid Conda environment name:" echo " $ENV_NAME" finish 1 fi if ! [[ "$BATCH_SIZE" =~ ^[1-9][0-9]*$ ]]; then echo "❌ Batch size must be a positive integer." finish 1 fi if ! [[ "$CPU_THREADS" =~ ^[1-9][0-9]*$ ]]; then echo "❌ CPU threads must be a positive integer." finish 1 fi if ! awk -v value="$CONFIDENCE_THRESHOLD" \ 'BEGIN {exit !(value >= 0 && value <= 1)}'; then echo "❌ Confidence threshold must be between 0 and 1." finish 1 fi if [ "$COMPUTE_MODE" != "auto" ] && [ "$COMPUTE_MODE" != "gpu" ] && [ "$COMPUTE_MODE" != "cpu" ]; then echo "❌ Unknown compute mode: $COMPUTE_MODE" finish 1 fi MODEL_ROOT="$(expand_user_path "$MODEL_ROOT_INPUT")" MODEL_ROOT="$(normalize_directory_path "$MODEL_ROOT")" OUTPUT_ROOT="$(expand_user_path "$OUTPUT_ROOT_INPUT")" OUTPUT_ROOT="$(normalize_directory_path "$OUTPUT_ROOT")" if [ "$INPUT_MODE" = "folder" ]; then PARENT_DIR="$(normalize_directory_path "$PARENT_DIR")" fi export OMP_NUM_THREADS="$CPU_THREADS" export MKL_NUM_THREADS="$CPU_THREADS" export OPENBLAS_NUM_THREADS="$CPU_THREADS" export NUMEXPR_NUM_THREADS="$CPU_THREADS" if [ "$COMPUTE_MODE" = "cpu" ]; then export CUDA_VISIBLE_DEVICES="" fi echo "============================================================" echo " Phynteny Transformer runner" echo "============================================================" echo echo "Input mode:" echo " $INPUT_MODE" echo echo "Conda environment:" echo " $ENV_NAME" echo echo "Model root:" echo " $MODEL_ROOT" echo echo "Compute mode:" echo " $COMPUTE_MODE" echo echo "ESM model:" echo " $ESM_MODEL" echo echo "Prediction batch size:" echo " $BATCH_SIZE" echo echo "Confidence threshold:" echo " $CONFIDENCE_THRESHOLD" echo echo "CPU threads:" echo " $CPU_THREADS" echo echo "Output root:" echo " $OUTPUT_ROOT" echo if [ "$INPUT_MODE" = "single" ]; then if [ ! -f "$SINGLE_GENBANK" ]; then echo "❌ Input GenBank file was not found:" echo " $SINGLE_GENBANK" finish 1 fi elif [ "$INPUT_MODE" = "folder" ]; then if [ ! -d "$PARENT_DIR" ]; then echo "❌ Parent folder was not found:" echo " $PARENT_DIR" finish 1 fi else echo "❌ Unknown input mode: $INPUT_MODE" finish 1 fi mkdir -p "$OUTPUT_ROOT" export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH" hash -r echo "🔎 Locating the Phynteny Transformer Conda environment..." collect_conda_candidates || finish 1 if ! select_existing_environment; then if [ "$INSTALL_ENVIRONMENT" = "yes" ]; then create_phynteny_environment || finish 1 else echo echo "❌ The Conda environment '$ENV_NAME' was not found." echo echo "Enable automatic environment installation in Observable" echo "or run the Phynteny installer first." finish 1 fi fi echo "✅ Conda base:" echo " $CONDA_BASE" echo echo "✅ Environment prefix:" echo " $ENV_PREFIX" echo "$CONDA_EXE" --version CONDA_RUN=( "$CONDA_EXE" run --prefix "$ENV_PREFIX" ) if "$CONDA_EXE" run --help 2>&1 | grep -q -- '--no-capture-output'; then CONDA_RUN=( "$CONDA_EXE" run --no-capture-output --prefix "$ENV_PREFIX" ) fi echo echo "🔍 Checking the Phynteny environment..." NEEDS_REPAIR="no" if [ ! -x "$ENV_PREFIX/bin/install_models" ]; then NEEDS_REPAIR="yes" elif ! phynteny_dependencies_available; then NEEDS_REPAIR="yes" elif ! "\${CONDA_RUN[@]}" install_models --help >/dev/null 2>&1; then NEEDS_REPAIR="yes" fi if [ "$NEEDS_REPAIR" = "yes" ]; then if [ "$REPAIR_ENVIRONMENT" = "yes" ]; then repair_phynteny_environment || finish 1 if ! phynteny_dependencies_available; then echo echo "❌ Environment repair completed, but the required" echo " Python 3.11 Phynteny dependencies are still unavailable." finish 1 fi else echo echo "❌ The environment is missing one or more required dependencies" echo " or is not using Python 3.11." echo " Enable automatic environment repair in Observable" echo " or run the updated Phynteny installer." finish 1 fi fi echo echo "🔍 Verifying Phynteny Transformer..." "\${CONDA_RUN[@]}" phynteny_transformer --version "\${CONDA_RUN[@]}" phynteny_transformer --help >/dev/null "\${CONDA_RUN[@]}" install_models --help >/dev/null "\${CONDA_RUN[@]}" python - <<'PY' import sys import requests import torch import Bio import transformers import sklearn import pandas import numpy import pkg_resources print(f"Python: {sys.version.split()[0]}") print(f"requests: {requests.__version__}") print(f"PyTorch: {torch.__version__}") print(f"Biopython: {Bio.__version__}") print(f"Transformers: {transformers.__version__}") print(f"scikit-learn: {sklearn.__version__}") print(f"pandas: {pandas.__version__}") print(f"NumPy: {numpy.__version__}") print("pkg_resources: available") PY echo echo "🔍 Checking PyTorch hardware access..." HARDWARE_REPORT="$("\${CONDA_RUN[@]}" python - <<'PY' import torch print(f"PyTorch: {torch.__version__}") print(f"CUDA available: {torch.cuda.is_available()}") if torch.cuda.is_available(): print(f"GPU: {torch.cuda.get_device_name(0)}") PY )" printf '%s\n' "$HARDWARE_REPORT" CUDA_AVAILABLE="$("\${CONDA_RUN[@]}" python - <<'PY' import torch print("yes" if torch.cuda.is_available() else "no") PY )" EFFECTIVE_COMPUTE_MODE="$COMPUTE_MODE" if [ "$COMPUTE_MODE" = "cpu" ]; then export CUDA_VISIBLE_DEVICES="" EFFECTIVE_COMPUTE_MODE="cpu" echo "ℹ️ CPU-only mode selected." elif [ "$CUDA_AVAILABLE" != "yes" ]; then if [ "$COMPUTE_MODE" = "gpu" ]; then echo echo "❌ GPU mode was selected, but CUDA is unavailable in PyTorch." finish 1 fi export CUDA_VISIBLE_DEVICES="" EFFECTIVE_COMPUTE_MODE="cpu" echo "ℹ️ CUDA is unavailable; Auto mode will use the CPU." else echo echo "🔍 Checking CUDA headers and Triton compilation..." GPU_READY="yes" if ! configure_cuda_header_path; then GPU_READY="no" echo "⚠️ cuda.h was not found in the Conda environment." fi if [ "$GPU_READY" = "yes" ] && ! gpu_compile_test; then GPU_READY="no" echo "⚠️ The initial CUDA/Triton compilation test failed." fi if [ "$GPU_READY" != "yes" ] && [ "$REPAIR_ENVIRONMENT" = "yes" ]; then if install_cuda_development_headers && gpu_compile_test; then GPU_READY="yes" else GPU_READY="no" fi fi if [ "$GPU_READY" = "yes" ]; then EFFECTIVE_COMPUTE_MODE="gpu" echo "✅ CUDA and Triton are ready for Phynteny." elif [ "$COMPUTE_MODE" = "gpu" ]; then echo echo "❌ GPU mode was selected, but CUDA/Triton compilation failed." echo " Review the CUDA header and compiler messages above." finish 1 else echo echo "⚠️ GPU compilation is unavailable." echo " Auto mode will safely rerun Phynteny on the CPU." export CUDA_VISIBLE_DEVICES="" EFFECTIVE_COMPUTE_MODE="cpu" fi fi echo echo "Effective compute mode:" echo " $EFFECTIVE_COMPUTE_MODE" if ! resolve_models_directory; then if [ "$CHECK_MODELS" = "yes" ]; then echo echo "📚 Installing or repairing the Phynteny models..." echo " Download root: $MODEL_ROOT" mkdir -p "$MODEL_ROOT" "\${CONDA_RUN[@]}" \ install_models \ -o "$MODEL_ROOT" if ! resolve_models_directory; then echo "❌ Model installation completed, but no .model files were found." echo " Checked:" echo " $MODEL_ROOT" echo " $MODEL_ROOT/models" finish 1 fi else echo "❌ Phynteny model files were not found." echo " Expected .model files in:" echo " $MODEL_ROOT" echo " or" echo " $MODEL_ROOT/models" echo echo "Enable model checking in Observable to install them." finish 1 fi elif [ "$CHECK_MODELS" = "yes" ]; then echo echo "✅ Existing Phynteny models found." else echo echo "✅ Existing Phynteny models found; installation check skipped." fi echo " Models directory:" echo " $MODELS_DIR" MODEL_COUNT="$(find "$MODELS_DIR" -maxdepth 1 -type f -name '*.model' | wc -l)" echo " Model files: $MODEL_COUNT" if [ "$MODEL_COUNT" -lt 10 ]; then echo echo "❌ Fewer than 10 Phynteny model files were found." echo " Enable model checking to repair the model installation." finish 1 fi TIMESTAMP="$(date +%Y%m%d_%H%M%S)" if [ "$INPUT_MODE" = "single" ]; then INPUT_FILENAME="$(basename "$SINGLE_GENBANK")" INPUT_PARENT="$(basename "$(dirname "$SINGLE_GENBANK")")" if [ "$INPUT_FILENAME" = "phold.gbk" ]; then RUN_NAME="$INPUT_PARENT" else RUN_NAME="\${INPUT_FILENAME%.*}" fi else RUN_NAME="$(basename "$PARENT_DIR")" fi if [ -z "$RUN_NAME" ]; then RUN_NAME="phynteny" fi SAFE_RUN_NAME="$(sanitize_name "$RUN_NAME")" RUN_ROOT="$OUTPUT_ROOT/\${SAFE_RUN_NAME}_\${TIMESTAMP}" RESULTS_ROOT="$RUN_ROOT/results" LOGS_ROOT="$RUN_ROOT/logs" SUMMARY_FILE="$RUN_ROOT/phynteny_run_summary.tsv" mkdir -p "$RESULTS_ROOT" mkdir -p "$LOGS_ROOT" printf 'Status\\tInput_GenBank\\tOutput_Directory\\tGenBank_Output\\tTSV_Output\\tPredictions_Added\\tCritical_Log_Issues\\n' \ > "$SUMMARY_FILE" COMMON_ARGUMENTS=( -m "$MODELS_DIR" --esm_model "$ESM_MODEL" --batch-size "$BATCH_SIZE" --confidence-threshold "$CONFIDENCE_THRESHOLD" ) if [ "$FORCE_OUTPUT" = "yes" ]; then COMMON_ARGUMENTS+=(-f) fi if [ "$DEBUG_MODE" = "yes" ]; then COMMON_ARGUMENTS+=(--debug) fi if [ "$SAVE_ATTENTION" = "yes" ]; then COMMON_ARGUMENTS+=(--save-attention-weights) fi run_phynteny_file() { local input_file="$1" local relative_path="$2" local input_filename="" local input_stem="" local input_parent_name="" local genome_name="" local safe_genome_name="" local relative_directory="" local output_directory="" local log_directory="" local stdout_log="" local stderr_log="" local genbank_output="" local table_output="" local combined_log="" local critical_issue_count=0 local predictions_added="NA" local unknown_genes="NA" local status_line="" input_filename="$(basename "$input_file")" input_stem="\${input_filename%.*}" input_parent_name="$(basename "$(dirname "$input_file")")" if [ "$input_filename" = "phold.gbk" ]; then genome_name="$input_parent_name" else genome_name="$input_stem" fi if [ -z "$genome_name" ]; then genome_name="phage" fi safe_genome_name="$(sanitize_name "$genome_name")" relative_directory="$(dirname "$relative_path")" if [ "$INPUT_MODE" = "single" ]; then output_directory="$RESULTS_ROOT/$safe_genome_name" log_directory="$LOGS_ROOT" elif [ "$input_filename" = "phold.gbk" ]; then if [ "$relative_directory" = "." ]; then output_directory="$RESULTS_ROOT/$safe_genome_name" log_directory="$LOGS_ROOT" else output_directory="$RESULTS_ROOT/$relative_directory" log_directory="$LOGS_ROOT/$relative_directory" fi else if [ "$relative_directory" = "." ]; then output_directory="$RESULTS_ROOT/$safe_genome_name" log_directory="$LOGS_ROOT" else output_directory="$RESULTS_ROOT/$relative_directory/$safe_genome_name" log_directory="$LOGS_ROOT/$relative_directory" fi fi stdout_log="$log_directory/\${safe_genome_name}_stdout.log" stderr_log="$log_directory/\${safe_genome_name}_stderr.log" combined_log="$log_directory/\${safe_genome_name}_combined.log" genbank_output="$output_directory/phynteny.gbk" table_output="$output_directory/phynteny.tsv" mkdir -p "$log_directory" TOTAL_COUNT=$((TOTAL_COUNT + 1)) echo echo "============================================================" echo "🧬 Genome $TOTAL_COUNT: $genome_name" echo "============================================================" echo echo "Input:" echo " $input_file" echo echo "Output:" echo " $output_directory" echo if [ -d "$output_directory" ] && [ "$FORCE_OUTPUT" != "yes" ]; then echo "⚠️ Output folder already exists." echo " Enable overwrite in Observable to replace it." FAILED_COUNT=$((FAILED_COUNT + 1)) printf 'Skipped_existing_output\\t%s\\t%s\\t%s\\t%s\\t%s\\t%s\\n' \ "$input_file" \ "$output_directory" \ "$genbank_output" \ "$table_output" \ "NA" \ "NA" \ >> "$SUMMARY_FILE" return fi if "\${CONDA_RUN[@]}" \ phynteny_transformer \ "$input_file" \ -o "$output_directory" \ "\${COMMON_ARGUMENTS[@]}" \ > >(tee "$stdout_log") \ 2> >(tee "$stderr_log" >&2); then cat "$stdout_log" "$stderr_log" > "$combined_log" critical_issue_count="$(grep -Eic \ 'Error processing batch|Some genomes are missing in prepared data|Prediction missing or index out of bounds|Missing in X:|Missing in y:' \ "$combined_log" || true)" status_line="$(grep -E \ 'Statistics: [0-9]+ genes with unknown function, [0-9]+ predictions added' \ "$combined_log" | tail -n 1 || true)" if [ -n "$status_line" ]; then unknown_genes="$(printf '%s\n' "$status_line" | \ sed -E 's/.*Statistics: ([0-9]+) genes with unknown function, ([0-9]+) predictions added.*/\\1/')" predictions_added="$(printf '%s\n' "$status_line" | \ sed -E 's/.*Statistics: ([0-9]+) genes with unknown function, ([0-9]+) predictions added.*/\\2/')" fi if [ ! -s "$genbank_output" ] || [ ! -s "$table_output" ]; then FAILED_COUNT=$((FAILED_COUNT + 1)) echo echo "❌ Phynteny exited without creating both expected outputs." echo " Expected:" echo " $genbank_output" echo " $table_output" printf 'Missing_expected_outputs\t%s\t%s\t%s\t%s\t%s\t%s\n' \ "$input_file" \ "$output_directory" \ "$genbank_output" \ "$table_output" \ "$predictions_added" \ "$critical_issue_count" \ >> "$SUMMARY_FILE" elif [ "$critical_issue_count" -gt 0 ]; then FAILED_COUNT=$((FAILED_COUNT + 1)) cat > "$output_directory/PHYNTENY_OUTPUT_INVALID.txt" <> "$SUMMARY_FILE" else SUCCESS_COUNT=$((SUCCESS_COUNT + 1)) echo echo "✅ Phynteny completed and passed output validation:" echo " $genome_name" echo " Predictions added: $predictions_added" echo " Unknown genes reported: $unknown_genes" if [ "$predictions_added" = "0" ]; then echo "⚠️ No new functions were added, but no technical failure" echo " patterns were detected in the Phynteny logs." fi printf 'Success\t%s\t%s\t%s\t%s\t%s\t%s\n' \ "$input_file" \ "$output_directory" \ "$genbank_output" \ "$table_output" \ "$predictions_added" \ "$critical_issue_count" \ >> "$SUMMARY_FILE" fi else cat "$stdout_log" "$stderr_log" > "$combined_log" 2>/dev/null || true FAILED_COUNT=$((FAILED_COUNT + 1)) echo echo "❌ Phynteny failed:" echo " $genome_name" echo echo "Combined log:" echo " $combined_log" printf 'Failed\t%s\t%s\t%s\t%s\t%s\t%s\n' \ "$input_file" \ "$output_directory" \ "" \ "" \ "NA" \ "NA" \ >> "$SUMMARY_FILE" fi } if [ "$INPUT_MODE" = "single" ]; then run_phynteny_file \ "$SINGLE_GENBANK" \ "$(basename "$SINGLE_GENBANK")" else FIND_DEPTH_ARGUMENTS=(-mindepth 1) if [ "$RECURSIVE_SEARCH" != "yes" ]; then FIND_DEPTH_ARGUMENTS+=(-maxdepth 2) fi GENBANK_FILES=() if [ "$FILE_PATTERN" = "phold" ]; then mapfile -d '' -t GENBANK_FILES < <( find "$PARENT_DIR" \ "\${FIND_DEPTH_ARGUMENTS[@]}" \ -type f \ -iname 'phold.gbk' \ -print0 | sort -z ) else mapfile -d '' -t GENBANK_FILES < <( find "$PARENT_DIR" \ "\${FIND_DEPTH_ARGUMENTS[@]}" \ -type f \ \$ \ -iname '*.gbk' \ -o -iname '*.gbff' \ -o -iname '*.gb' \ \$ \ ! -iname 'phynteny.gbk' \ ! -iname 'phynteny_transformer.gbk' \ -print0 | sort -z ) fi if [ "\${#GENBANK_FILES[@]}" -eq 0 ]; then echo echo "❌ No matching GenBank files were found." echo echo "Searched folder:" echo " $PARENT_DIR" echo echo "File search mode:" echo " $FILE_PATTERN" finish 1 fi echo echo "🔍 GenBank files found:" echo " \${#GENBANK_FILES[@]}" for input_file in "\${GENBANK_FILES[@]}"; do relative_path="\${input_file#"$PARENT_DIR"/}" run_phynteny_file \ "$input_file" \ "$relative_path" done fi echo echo "============================================================" echo " Phynteny processing summary" echo "============================================================" echo echo "Total GenBank files:" echo " $TOTAL_COUNT" echo echo "Successful:" echo " $SUCCESS_COUNT" echo echo "Failed or skipped:" echo " $FAILED_COUNT" echo echo "Results root:" echo " $RESULTS_ROOT" echo echo "Log files:" echo " $LOGS_ROOT" echo echo "Summary table:" echo " $SUMMARY_FILE" echo if [ "$OPEN_EXPLORER" = "yes" ]; then WINDOWS_RUN_ROOT="$(wslpath -w "$RUN_ROOT" 2>/dev/null || true)" if command -v explorer.exe >/dev/null 2>&1; then echo "📂 Opening the results in Windows Explorer..." explorer.exe "$WINDOWS_RUN_ROOT" >/dev/null 2>&1 || true elif [ -x /mnt/c/Windows/explorer.exe ]; then echo "📂 Opening the results in Windows Explorer..." /mnt/c/Windows/explorer.exe "$WINDOWS_RUN_ROOT" \ >/dev/null 2>&1 || true else echo "⚠️ Windows Explorer could not be opened automatically." fi fi if [ "$FAILED_COUNT" -gt 0 ]; then echo echo "⚠️ Processing completed, but one or more files failed," echo " were skipped, or failed output validation." echo " Review the summary table and combined logs." finish 1 fi echo echo "✅ All Phynteny analyses completed and passed validation." finish 0`; textarea.value = `bash <<'PHYNTENY_RUN_SCRIPT' ${innerScript} PHYNTENY_RUN_SCRIPT`; root.value = textarea.value; copyStatus.textContent = ""; } for (let control of [ pt_mode, pt_singlePath, pt_parentPath, pt_recursive, pt_filePattern, pt_outputPath, pt_condaBase, pt_env, pt_installEnvironment, pt_repairEnvironment, pt_modelRoot, pt_checkModels, pt_computeMode, pt_esmModel, pt_batchSize, pt_confidence, pt_cpuThreads, pt_force, pt_debug, pt_attention, pt_open ]) { control.addEventListener("input", () => { updateDisplay(); generateScript(); }); control.addEventListener("change", () => { updateDisplay(); generateScript(); }); } copyButton.addEventListener("click", async () => { try { await navigator.clipboard.writeText(textarea.value); copyButton.textContent = "✅ Copied!"; copyStatus.textContent = ""; setTimeout(() => { copyButton.textContent = "📋 Copy Phynteny script"; }, 1500); } catch (error) { textarea.focus(); textarea.select(); let copied = document.execCommand("copy"); copyStatus.textContent = copied ? "Copied" : "Select the code and copy it manually"; } }); downloadButton.addEventListener("click", () => { let blob = new Blob([textarea.value], { type: "text/x-shellscript;charset=utf-8" }); let objectUrl = URL.createObjectURL(blob); let link = Object.assign(document.createElement("a"), { href: objectUrl, download: "run_phynteny.sh" }); document.body.appendChild(link); link.click(); link.remove(); setTimeout(() => { URL.revokeObjectURL(objectUrl); }, 1000); }); updateDisplay(); generateScript(); return root; }4d:T172e,viewof powershellRenameGenerator = { const parentFolderInput = Inputs.text({ label: "📁 Parent folder path", placeholder: "C:\\Users\\kiera\\Desktop\\Nanopore_Results", style: "width: 600px" }); const renameModeInput = Inputs.select( new Map([ ["Rename files only", "files"], ["Rename folders only", "folders"], ["Rename files and folders", "both"] ]), { label: "Rename mode", value: "files" } ); const searchSubfoldersInput = Inputs.toggle({ label: "🔍 Search inside subfolders", value: true }); const findTextInput = Inputs.text({ label: "Text to replace", placeholder: "filtered_PL27f_", style: "width: 420px" }); const replaceTextInput = Inputs.text({ label: "Replace with", placeholder: "", style: "width: 420px" }); const outputBox = html``; const copyButton = html``; function psQuote(s) { return `"${String(s).replace(/`/g, "``").replace(/"/g, '`"')}"`; } function updateScript() { const parentFolder = parentFolderInput.value.trim(); const mode = renameModeInput.value; const recursive = searchSubfoldersInput.value; const findText = findTextInput.value; const replaceText = replaceTextInput.value; if (!parentFolder || !findText) { outputBox.value = ""; copyButton.disabled = true; return; } copyButton.disabled = false; outputBox.value = `# Rename files/folders under one parent folder $ErrorActionPreference = "Stop" $ParentFolder = ${psQuote(parentFolder)} $Mode = ${psQuote(mode)} $SearchSubfolders = $${recursive ? "true" : "false"} $FindText = ${psQuote(findText)} $ReplaceWith = ${psQuote(replaceText)} if (!(Test-Path -LiteralPath $ParentFolder -PathType Container)) { Write-Host "ERROR: Parent folder not found:" -ForegroundColor Red Write-Host " $ParentFolder" Read-Host "Press Enter to exit" exit 1 } if ([string]::IsNullOrWhiteSpace($FindText)) { Write-Host "ERROR: Text to replace is empty." -ForegroundColor Red Read-Host "Press Enter to exit" exit 1 } Write-Host "" Write-Host "Parent folder: $ParentFolder" -ForegroundColor Cyan Write-Host "Mode: $Mode" Write-Host "Search subfolders: $SearchSubfolders" Write-Host "Replace: '$FindText' -> '$ReplaceWith'" Write-Host "" function Rename-MatchingItems { param( [string]$ItemType ) if ($ItemType -eq "files") { $items = Get-ChildItem -LiteralPath $ParentFolder -File -Recurse:$SearchSubfolders } if ($ItemType -eq "folders") { $items = Get-ChildItem -LiteralPath $ParentFolder -Directory -Recurse:$SearchSubfolders | Sort-Object FullName -Descending } if ($ItemType -ne "files" -and $ItemType -ne "folders") { throw "Unknown item type: $ItemType" } $matches = $items | Where-Object { $_.Name -like "*$FindText*" } if (!$matches -or $matches.Count -eq 0) { Write-Host "No matching $ItemType found." -ForegroundColor Yellow return } foreach ($item in $matches) { $newName = $item.Name.Replace($FindText, $ReplaceWith) if ($newName -eq $item.Name) { continue } if ($ItemType -eq "files") { $parentPath = $item.DirectoryName } else { $parentPath = $item.Parent.FullName } $newPath = Join-Path -Path $parentPath -ChildPath $newName Write-Host "" Write-Host "OLD: $($item.FullName)" -ForegroundColor DarkGray Write-Host "NEW: $newPath" -ForegroundColor Green if (Test-Path -LiteralPath $newPath) { Write-Host "SKIP: Target already exists." -ForegroundColor Yellow continue } Rename-Item -LiteralPath $item.FullName -NewName $newName Write-Host "RENAMED" -ForegroundColor Green } } switch ($Mode) { "files" { Rename-MatchingItems -ItemType "files" } "folders" { Rename-MatchingItems -ItemType "folders" } "both" { Rename-MatchingItems -ItemType "files" Rename-MatchingItems -ItemType "folders" } default { Write-Host "ERROR: Unknown mode: $Mode" -ForegroundColor Red Read-Host "Press Enter to exit" exit 1 } } Write-Host "" Write-Host "Renaming complete." -ForegroundColor Green Read-Host "Press Enter to exit" `; } [ parentFolderInput, renameModeInput, searchSubfoldersInput, findTextInput, replaceTextInput ].forEach((input) => input.addEventListener("input", updateScript)); copyButton.onclick = () => { navigator.clipboard.writeText(outputBox.value); copyButton.textContent = "✅ Copied!"; setTimeout(() => { copyButton.textContent = "📋 Copy PowerShell Script"; }, 1500); }; updateScript(); return html`

⬇ PowerShell rename files/folders script

${parentFolderInput} ${renameModeInput} ${searchSubfoldersInput} ${findTextInput} ${replaceTextInput}

Copy the generated script, paste it into PowerShell, and press Enter.

${copyButton}

${outputBox}

`; }4e:T89bc,{ const wrap = document.createElement("div"); wrap.style.maxWidth = "980px"; wrap.style.fontFamily = "system-ui, -apple-system, Segoe UI, sans-serif"; wrap.innerHTML = `

Nanopore phage assembly — WSL script generator

Windows and WSL paths are accepted. Windows drive paths are automatically converted to WSL paths. The generated pipeline deduplicates FASTQ read IDs before Filtlong and uses the deduplicated full read set for Medaka.

1. Input

Input mode

Parent folder — Windows or WSL path

No path entered

Folder behavior

Separate-file mode is recommended when each phage already has one FASTQ. Combine mode remains available for older split/chunked datasets.

Output parent folder — Windows or WSL path

No path entered

Overwrite completed sample outputs and rerun them

Existing sample folders without a final Medaka FASTA are treated as incomplete and automatically removed/restarted.

2. Conda environments

Filtlong + Flye environment

Medaka environment

3. Read preparation + Filtlong

Before Filtlong, the script keeps the first occurrence of each Nanopore read ID and removes later duplicates. A deduplication report is saved for every sample.

Minimum read length (bp)

Minimum mean read Q (Phred)

Target bases (Mb) Use --target_bases

Keep best reads (%) Use --keep_percent

4. Flye

Nanopore read type

Estimated genome size

Flye polishing iterations

Flye threads

Minimum overlap (bp)

0 = Flye chooses automatically.

Assembly coverage

0 = disabled.

Use --meta Use --scaffold

5. Medaka

Medaka threads

Medaka model

Example: r1041_e82_400bps_sup_v5.2.0. Blank omits -m.

Flye uses the Filtlong-filtered subset. Medaka uses all deduplicated reads before Filtlong subsampling.

Download .sh

`; const $ = (id) => wrap.querySelector(`#${id}`); let downloadUrl = null; const shQuote = (value) => "'" + String(value).replace(/'/g, "'\"'\"'") + "'"; const boolWord = (value) => (value ? "true" : "false"); function toWslPath(value) { let p = String(value ?? "").trim(); if (!p) { return ""; } if ( (p.startsWith('"') && p.endsWith('"')) || (p.startsWith("'") && p.endsWith("'")) ) { p = p.slice(1, -1).trim(); } if (/^file:\/\//i.test(p)) { try { const u = new URL(p); p = decodeURIComponent(u.pathname); if (/^\/[A-Za-z]:\//.test(p)) { p = p.slice(1); } } catch (err) { return p; } } p = p.replace(/^\\\\\?\\/, ""); const driveMatch = p.match(/^([A-Za-z]):[\\/](.*)$/); if (driveMatch) { const drive = driveMatch[1].toLowerCase(); const rest = driveMatch[2].replace(/\\/g, "/").replace(/^\/+/, ""); return `/mnt/${drive}/${rest}`; } if (/^\/mnt\/[A-Za-z](\/|$)/.test(p)) { p = p.replace( /^\/mnt\/([A-Za-z])/, (_, drive) => `/mnt/${drive.toLowerCase()}` ); } return p; } function makeScript() { const inputMode = $("inputMode").value; const folderBehavior = $("folderBehavior").value; const parentFolder = toWslPath($("parentFolder").value); const singleFastq = toWslPath($("singleFastq").value); const outputFolder = toWslPath($("outputFolder").value); const coreEnv = $("coreEnv").value.trim(); const medakaEnv = $("medakaEnv").value.trim(); const overwrite = $("overwrite").checked; // -------------------------------------------------------- // Filtlong // -------------------------------------------------------- const minLength = Math.max(0, Number($("minLength").value) || 0); const minQ = Math.max(0, Number($("minQ").value) || 0); const filtIdentity = 100 * (1 - Math.pow(10, -minQ / 10)); const targetMb = Math.max(0, Number($("targetMb").value) || 0); const targetBases = Math.round(targetMb * 1000000); const keepPercent = Math.min( 100, Math.max(0.1, Number($("keepPercent").value) || 100) ); const useTarget = $("useTarget").checked && targetBases > 0; const useKeep = $("useKeep").checked; // -------------------------------------------------------- // Flye // -------------------------------------------------------- const flyeReadType = $("flyeReadType").value; const genomeSize = $("genomeSize").value.trim(); const flyeIterations = Math.max( 0, Math.floor(Number($("flyeIterations").value) || 0) ); const flyeThreads = Math.max( 1, Math.floor(Number($("flyeThreads").value) || 1) ); const minOverlap = Math.max( 0, Math.floor(Number($("minOverlap").value) || 0) ); const asmCoverage = Math.max( 0, Math.floor(Number($("asmCoverage").value) || 0) ); const flyeMeta = $("flyeMeta").checked; const flyeScaffold = $("flyeScaffold").checked; // -------------------------------------------------------- // Medaka // -------------------------------------------------------- const medakaThreads = Math.max( 1, Math.floor(Number($("medakaThreads").value) || 1) ); const medakaModel = $("medakaModel").value.trim(); // -------------------------------------------------------- // Build arguments // -------------------------------------------------------- const filtArgs = []; if (minLength > 0) { filtArgs.push(`--min_length ${Math.round(minLength)}`); } if (minQ > 0) { filtArgs.push(`--min_mean_q ${filtIdentity.toFixed(4)}`); } if (useTarget) { filtArgs.push(`--target_bases ${targetBases}`); } if (useKeep) { filtArgs.push(`--keep_percent ${keepPercent}`); } const flyeArgs = [ `${flyeReadType} "$filtered_reads"`, '--out-dir "$FLYE_DIR"', `--threads ${flyeThreads}`, `--iterations ${flyeIterations}` ]; if (genomeSize) { flyeArgs.push(`--genome-size ${shQuote(genomeSize)}`); } if (minOverlap > 0) { flyeArgs.push(`--min-overlap ${minOverlap}`); } if (asmCoverage > 0) { flyeArgs.push(`--asm-coverage ${asmCoverage}`); } if (flyeMeta) { flyeArgs.push("--meta"); } if (flyeScaffold) { flyeArgs.push("--scaffold"); } const medakaArgs = [ '-i "$unique_reads"', '-d "$FLYE_DIR/assembly.fasta"', '-o "$MEDAKA_DIR"', `-t ${medakaThreads}` ]; if (medakaModel) { medakaArgs.push(`-m ${shQuote(medakaModel)}`); } const inputPath = inputMode === "single" ? singleFastq : parentFolder; // ======================================================== // Generated Bash // ======================================================== const L = [ "#!/usr/bin/env bash", "", "# Nanopore bacteriophage assembly pipeline", "# Original FASTQ -> deduplicate -> Filtlong -> Flye -> Medaka", "# Flye uses filtered reads; Medaka uses the full deduplicated read set.", "", "(", "set -Eeuo pipefail", "trap 'status=$?; echo; echo \"❌ Pipeline failed at line $LINENO (exit $status)\" >&2; exit $status' ERR", "", `INPUT_MODE=${shQuote(inputMode)}`, `FOLDER_BEHAVIOR=${shQuote(folderBehavior)}`, `INPUT_PATH=${shQuote(inputPath)}`, `OUT_ROOT=${shQuote(outputFolder)}`, `CORE_ENV=${shQuote(coreEnv)}`, `MEDAKA_ENV=${shQuote(medakaEnv)}`, `OVERWRITE=${boolWord(overwrite)}`, "", 'export PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:$PATH"', "", "echo", 'echo "=============================================================="', 'echo "Nanopore phage assembly pipeline"', 'echo "=============================================================="', 'echo "Input mode: $INPUT_MODE"', 'echo "Input path: $INPUT_PATH"', 'echo "Output root: $OUT_ROOT"', "", '[[ -n "$INPUT_PATH" ]] || { echo "❌ No input path supplied." >&2; exit 1; }', '[[ -n "$OUT_ROOT" ]] || { echo "❌ No output path supplied." >&2; exit 1; }', "", 'if [[ "$INPUT_MODE" == "single" ]]; then', ' [[ -f "$INPUT_PATH" ]] || { echo "❌ FASTQ file does not exist: $INPUT_PATH" >&2; exit 1; }', "else", ' [[ -d "$INPUT_PATH" ]] || { echo "❌ Input folder does not exist: $INPUT_PATH" >&2; exit 1; }', "fi", "", 'mkdir -p "$OUT_ROOT"', '[[ -w "$OUT_ROOT" ]] || { echo "❌ Output folder is not writable: $OUT_ROOT" >&2; exit 1; }', "", "# Locate a Conda installation containing both requested environments.", 'CONDA_SH=""', 'CONDA_CANDIDATES=("$HOME/miniconda/etc/profile.d/conda.sh" "$HOME/miniconda3/etc/profile.d/conda.sh" "$HOME/anaconda3/etc/profile.d/conda.sh" "$HOME/miniforge3/etc/profile.d/conda.sh")', "", 'for candidate in "${CONDA_CANDIDATES[@]}"; do', ' [[ -f "$candidate" ]] || continue', ' conda_base="${candidate%/etc/profile.d/conda.sh}"', ' if [[ -d "$conda_base/envs/$CORE_ENV" && -d "$conda_base/envs/$MEDAKA_ENV" ]]; then', ' CONDA_SH="$candidate"', " break", " fi", "done", "", '[[ -n "$CONDA_SH" ]] || { echo "❌ Could not find one Conda installation containing $CORE_ENV and $MEDAKA_ENV." >&2; exit 1; }', "", '. "$CONDA_SH"', 'echo "Using Conda: $CONDA_SH"', "", 'conda activate "$CORE_ENV"', 'command -v python3 >/dev/null || { echo "❌ python3 not found in $CORE_ENV" >&2; exit 1; }', 'command -v filtlong >/dev/null || { echo "❌ filtlong not found in $CORE_ENV" >&2; exit 1; }', 'command -v flye >/dev/null || { echo "❌ flye not found in $CORE_ENV" >&2; exit 1; }', 'command -v gzip >/dev/null || { echo "❌ gzip not found" >&2; exit 1; }', "", 'echo "Python: $(command -v python3)"', 'echo "Filtlong: $(command -v filtlong)"', 'echo "Flye: $(command -v flye)"', "", 'conda activate "$MEDAKA_ENV"', 'command -v medaka_consensus >/dev/null || { echo "❌ medaka_consensus not found in $MEDAKA_ENV" >&2; exit 1; }', 'echo "Medaka: $(command -v medaka_consensus)"', 'echo "✅ Preflight checks passed."', "", // ------------------------------------------------------ // Filename helpers // ------------------------------------------------------ "strip_fastq_ext() {", ' local x="$1"', ' case "$x" in', ' *.fastq.gz) x="${x%.fastq.gz}" ;;', ' *.fq.gz) x="${x%.fq.gz}" ;;', ' *.fastq) x="${x%.fastq}" ;;', ' *.fq) x="${x%.fq}" ;;', ' *.FASTQ.GZ) x="${x%.FASTQ.GZ}" ;;', ' *.FQ.GZ) x="${x%.FQ.GZ}" ;;', ' *.FASTQ) x="${x%.FASTQ}" ;;', ' *.FQ) x="${x%.FQ}" ;;', " esac", ' printf "%s" "$x"', "}", "", "sanitize_sample() {", ' local x="$1"', ' x="${x//\\//__}"', ' x="$(printf "%s" "$x" | tr "[:space:]" "_")"', ' x="$(printf "%s" "$x" | LC_ALL=C tr -c "A-Za-z0-9._-" "_")"', ' while [[ "$x" == _* ]]; do x="${x#_}"; done', ' while [[ "$x" == *_ ]]; do x="${x%_}"; done', ' printf "%s" "$x"', "}", "", // ------------------------------------------------------ // Deduplicate FASTQ // ------------------------------------------------------ "deduplicate_fastq() {", ' local input="$1"', ' local output="$2"', ' local report="$3"', "", ' python3 - "$input" "$output" "$report" <<\'PY\'', "import sys", "import gzip", "", "input_path, output_path, report_path = sys.argv[1:4]", "", "def open_input(path):", ' if path.lower().endswith(".gz"):', ' return gzip.open(path, "rt")', ' return open(path, "rt")', "", "seen = set()", "duplicate_ids = []", "total = 0", "duplicates_removed = 0", "", 'with open_input(input_path) as fin, gzip.open(output_path, "wt") as fout:', " while True:", " header = fin.readline()", " if not header:", " break", "", " sequence = fin.readline()", " plus = fin.readline()", " quality = fin.readline()", "", " if not sequence or not plus or not quality:", ' raise RuntimeError("Incomplete FASTQ record at end of file")', "", ' if not header.startswith("@"):', ' raise RuntimeError(f"Invalid FASTQ header: {header.rstrip()}")', "", ' if not plus.startswith("+"):', ' raise RuntimeError(f"Invalid FASTQ plus line for {header.rstrip()}")', "", " read_id = header[1:].split()[0]", " total += 1", "", " if read_id in seen:", " duplicates_removed += 1", " if len(duplicate_ids) < 25:", " duplicate_ids.append(read_id)", " continue", "", " seen.add(read_id)", " fout.write(header)", " fout.write(sequence)", " fout.write(plus)", " fout.write(quality)", "", "unique = len(seen)", "", 'with open(report_path, "wt") as report:', ' report.write(f"Input FASTQ: {input_path}\\n")', ' report.write(f"Total records: {total}\\n")', ' report.write(f"Unique read IDs: {unique}\\n")', ' report.write(f"Duplicate records removed: {duplicates_removed}\\n")', "", " if duplicate_ids:", ' report.write("\\nFirst duplicate read IDs encountered:\\n")', " for read_id in duplicate_ids:", ' report.write(read_id + "\\n")', "", 'print(f"FASTQ records: {total}")', 'print(f"Unique read IDs: {unique}")', 'print(f"Duplicates removed: {duplicates_removed}")', "PY", "}", "", // ------------------------------------------------------ // Combine FASTQs for optional legacy group mode // ------------------------------------------------------ "combine_fastqs() {", ' local output="$1"', " shift", "", ' (( $# > 0 )) || { echo "❌ No FASTQ files supplied for concatenation." >&2; return 1; }', "", " {", " local f", ' for f in "$@"; do', ' case "$f" in', ' *.gz|*.GZ) gzip -cd -- "$f" ;;', ' *) cat -- "$f" ;;', " esac", " done", ' } | gzip -c > "$output"', "}", "", // ------------------------------------------------------ // FASTQ discovery // ------------------------------------------------------ "find_fastqs_recursive() {", " local f", "", ' while IFS= read -r -d "" f; do', ' case "${f,,}" in', " *.fastq|*.fastq.gz|*.fq|*.fq.gz)", ' printf "%s\\0" "$f"', " ;;", " esac", " done < <(", ' find "$INPUT_PATH" -type f ! -path "$OUT_ROOT/*" -print0', " )", "}", "", "find_fastqs_in_dir() {", ' local d="$1"', " local f", "", ' while IFS= read -r -d "" f; do', ' case "${f,,}" in', " *.fastq|*.fastq.gz|*.fq|*.fq.gz)", ' printf "%s\\0" "$f"', " ;;", " esac", " done < <(", ' find "$d" -maxdepth 1 -type f ! -path "$OUT_ROOT/*" -print0', " )", "}", "", // ------------------------------------------------------ // Process one sample // ------------------------------------------------------ "process_sample() {", ' local sample="$1"', ' local original_reads="$2"', "", ' local SAMPLE_DIR="$OUT_ROOT/$sample"', ' local INPUT_DIR="$SAMPLE_DIR/input"', ' local FILTER_DIR="$SAMPLE_DIR/filtered"', ' local FLYE_DIR="$SAMPLE_DIR/flye"', ' local MEDAKA_DIR="$SAMPLE_DIR/medaka"', "", ' local unique_reads="$INPUT_DIR/${sample}.unique.fastq.gz"', ' local dedup_report="$INPUT_DIR/${sample}.deduplication.txt"', ' local filtered_reads="$FILTER_DIR/${sample}.filtered.fastq.gz"', ' local final_result="$SAMPLE_DIR/${sample}.medaka.fasta"', "", " echo", ' echo "=============================================================="', ' echo "Sample: $sample"', ' echo "Original reads: $original_reads"', ' echo "Output: $SAMPLE_DIR"', ' echo "=============================================================="', "", ' if [[ -d "$SAMPLE_DIR" ]]; then', ' if [[ -s "$final_result" && "$OVERWRITE" != "true" ]]; then', ' echo "✅ Completed Medaka output already exists; skipping sample."', ' echo " $final_result"', " return 0", " fi", "", ' if [[ "$OVERWRITE" == "true" ]]; then', ' echo "Removing existing sample output and rerunning."', " else", ' echo "⚠️ Existing sample directory is incomplete; removing it and restarting."', " fi", "", ' rm -rf -- "$SAMPLE_DIR"', " fi", "", ' mkdir -p "$INPUT_DIR" "$FILTER_DIR" "$FLYE_DIR" "$MEDAKA_DIR"', "", // Deduplicate ' conda activate "$CORE_ENV"', ' echo "[0/3] Deduplicating FASTQ read IDs"', ' deduplicate_fastq "$original_reads" "$unique_reads" "$dedup_report"', "", ' [[ -s "$unique_reads" ]] || { echo "❌ Deduplication produced an empty FASTQ." >&2; return 1; }', ' echo "Deduplicated reads: $unique_reads"', ' echo "Deduplication report: $dedup_report"', "", // Filtlong ' echo "[1/3] Filtlong"', "", " filtlong \\", ...filtArgs.map((arg) => ` ${arg} \\`), ' "$unique_reads" \\', ' 2> >(tee "$SAMPLE_DIR/filtlong.log" >&2) \\', ' | gzip -c > "$filtered_reads"', "", ' [[ -s "$filtered_reads" ]] || { echo "❌ Filtlong produced an empty FASTQ. See $SAMPLE_DIR/filtlong.log" >&2; return 1; }', "", // Flye ' echo "[2/3] Flye"', "", " flye \\", ...flyeArgs.map( (arg, i) => ` ${arg}${i === flyeArgs.length - 1 ? "" : " \\"}` ), "", ' [[ -s "$FLYE_DIR/assembly.fasta" ]] || { echo "❌ Flye did not produce assembly.fasta." >&2; return 1; }', ' cp -- "$FLYE_DIR/assembly.fasta" "$SAMPLE_DIR/${sample}.flye.fasta"', "", // Medaka ' conda activate "$MEDAKA_ENV"', ' echo "[3/3] Medaka using all deduplicated reads"', "", " medaka_consensus \\", ...medakaArgs.map( (arg, i) => ` ${arg}${i === medakaArgs.length - 1 ? "" : " \\"}` ), "", ' [[ -s "$MEDAKA_DIR/consensus.fasta" ]] || { echo "❌ Medaka did not produce consensus.fasta." >&2; return 1; }', ' cp -- "$MEDAKA_DIR/consensus.fasta" "$final_result"', "", ' echo "✅ Finished $sample"', ' echo "Flye: $SAMPLE_DIR/${sample}.flye.fasta"', ' echo "Medaka: $final_result"', "}", "", // ------------------------------------------------------ // Input mode // ------------------------------------------------------ 'if [[ "$INPUT_MODE" == "single" ]]; then', ' base="$(basename -- "$INPUT_PATH")"', ' sample="$(sanitize_sample "$(strip_fastq_ext "$base")")"', ' [[ -n "$sample" ]] || sample="sample"', ' process_sample "$sample" "$INPUT_PATH"', "", "else", "", ' INPUT_PATH="${INPUT_PATH%/}"', ' OUT_ROOT="${OUT_ROOT%/}"', "", ' if [[ "$FOLDER_BEHAVIOR" == "file" ]]; then', "", " found=0", "", ' while IFS= read -r -d "" f; do', " found=1", ' rel="${f#"$INPUT_PATH"/}"', ' sample="$(sanitize_sample "$(strip_fastq_ext "$rel")")"', ' [[ -n "$sample" ]] || sample="sample"', ' process_sample "$sample" "$f"', " done < <(find_fastqs_recursive | sort -z)", "", ' [[ "$found" -eq 1 ]] || { echo "❌ No FASTQ files found under $INPUT_PATH" >&2; exit 1; }', "", " else", "", " declare -A seen_dirs=()", "", ' while IFS= read -r -d "" f; do', ' seen_dirs["$(dirname -- "$f")"]=1', " done < <(find_fastqs_recursive)", "", ' (( ${#seen_dirs[@]} > 0 )) || { echo "❌ No FASTQ files found under $INPUT_PATH" >&2; exit 1; }', "", " mapfile -t dirs < <(", ' printf "%s\\n" "${!seen_dirs[@]}" | sort', " )", "", ' for d in "${dirs[@]}"; do', "", ' mapfile -d "" -t files < <(', ' find_fastqs_in_dir "$d" | sort -z', " )", "", " (( ${#files[@]} > 0 )) || continue", "", ' if [[ "$d" == "$INPUT_PATH" ]]; then', ' rel="$(basename -- "$INPUT_PATH")"', " else", ' rel="${d#"$INPUT_PATH"/}"', " fi", "", ' sample="$(sanitize_sample "$rel")"', ' [[ -n "$sample" ]] || sample="sample"', "", ' tmp_original="$(mktemp -p "$OUT_ROOT" ".${sample}.original.XXXXXX.fastq.gz")"', "", ' echo "Combining ${#files[@]} FASTQ file(s) for sample $sample"', ' combine_fastqs "$tmp_original" "${files[@]}"', ' process_sample "$sample" "$tmp_original"', ' rm -f -- "$tmp_original"', "", " done", "", " fi", "", "fi", "", "echo", 'echo "✅ All requested samples are complete."', 'echo "Output root: $OUT_ROOT"', "", ")", "", "PIPELINE_STATUS=$?", "", 'if [[ "$PIPELINE_STATUS" -ne 0 ]]; then', " echo", ' echo "Pipeline stopped with exit code $PIPELINE_STATUS."', ' echo "Your WSL shell remains open."', "fi", "" ]; return L.join("\n"); } function updateVisibility() { const folderMode = $("inputMode").value === "folder"; wrap .querySelectorAll(".folder-only") .forEach((el) => (el.style.display = folderMode ? "" : "none")); wrap .querySelectorAll(".single-only") .forEach((el) => (el.style.display = folderMode ? "none" : "")); } function updatePathPreviews() { const parent = $("parentFolder").value.trim(); const single = $("singleFastq").value.trim(); const output = $("outputFolder").value.trim(); $("parentFolderPreview").textContent = parent ? `WSL: ${toWslPath(parent)}` : "No path entered"; $("singleFastqPreview").textContent = single ? `WSL: ${toWslPath(single)}` : "No path entered"; $("outputFolderPreview").textContent = output ? `WSL: ${toWslPath(output)}` : "No path entered"; } function validateInputs() { const errors = []; const warnings = []; const inputMode = $("inputMode").value; const parent = $("parentFolder").value.trim(); const single = $("singleFastq").value.trim(); const output = $("outputFolder").value.trim(); const activeInput = inputMode === "single" ? single : parent; if (inputMode === "single" && !single) { errors.push("Enter a FASTQ file path."); } if (inputMode === "folder" && !parent) { errors.push("Enter a parent folder path."); } if (!output) { errors.push("Enter an output folder path."); } if (!$("coreEnv").value.trim()) { errors.push("Enter the Filtlong + Flye Conda environment."); } if (!$("medakaEnv").value.trim()) { errors.push("Enter the Medaka Conda environment."); } if (inputMode === "folder" && /\.(fastq|fq)(\.gz)?$/i.test(parent)) { errors.push( "The parent-folder path looks like a FASTQ file. Choose 'One FASTQ file' or remove the FASTQ filename from the path." ); } if ( inputMode === "single" && single && !/\.(fastq|fq)(\.gz)?$/i.test(single) ) { warnings.push( "The single-file path does not end in .fastq, .fastq.gz, .fq, or .fq.gz." ); } if (/^\\\\[^\\]/.test(activeInput)) { errors.push("UNC network paths cannot be converted automatically."); } if (/^\\\\[^\\]/.test(output)) { errors.push("The output path appears to be a UNC network path."); } if (Number($("minQ").value) >= 20) { warnings.push( "Q20 is stringent; confirm that enough phage reads remain after filtering." ); } if (Number($("medakaThreads").value) > 4) { warnings.push( "Medaka generally does not benefit much from a large thread count." ); } return { errors, warnings }; } function validateGeneratedScript(script) { const errors = []; if (!script.startsWith("#!/usr/bin/env bash\n")) { errors.push("Generated script does not start with a valid Bash shebang."); } if (script.split("\n").length < 100) { errors.push("Generated script appears to be missing real line breaks."); } if (/^(INPUT_PATH|OUT_ROOT)='[A-Za-z]:[\\/]/m.test(script)) { errors.push("A Windows path was not converted to WSL format."); } return errors; } function update() { updateVisibility(); updatePathPreviews(); const q = Math.max(0, Number($("minQ").value) || 0); const identity = 100 * (1 - Math.pow(10, -q / 10)); $( "qConversion" ).textContent = `Q${q} → Filtlong --min_mean_q ${identity.toFixed(4)}%`; const { errors: inputErrors, warnings } = validateInputs(); const script = inputErrors.length === 0 ? makeScript() : ""; const scriptErrors = script ? validateGeneratedScript(script) : []; const errors = [...inputErrors, ...scriptErrors]; const warningBox = $("warning"); warningBox.classList.toggle("phage-error", errors.length > 0); if (errors.length || warnings.length) { warningBox.style.display = "block"; warningBox.textContent = [ ...errors.map((x) => `❌ ${x}`), ...warnings.map((x) => `⚠️ ${x}`) ].join(" "); } else { warningBox.style.display = "none"; warningBox.textContent = ""; } $("code").textContent = errors.length ? "Complete/correct the required fields above to generate the WSL script." : script; if (downloadUrl) { URL.revokeObjectURL(downloadUrl); downloadUrl = null; } if (errors.length === 0) { downloadUrl = URL.createObjectURL( new Blob([script], { type: "text/x-shellscript" }) ); $("downloadBtn").href = downloadUrl; $("downloadBtn").style.pointerEvents = "auto"; $("downloadBtn").style.opacity = "1"; $("copyBtn").disabled = false; } else { $("downloadBtn").removeAttribute("href"); $("downloadBtn").style.pointerEvents = "none"; $("downloadBtn").style.opacity = "0.5"; $("copyBtn").disabled = true; } $("copyStatus").textContent = ""; } wrap.querySelectorAll("input, select").forEach((el) => { el.addEventListener("input", update); el.addEventListener("change", update); }); $("copyBtn").addEventListener("click", async () => { try { await navigator.clipboard.writeText($("code").textContent); $("copyStatus").textContent = "Copied."; } catch (err) { $("copyStatus").textContent = "Copy failed; select the script manually."; } }); update(); return wrap; }4f:[{"id":"ec49f2fe9b16cafc747859ead1bd81591e1d552306f1ff15088e514be2d41de7c014658f4378bb049766a5507da47e37a506fdcf033ec2e202735cc2efa1a885","href":"https://static.observableusercontent.com/files/ec49f2fe9b16cafc747859ead1bd81591e1d552306f1ff15088e514be2d41de7c014658f4378bb049766a5507da47e37a506fdcf033ec2e202735cc2efa1a885","name":"image@2.png","type":"image/png","lastModified":1751220008793.075,"size":6090}] 2d:{"id":"7dbec19870ae088c","slug":"nanopore-sequencing-scripts","title":"Nanopore Sequencing Scripts","thumbnail":"4a45afe2e2097537bf6aeae21ca4a10a7c587dad6f89f0bed023b60b3820b46c","access_level":"public","create_time":"$D2025-06-29T10:24:18.622Z","update_time":"$D2026-09-28T15:40:35.215Z","view_time":null,"first_public_time":"$D2025-06-30T15:51:42.453Z","has_unpublished_edits":null,"fork":false,"stars":0,"starred":false,"owner":{"type":"team","id":"d1dfc5f860fb8699","login":"phage","name":"WabushOnce","avatar_url":"https://avatars-next.observableusercontent.com/9303f26b98fd8f7e04b66a6ac3218bfd2ca7e146f15ba24ca32ffac15f802763","tier":"starter_2024"},"version":2104,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"3035ce62f81ebcd9","login":"phage","name":"WabushOnce","avatar_url":"https://avatars-next.observableusercontent.com/9303f26b98fd8f7e04b66a6ac3218bfd2ca7e146f15ba24ca32ffac15f802763","tier":null}],"members":[{"type":"individual","id":"3035ce62f81ebcd9","login":"phage","name":"WabushOnce","avatar_url":"https://avatars-next.observableusercontent.com/9303f26b98fd8f7e04b66a6ac3218bfd2ca7e146f15ba24ca32ffac15f802763","tier":null,"role":"editor"},{"type":"team","id":"d1dfc5f860fb8699","login":"phage","name":"WabushOnce","avatar_url":"https://avatars-next.observableusercontent.com/9303f26b98fd8f7e04b66a6ac3218bfd2ca7e146f15ba24ca32ffac15f802763","tier":"starter_2024","role":"editor"}],"comments":[],"custom_thumbnail":true,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"title":"Nanopore Sequencing Scripts","cells":[{"id":0,"value":"$2f","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1366,"value":"DoradoVersion = md`## 📦 Dorado - Check for the latest version\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1336,"value":"$30","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1314,"value":"$31","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":213,"value":"Dorado = md`\n---\n## 📦 Dorado - Basecalling via WSL\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":153,"value":"dorado_kitInfo = html`\n

\n 🔍 Need to verify the kit name? \n \n Check the official Dorado documentation ↗\n \n

\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":958,"value":"dorado_modelInfo = html`\n

\n 🔍 Need to verify the model name? \n \n Check the official Dorado documentation ↗\n \n

\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2020,"value":"$32","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2082,"value":"$33","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1431,"value":"Demultiplexing = md`\n--- \n## 📦 Demultiplexing: Split Barcodes \n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":924,"value":"$34","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1442,"value":"Merge_Barcodes = md`\n---\n## 📦 Merge Multiple Barcode FastQ Files \n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":64,"value":"- Merge multiple FASTQ files into one FASTQ file.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1638,"value":"$35","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1460,"value":"Remove_Host = md`\n---\n## 📦 Remove Bacterial Reads\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1013,"value":"\n- Use Minimap2 to remove bacterial host reads from fastq files","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1011,"value":"$36","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1655,"value":"NanoComp = md`\n---\n## 📦 Create NanoComp Files\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1662,"value":"- NanoComp compares multiple runs of long read sequencing data and alignments\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1651,"value":"$37","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1465,"value":"NanoPlot = md`\n---\n## 📦 Create Nanoplot Files\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":377,"value":"- For each genome, NanoPlot summarizes read data such as: number of bases and reads, read length, quality, N50, and more.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":375,"value":"$38","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":381,"value":"- Create a spreadsheet (Excel) to summarize all the Nanoplot files in a folder","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":453,"value":"$39","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1469,"value":"Filter_Filtlong = md`\n---\n## 📦 Filter Nanopore Reads - Using Filtlong or Chopper\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":654,"value":"\nFor a 50 kb genome, empirical non-experimental experience suggests 6-10 megabases of sequencing data for assembly of circularized contig.\n- Currently, for picked-plaque DNA treated with T4 ligase on phage genomes containing cohesive overhangs, 4-5 megabases of sequencing data has been sufficient to generate complete assemblies.\n- In one case, assembling with 30 megabases of data produced NO contigs. However, filtering the dataset to retain the top 10 megabases of reads, resulted in a single circularized contig.\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1866,"value":"- Filtlong filters reads based on length and quality.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":652,"value":"$3a","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":480,"value":"- Chopper filters reads by length, Q-score, and/or total bases.\n ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":483,"value":"$3b","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1473,"value":"Flye = md`\n---\n## 📦 Flye - Assemble Reads\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1810,"value":"$3c","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1815,"value":"$3d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":504,"value":"- Use Flye to assemble a folder of fastq files ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":582,"value":"$3e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2015,"value":"$3f","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":508,"value":"\n- Create a spreadsheet of each barcode's assembly info (contig #, size, coverage, circularization, ...) ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":2012,"value":"$40","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1510,"value":"Create_BAM = md`\n---\n## 📦 View Read Coverage\n- Create .BAM Files\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2008,"value":"$41","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1668,"value":"$42","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1875,"value":"- When aligning reads to a PhagesDB FASTA to create a BAM file, you may need to use the reverse-complemented FASTA if the reads do not align due to the reference being in the opposite orientation ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1872,"value":"$43","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1678,"value":"- Use QUAST to find SNPs and indels between two FastA files.\n- QUAST Install Script","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1680,"value":"$44","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1882,"value":"$45","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1001,"value":"---\n## Run Medaka to Polish Contigs\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":998,"value":"$46","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1900,"value":"PharokkaAnnotation = md`\n---\n## 📦 Annotate Phages (Pharokka, Phold, and Phynteny)\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2031,"value":"- Pharokka: run on a phage .FASTA - predicts genes and functions\n- Phold: run on the Pharokka generated .GBK file - creates ProsT5 3Di tokens run in Foldseek to predict protein structures\n- Phynteny: (optional) run on the Phold .GBK file - predicts functional categories for hypothetical proteins\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":2047,"value":"### Program Installers","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":2000,"value":"$47","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2037,"value":"$48","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2040,"value":"$49","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2049,"value":"### Run Programs","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":1161,"value":"$4a","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2052,"value":"$4b","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2060,"value":"$4c","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1960,"value":"BatchRename = md`\n---\n## 📦 Batch Rename Folders and Files\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1956,"value":"$4d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":9,"value":"---\n# Code\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":11,"value":"import { Inputs } from \"@observablehq/inputs\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":13,"value":"import { html } from \"@observablehq/htl\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2081,"value":"$4e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W4f","stdlib":"1","titling":"auto"}} 2e:"$2d:body"