dataviz & research
//chart_treatment_by_severity_prevention = generate_research_evolution_treatment_type(
// "Research evolution for treatment types : prevention",
// d => d["Study aim"] === "Prevention"
//)
// No more relevant as
//chart_treatment_by_severity_post = generate_research_evolution_treatment_type(
// "Research evolution for treatment types : post treatment",
// d => d["Study aim"] === "Post treatment"
//)
data_agg_authors = d3.nest().key(function(d) { return d["First author"]; }).rollup(function(values) { return values.length }).entries(data_agg.map(d => d.value))
data_agg
data_agg_treatment1 = get_data_agg_field("Treatment type")
data_agg_treatment1
data_agg.map(d => d.value.values)
import {vl} from '@vega/vega-lite-api'
data
data_agg
data_agg.map(function(d, i) { return d.value.values.map(e => e["Treatment type"]) })
html`<img src="https://covid-nma.com/wp-content/uploads/2020/04/network-mild-070420-800x450.jpg" width=500>
<br>
`
html`<img src="https://covid-nma.com/wp-content/uploads/2020/04/network-severe-070420-800x450.jpg" width=500>
<br>
`
html`<img src="https://covid-nma.com/wp-content/uploads/2020/04/network-heterogeneous-070420-800x450.jpg" width=600>
<br>
`
html`<img src="https://covid-nma.com/wp-content/uploads/2020/04/Network-europe-070420-800x450.jpg" width=600>
<br>`
html`<img src="https://covid-nma.com/wp-content/uploads/2020/04/Rplot-treatment-type-eu_page-0001-800x566.jpg" width=600>
<br>
`
d3 = require("d3@5")
vegalite = require("@observablehq/vega-lite@0.2")
data = d3.csvParse(await FileAttachment("Database 080420_140420.csv").text())
data_agg = d3.nest().key(function(d) { return d["Trial ID"]; }).rollup(function(values) {
return {values: values,
"Recruitment status": values[0]["Recruitment status"],
"Countries": values[0]["Countries"],
"Center": values[0]["Center"],
"First author": values[0]["First author"],
"Publication date online": values[0]["Publication date online"],
"Trial registration number": values[0]["Trial registration number"],
"Funding": values[0]["Funding"],
"Severity": values[0]["Severity scale"],
"Study design": values[0]["Study design"],
"registration_code": values[0]["Trial registration number"].slice(0, 3)
}
}).entries(data);
Treatment_type1 = new Set(data.map(d => d["Treatment type"]))
Treatment_type1_list = Array.from(Treatment_type1)
Array.from(Treatment_type1).map(d => 0)
md`Using [UpSets](https://observablehq.com/@sgratzl/upset-observable-example?collection=@sgratzl/upset-visualization-technique) to visualize co-occurences`
all_co_binary = data_agg.map(function(d, i) {
let a = Array.from(Treatment_type1).map(d => 0)
d.value.values.forEach(function(e) {
i = Treatment_type1_list.indexOf(e["Treatment type1"])
a[i] = 1
})
return a
})
// EXPERIMENTAL
// Only with a dataset otherwise too many intersections
// We need to reduce the number of categories
all_co_data = data_agg.filter((d, i) => i < 30).map(function(d, i) {
let sets = []
d.value.values.forEach(function(e) {
// if(e["Treatment type1"] != "*")
sets.push(e["Treatment type1"])
})
let a = {}
a["name"] = d.value["Trial registration number"]
a["Trial registration number"] = d.value["Trial registration number"]
a["First author"] = d.value["First author"]
a["Publication date online"] = d.value["Publication date online"]
a["sets"] = sets
return a
})
sets = extractSets(all_co_data)
intersections = generateIntersections(sets).sort(
(a, b) => b.cardinality - a.cardinality
)
intersections
viewof selection = UpSet(sets, intersections)
printTable(selection ? selection.elems.map(d => d): "")
selection
bycontinent = new Map(d3.nest().key(d => d).rollup(v => v.length).entries(Array.from(countriesIndex).map(d => Array.from(new Set(d[1].map(d => continents.get(d))))).flat()).map(d => [d.key, d.value]))
import {
checkbox,
slider,
button,
text,
select,
autoSelect,
radio
} from "@jashkenas/inputs"
import {printTable} from '@uwdata/data-utilities'
// https://observablehq.com/d/3c709ffbafcdb388
import {
map,
viewof textFilter,
viewof recruitmentFilter,
viewof zoomType,
countriesIndex,
continents
} from "3c709ffbafcdb388"
md`Other Viz
https://observablehq.com/d/8215857ffae42ded
`
stackedBar = vl.markBar()
.data(data_agg.map(d => d.value))
.encode(
vl.x().fieldO("Countries"),//.timeUnit("utcmonth"),
vl.y().count(), // .stack("normalize").axis({ format: "%" }),
vl.color().fieldN("continent") // .scale({ range: weatherColors })
)
.render()
// Without Date filter
chart_date_strip = embed({
data: {values: data},
mark: "tick",
width: width/2,
encoding: {
x: {field: "Registration date", type: "temporal"}
}
})
embed = require("vega-embed@6")
vegalite({
data: {values: data.filter(d => (new Date(d["Registration date"]) > new Date("2019")) )},
mark: "area",
width: width/2,
encoding: {
y: {aggregate: "count", field: "Treatment type1", type: "quantitative"},
x: {field: "Registration date", type: "temporal"},
color: {field: "Treatment type", "type":"nominal", "scale":{"scheme": "category20b"}}
},
})
md`OLD CHARTS`
html`<img src="https://covid-nma.com/wp-content/uploads/2020/04/Rplot-countries_page-0001-800x566.jpg" width=200>`
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.
