Published
Edited
Apr 2, 2022
plot_tox21_data_for_all_compounds = function(array) {
var i;
var vl_array = [];
for (i = 0; i < array.length; i++) {
var data = array[i]["Results"];
var compound_name = array[i]["Compound"];
vl_array.push(
vl.markRect().data(data)
.encode(
vl.x({"grid": true}).fieldN("Dataset name"),
vl.color().fieldN("Hit call")
.legend({"orient": "left"})
.scale({
"type": "ordinal",
"domain": ["active", "inactive", "ambiguous"],
"range": ["#ff6f69", "#96ceb4", "#ffcc5c"]}),
vl.tooltip([
vl.fieldN("Dataset name"),
{
"field": "IC50",
"type": "quantitative",
"format": ",.2f"
},
vl.fieldN("Organism"),
vl.fieldN("Biological process")]),
).title({"text": compound_name + " (Tanimoto similarity: " + tox21_hits.filter(function(item) {return item["Substance name"] == compound_name})[0].SMILES.similarity + ")", "anchor": "start"})
);
}
return vl.vconcat(vl_array).render();
}
plot_drugmatrix_tggates_data_for_all_compounds = function(array, assay_hits) {
if (array.length < 1) {
return "No data to display";
} else {
var i;
var vl_array = [];
for (i = 0; i < array.length; i++) {
var data = array[i]["Actual datasets with sample names"];
var compound_name = array[i]["Compound"];
vl_array.push(
vl.markRect().data(data)
.encode(
vl.y().fieldN("Sample"),
vl.x().fieldN("SYMBOL"),
vl.color().fieldQ("logFC").legend({"orient": "left"}).scale({scheme: "lightmulti", reverse: false, zero: false}),
vl.tooltip([
vl.fieldN("Sample"),
{
"field": "logFC",
"type": "quantitative",
"format": ",.2f"
},
vl.fieldN("ENSEMBL"),
vl.fieldN("ENTREZID")]),
).title({"text": compound_name + " (Tanimoto similarity: " + assay_hits.filter(function(item) {return item["Compound"] == compound_name})[0].SMILES.similarity + ")", "anchor": "start"})
);
}
return vl.vconcat(vl_array).render();
}
}
reshaped_tox21_datasets_for_all_compounds = reshape_tox21_datasets_for_all_compounds(filtered_tox21_hits, selected_tox21_compounds)
tox21_similar_compounds= Array.from(new Set(filtered_tox21_hits.map(item => item["Substance name"])))
reshape_tox21_datasets_for_all_compounds = function(array_of_datasets, array_of_compounds) {
var i;
var arr = [];
var j;
var hitcall;
var ic50;
for (i = 0; i < array_of_compounds.length; i++) {
const compound = array_of_compounds[i];
var compound_datasets = array_of_datasets.filter(item => item["Substance name"] == compound);
var unique_compound_datasets = Array.from(new Set(compound_datasets.map(item => item["Summary data"])));
var results = [];
// loop over unique compound datasets and evaluate
for (j = 0; j < unique_compound_datasets.length; j++) {
const dataset_id = unique_compound_datasets[j];
const datasets = compound_datasets.filter(item => item["Summary data"] == dataset_id);
const hitcalls = datasets.map(item => item["Hit call"]);
const unique_hitcalls = Array.from(new Set(hitcalls));
const ic50_values = datasets.map(item => item["IC50"]);
if (unique_hitcalls.length > 1) {
hitcall = "ambiguous";
} else {
hitcall = unique_hitcalls[0];
};
if (hitcall == "active") {
const arrSum = ic50_values => ic50_values.reduce((a,b) => a + b, 0);
ic50 = arrSum / (ic50_values.length);
} else {
ic50 = null;
}
const dataset_object = datasets[0];
if (colors.includes(hitcall)) {
results.push({
"Hit call": hitcall,
"Organism": dataset_object["Organism"],
"Cell type": dataset_object["Cell type"],
"Design type": dataset_object["Design Type"],
"Target type": dataset_object["Target type"],
"Target family": dataset_object["Target family"],
"Biological process": dataset_object["Biolological process"],
"Dataset name": dataset_object["Dataset name"].replace("EPA-InVitroDBV3.2-", "").replace("_", " "),
"IC50": ic50,
});
}
};
var compound_object = {
"Compound": compound,
"Results": results,
};
arr.push(compound_object);
}
return arr;
}
filtered_tox21_hits = tox21_hits
.filter(item => selected_cell_types.includes(item["Cell type"]))
.filter(item => selected_target_types.includes(item["Target type"]))
.filter(item => selected_target_families.includes(item["Target family"]))
target_families = Array.from(new Set(tox21_hits.map(item => item["Target family"])))
target_types = Array.from(new Set(tox21_hits.map(item => item["Target type"])))
selected_cell_types = selected_cell_types_with_descriptions.map(item => item.split(" ")[0])
cell_types_with_descriptions = [
"C3H10T1/2 (Mouse mesenchymal cell line)",
"CHO-K1 (Chinese Hamster Ovary cell line)",
"DT40 (Chicken bursal lymphoma cell line)",
"ERR-HEK293T (Human embryonic kidney 293 cell line)",
"GH3 (Rat pituitary tumor cell line)",
"HCT-116 (Human colon carcinoma cell line)",
"HEK293 (Human embryonic kidney 293 cell line)",
"HEK293T (Human embryonic kidney 293 cell line)",
"HeLa (Human epithelial cell line)",
"HepG2 (Human hepatocellular carcinoma cell line)",
"MCF-7 (Human breast cancer cell line)",
"MDA-kb2 (Human breast cancer cell line)",
"ME-180 (Human epithelial cell line)",
"NIH/3T3 (Mouse embryo fibroblast cell line)",
"PGC/ERR HEK293T (Human embryonic kidney 293 cell line)",
"PR-UAS-bla-HEK293T (Human embryonic kidney 293 cell line)",
"VM7 (Human breast cancer cell line)",
]
cell_types = cell_types_with_null.filter(item => item != null).sort()
cell_types_with_null = Array.from(new Set(tox21_hits.map(item => item["Cell type"])))
reshaped_tggates_datasets_for_all_compounds = reshape_datasets_for_all_compounds(tggates_datasets_for_all_compounds, selected_tggates_compounds)
reshaped_drugmatrix_datasets_for_all_compounds = reshape_datasets_for_all_compounds(drugmatrix_datasets_for_all_compounds, selected_drugmatrix_compounds)
reshape_datasets_for_all_compounds = function(array_of_compound_datasets, array_of_compounds) {
var i;
var arr = [];
for (i = 0; i < array_of_compound_datasets.length; i++) {
var compound_object = array_of_compound_datasets[i];
var compound = compound_object.Compound;
if (array_of_compounds.includes(compound)) {
var compound_samplenames = compound_object["Sample names"];
var compound_datasets = compound_object["Actual datasets"];
var compound_datasets_with_samplenames = [];
// loop over datasets for this compound
var j;
for (j = 0; j < compound_datasets.length; j++) {
var dataset = compound_datasets[j];
var dataset_with_samplename = dataset.map(item => {
item["Sample"] = compound_samplenames[j];
return item
})
compound_datasets_with_samplenames.push(dataset_with_samplename);
}
compound_object["Actual datasets with sample names"] = [].concat(...compound_datasets_with_samplenames);
arr.push(compound_object);
}
}
return arr;
}
tggates_datasets_for_all_compounds = get_promised_fold_changes_datasets_for_all_compounds(reshaped_tggates_hits)
drugmatrix_datasets_for_all_compounds = get_promised_fold_changes_datasets_for_all_compounds(reshaped_drugmatrix_hits)
get_datasets_from_promises = function(array_of_promises) {
return Promise.all(array_of_promises).then(item => item.map(dataset => dataset.results.map(row => row.data)))
}
get_promised_fold_changes_datasets_for_all_compounds = async function(array) {
var i;
var new_array = [];
for (i = 0; i < array.length; i++) {
var object = array[i];
const promised_datasets = get_promised_fold_changes_datasets(object["Fold changes datasets"]);
object["Promised datasets"] = promised_datasets;
object["Actual datasets"] = await get_datasets_from_promises(promised_datasets);
new_array.push(object);
}
return new_array;
}
get_promised_fold_changes_datasets = function(array_of_datasets) {
return array_of_datasets.map(dataset => {
return fetch(base_url_edelweiss + dataset + "/versions/latest/data", query_options_for_fold_changes).then(response_dataset => {
if (!response_dataset.ok) throw new Error(response_dataset.status);
return response_dataset.json();
})
})
}
query_options_for_fold_changes = {
return {
method: 'POST',
body: JSON.stringify({
condition: {
or: [
{lt: [{column: ["logFC"]}, -logfc_threshold]},
{gt: [{column: ["logFC"]}, logfc_threshold]},
]
},
columns: ["logFC", "SYMBOL", "ENSEMBL", "ENTREZID"]
})
}
}
reshaped_tggates_hits = reshape_tggates_hits(tggates_hits, tggates_unique_compounds)
reshape_tggates_hits = function(array_of_datasets, compounds) {
var arr = [];
compounds.forEach(compound => {
var compound_datasets = array_of_datasets.filter(object => object["Compound"] == compound);
var unique_compound_datasets = function_get_unique_fold_changes_datasets(compound_datasets);
var i;
var sample_names = [];
for (i = 0; i < unique_compound_datasets.length; i++) {
var dataset_id = unique_compound_datasets[i];
var obj = array_of_datasets.filter(item => item["Dataset ID"] == dataset_id)[0];
var sample_name = obj["Organism"] + " / " + obj["Study type"] + " / " + obj["Organ"] + " / " + obj["Dosing"] + " / " + obj["Dose"] + " / " + obj["Duration"] + " " + obj["Duration unit"];
sample_names.push(sample_name);
}
arr.push({
"Compound": compound,
"Fold changes datasets": unique_compound_datasets,
"Sample names": sample_names
})
});
return arr;
}
reshaped_drugmatrix_hits = reshape_drugmatrix_hits(drugmatrix_hits, drugmatrix_unique_compounds)
reshape_drugmatrix_hits = function(array_of_datasets, compounds) {
var arr = [];
compounds.forEach(compound => {
var compound_datasets = array_of_datasets.filter(object => object["Compound"] == compound);
var unique_compound_datasets = function_get_unique_fold_changes_datasets(compound_datasets);
var i;
var sample_names = [];
for (i = 0; i < unique_compound_datasets.length; i++) {
var dataset_id = unique_compound_datasets[i];
var obj = array_of_datasets.filter(item => item["Dataset ID"] == dataset_id)[0];
var sample_name = obj["Strain"] + " / " + obj["Route"] + " / " + obj["Organ or cell type"] + " / " + obj["Dose"] + " " + obj["Dose unit"] + " / " + obj["Duration"] + " " + obj["Duration unit"];
sample_names.push(sample_name);
}
arr.push({
"Compound": compound,
"Fold changes datasets": unique_compound_datasets,
"Sample names": sample_names
})
});
return arr;
}
tox21_unique_compounds_similarities = tox21_unique_compounds.map(compound => {
return tox21_hits.filter(item => item["Substance name"] == compound).map(item => item.SMILES.similarity)[0]
})
tox21_unique_svg = Promise.all(tox21_unique_smiles.map(smiles => function_get_svg_from_smiles(smiles)))
tox21_unique_smiles = tox21_unique_compounds.map(compound => {
return tox21_hits.filter(item => item["Substance name"] == compound).map(item => item.SMILES.canonical)[0]
})
tox21_unique_compounds = get_unique_compounds(tox21_hits, "Substance name")
tggates_unique_compounds_similarities = tggates_unique_compounds.map(compound => {
return tggates_hits.filter(item => item.Compound == compound).map(item => item.SMILES.similarity)[0]
})
tggates_unique_svg = Promise.all(tggates_unique_smiles.map(smiles => function_get_svg_from_smiles(smiles)))
tggates_unique_smiles = tggates_unique_compounds.map(compound => {
return tggates_hits.filter(item => item.Compound == compound).map(item => item.SMILES.canonical)[0]
})
tggates_unique_compounds = get_unique_compounds(tggates_hits, "Compound")
drugmatrix_unique_svg = Promise.all(drugmatrix_unique_smiles.map(smiles => function_get_svg_from_smiles(smiles)))
drugmatrix_unique_smiles = drugmatrix_unique_compounds.map(compound => {
return drugmatrix_hits.filter(item => item.Compound == compound).map(item => item.SMILES.canonical)[0]
})
drugmatrix_unique_compounds_similarities = drugmatrix_unique_compounds.map(compound => {
return drugmatrix_hits.filter(item => item.Compound == compound).map(item => item.SMILES.similarity)[0]
})
drugmatrix_hits.filter(item => item.Compound == drugmatrix_unique_compounds[0]).map(item => item.SMILES.canonical)[0]
drugmatrix_unique_compounds = get_unique_compounds(drugmatrix_hits, "Compound")
function_get_unique_fold_changes_datasets = function get_unique_fold_changes_datasets(array_of_datasets) {
var set_of_datasets = new Set(array_of_datasets.map(x => x["Dataset ID"]));
return Array.from(set_of_datasets).filter(function(item) {return item != null;})
}
get_unique_compounds = function(array_of_datasets, compound_name_column) {
var set_of_datasets = new Set(array_of_datasets.map(x => x[compound_name_column]));
return Array.from(set_of_datasets).filter(function(item) {return item != null;})
}
tox21_hits.filter(function(item) {return item["Substance name"] == "Zileuton"})[0].SMILES.similarity
tox21_hits = function_fetch_dataset_to_array(tox21_summary_url, query_options)
tggates_hits = function_fetch_dataset_to_array(tggates_summary_url, query_options)
drugmatrix_hits = function_fetch_dataset_to_array(drugmatrix_summary_url, query_options)
function_fetch_dataset_to_array = function fetch_dataset_to_array(dataset_url, query_options) {
return fetch(dataset_url, query_options).then(response_dataset => {
if (!response_dataset.ok) throw new Error(response_dataset.status);
return response_dataset.json();}).then(json => json.results.map(x => x.data));
}
tox21_summary_url = base_url_edelweiss + tox21_summary_id + "/versions/latest/data"
tggates_summary_url = base_url_edelweiss + tggates_summary_id + "/versions/latest/data"
drugmatrix_summary_url = base_url_edelweiss + drugmatrix_summary_id + "/versions/latest/data"
tox21_summary_id = "8cc0534a-5e52-4253-a736-40f616b84a41"
tggates_summary_id = '16a4418e-01ff-4dcd-9b2d-90ef00565655'
drugmatrix_summary_id = '9c1c25e0-2943-4c36-9123-ccbb1f4ef2ae'
base_url_edelweiss = 'https://api.kit.edelweissconnect.com/datasets/'
query_options = {
return {
method: 'POST',
body: JSON.stringify({
condition: {
gt: [{tanimotoSimilarity: [smiles, { column: ["SMILES"] }]}, tanimoto_threshold]
},
orderBy: [
{expression: {tanimotoSimilarity: [smiles, { column: ["SMILES"] }]}, ordering: "descending"}
]
})
}
}
svg = function_get_svg_from_smiles(smiles)
function_get_svg_from_smiles = function get_svg_from_smiles(smiles) {
const safeSmiles = encodeURIComponent(smiles);
const svg_url = "https://chemidconvert.cloud.douglasconnect.com/v1/asSvg?width=200&height=200&smiles=" + safeSmiles;
return fetch(svg_url).then(response => response.text());
}
import {vl} from "@vega/vega-lite-api"
import {select, textarea, text, slider} from "@jashkenas/inputs"
import {Select, Checkbox} from "@observablehq/inputs"
Purpose-built for displays of data
Observable is your go-to platform for exploring data and creating expressive data visualizations. Use reactive JavaScript notebooks for prototyping and a collaborative canvas for visual data exploration and dashboard creation.
