gwasviewer

A JBrowse plugin that adds some custom glyphs for variants and GWAS data

Options

heightScaler - Numerical value to collapse or exapand y-axis. By default 1px is 1 unit in -log10(pvalue) units, so making heightScaler: 2 makes smaller -log10(pvalue) scores more visible. Note: if your scoreFun is just returning feature.get('score'), make this equal to maxHeight to just be scale between 0.0-1.0
maxHeight - Numerical value to set track height. Default: 210px
showLabels - False by default
scoreFun - A function to give the score for the graph. Default: "function(feature) { Math.log(feature.get('score')) }", can define similar in config file
useYAxis - Shows a y-axis bar. Default: true
style->color - A color callback for the variables
style->label - A string or callback for the feature label.
style->width - Size of the circle glyph. Default 10
style->borderColor - Default black, set to empty string or null to remove
clearRect - runs clear rect after mouseovers. default: false

Fetch API options

useMyVariantInfo - Clicking on variant launches myvariant.info popup. Default: false
useMyVariantInfoURL - Default: 'https://myvariant.info/v1/query?q='
useMyVariantInfoArgs - Default: '&fields=all'
useEnsemblR2 - Clicking on a variant gets R^2 data from Ensembl REST API
useEnsemblURL - Default: 'https://rest.ensembl.org/ld/human'
useEnsemblArgs - Default: '?content-type=application/json;population_name=1000GENOMES:phase_3:KHV'

Notes

It could be useful to only show labels above a certain score for style->label, e.g.

"style": { "label": "function(feature) { return -Math.log(feature.get('score'))>50 ? feature.get('name') : null; }" }
The circle glyph is used just for convenience, but it is by nature larger than the actual variant in genome coordinates on the screen, which can cause minor rendering issues like having half circles at block boundaries
The useMyVariantInfo and useEnsemblR2 are exclusive to one another, as they both use click action

Example configuration

Using the "BEDTabix" class

  {
     "maxFeatureScreenDensity": 6,
     "storeClass" : "JBrowse/Store/SeqFeature/BEDTabix",
     "urlTemplate" : "HDL_pval_name.sort.bed.gz",
     "label" : "HDL_pvals",
     "type" : "GWASViewer/View/Track/VariantPlotter"
  }

Using a track loaded with flatfile-to-json.pl (GWASViewer can load GFF file with scores, or a BED with a score column)

  {
     "maxFeatureScreenDensity": 6,
     "storeClass" : "JBrowse/Store/SeqFeature/NCList",
     "urlTemplate" : "tracks/HDL_pval_name/{refseq}/trackData.json",
     "label" : "HDL_pvals",
     "type" : "GWASViewer/View/Track/VariantPlotter"
  }

Note you can use flatfile-to-json.pl with --trackType "GWASViewer/View/Track/VariantPlotter" to automatically set this up

Screenshot

Basic GWASViewer track

Shows track with useEnsemblR2 enabled, which fetches R^2 from Ensembl for variants in linkage with a variant that is clicked (pink diamond)

Install

Clone repo into plugins folder in JBrowse and name folder GWASViewer
Add "plugins": ["GWASViewer"] to trackList.json or jbrowse_conf.json

Still in beta! Feel free to provide feedback

Test data

Test GWASViewer tracks are in the test/data/ directory, visit http://localhost/jbrowse/?data=plugins/GWASViewer/test/data or similar to view

elsiklab / gwasviewer

readme