Published
Edited
Apr 17, 2020
1 fork
1 star
md`# The informality of InDel Genome Variation in SARS-CoV-2`
md`### Figures
#### (Figure 1)
Demonstrates the distribution of Whole Genome lengths (in bp) we observe in sequences submitted for the SARS-CoV-2 entries in GISAID. This dataset compromises XXX whole genomes which were pulled down from the GISAID repository on **04-14-2020**. Additionally, sequences were filtered to include whole genomes (> 29,000 bp) and represent samples from Human hosts only in order to examine intrahost evolution. This dataset is comprised of **n=8,064** whole genome sequences`
md`#### (Figure 2)
Breakdown of Whole Genome length sequence distributions by Collection Date and colored by submitting Country
`
vl.markPoint().data(flt_data2).width(width - 200).encode(
vl.x().fieldT('Date'),
vl.y().fieldQ('WG_Length')
.scale({domain: [29000, 30400]}).scale({domain: [29000, 30400]}),
vl.color().fieldN('Locale'),
vl.tooltip(['ID', 'Locale', 'Date', 'WG_Length']) // show the Name and Origin fields in a tooltip
).render()
md`#### (Figure 3)
What does whole genome length variability mean in the context of an individual gene?
For this, we examine a single gene within SARS-CoV-2, the spike protein as an example
The figure below shows spike gene size (in bp) distributed across the submission date to GISAID.
`
//Additionally, whole genomes display wide variability but genes, such as spike, have relatively low sequence length variance. Question is what are the hot spots?
viewof layeredTrend = vl.data(spikeData).width(width - 200)
.encode(
vl.x().fieldT(" Date"),
vl.y().fieldQ(" Nucleotides")
.scale({domain: [3780
, 3840]}),
vl.tooltip([' ID', ' Date', ' Nucleotides'])
)
.layer(
vl.markCircle()
,
vl.markErrorband({ extent: "iqr" , interpolate: "basis"}),
vl.markLine()
.encode(
vl.y().mean("Nucleotides")
)
)
.render()
md`#### (Figure 4)
Inferred Indel Sites from protein alignments.
Note: Site 265 was filtered out of this alignment, it seems to be the result of an insertion in a single sequence and warrants further investigation. It also forces the rest of the sequences to incur a gap at that particlar site in the alignment and overstates the reporting in the table below.
Table of Sites and Frequencies.
`
//Can point to alignment errors
table(flt_spikeInDels, {
rank: true,
style: 'compact',
columns: {
Track: {
formatter(val, i) {
return html`<strong>${val[0]}</strong> by ${val[1]}`;
}
},
Streams: {
formatter: d3.format(',')
}
}
})
// md`### Figure 5. InDel rate and Genome coverage. maybe it is an issue with long read seqs or low coverage? `
md`### Data `
data = Object.assign(d3.csvParse(await FileAttachment("test_wg_variation_daybyday_coloredbycountry.csv").text(), ({WG_Length}) => +WG_Length), {x: "WG Length (bp)", y: "Occurences (#)"})
data2 = d3.csvParse(await FileAttachment("test_wg_variation_daybyday_coloredbycountry.csv").text(), d3.autoType)
//flt_data2 = data2.filter(d => d.Date > 2018)
flt_data2 = data2.filter(d => d.Date > new Date(2018, 0, 1))
//Nucleotide length variability
spikeData = d3.csvParse(await FileAttachment("gisaid_cov2020_sequences.fasta.S_nuc.fas@3.csv").text(), d3.autoType)
spikeInDels = d3.csvParse(
await FileAttachment("gisaid_cov2020_sequences.fasta.S.protein_all_withref.fas_table.csv").text()
)
//Filter SpikeIndels
flt_spikeInDels = spikeInDels.filter(d => d.SiteDeletion != 265)
md`### Plotting `
bins = d3.histogram()
.domain(x.domain())
.thresholds(x.ticks(40))
(data)
x = d3.scaleLinear()
.domain(d3.extent(data)).nice()
.range([margin.left, width - margin.right])
y = d3.scaleLinear()
.domain([0, d3.max(bins, d => d.length)]).nice()
.range([height - margin.bottom, margin.top])
md`### Declares `
margin = ({top: 25, right: 20, bottom: 35, left: 40})
height = 600
grid = g => g
.attr("stroke", "currentColor")
.attr("stroke-opacity", 0.1)
.call(g => g.append("g")
.selectAll("line")
.data(x.ticks())
.join("line")
.attr("x1", d => 0.5 + x(d))
.attr("x2", d => 0.5 + x(d))
.attr("y1", margin.top)
.attr("y2", height - margin.bottom))
.call(g => g.append("g")
.selectAll("line")
.data(y.ticks())
.join("line")
.attr("y1", d => 0.5 + y(d))
.attr("y2", d => 0.5 + y(d))
.attr("x1", margin.left)
.attr("x2", width - margin.right));
xAxis = g => g
.attr("transform", `translate(0,${height - margin.bottom})`)
.call(d3.axisBottom(x).ticks(width / 80 ).tickSizeOuter(0))
.call(g => g.append("text")
.attr("x", width - margin.right)
.attr("y", -4)
.attr("fill", "currentColor")
.attr("font-weight", "bold")
.attr("text-anchor", "end")
.text(data.x))
yAxis = g => g
.attr("transform", `translate(${margin.left},0)`)
.call(d3.axisLeft(y).ticks(height / 40))
.call(g => g.select(".domain").remove())
.call(g => g.select(".tick:last-of-type text").clone()
.attr("x", 4)
.attr("text-anchor", "start")
.attr("font-weight", "bold")
.text(data.y))
md`### Dependencies `
import {table} from "@tmcw/tables@513"
import {vl} from '@vega/vega-lite-api'
import {printTable} from '@uwdata/data-utilities'
d3 = require("d3@5")
embed = require("vega-embed@3")
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.