Merge pull request #441 from piotr-gawron/vcf-allel-vis

add bar graph support for VCFs and enable subscribing to variant track clicks (thanks @piotr-gawron!)
hammerlab · May 17, 2017 · 6399869 · 6399869
2 parents 189b18b + 8f50dd4
commit 6399869
Show file tree

Hide file tree

Showing 13 changed files with 236 additions and 14 deletions.
diff --git a/examples/data.js b/examples/data.js
@@ -28,6 +28,16 @@ var sources = [
     data: pileup.formats.vcf({
       url: '/test-data/snv.chr17.vcf'
     }),
+    options: {
+      variantHeightByFrequency: true,
+      onVariantClicked: function(data) {
+        var content = "Variants:\n";
+        for (var i =0;i< data.length;i++) {
+          content +=data[i].id+" - "+data[i].vcfLine+"\n";
+        }
+        alert(content);
+      },
+    },
     name: 'Variants'
   },
   {

diff --git a/examples/index.html b/examples/index.html
@@ -3,6 +3,7 @@
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
 <link rel="stylesheet" href="../style/pileup.css" />
 <link rel="stylesheet" href="demo.css" />
+<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/latest/css/bootstrap.min.css">
 </head>
 
 <body>

diff --git a/src/main/Root.js b/src/main/Root.js
@@ -97,6 +97,7 @@ class Root extends React.Component {
             range={this.state.range}
             onRangeChange={this.handleRangeChange.bind(this)}
             source={track.source}
+            options={track.track.options}
             referenceSource={this.props.referenceSource}
             ref = {(c) => {this.trackReactElements[intKey]=c}}
           />);

diff --git a/src/main/VisualizationWrapper.js b/src/main/VisualizationWrapper.js
@@ -27,6 +27,7 @@ type Props = {
   onRangeChange: (newRange: GenomeRange) => void;
   referenceSource: TwoBitSource;
   source: any;
+  options: ?Object;
 };
 
 class VisualizationWrapper extends React.Component {
@@ -136,14 +137,15 @@ class VisualizationWrapper extends React.Component {
     if (!range) {
       return <EmptyTrack className={component.displayName} />;
     }
+    var options = _.extend({},this.props.visualization.options,this.props.options);
 
     var el = React.createElement(component, ({
       range: range,
       source: this.props.source,
       referenceSource: this.props.referenceSource,
       width: this.state.width,
       height: this.state.height,
-      options: this.props.visualization.options
+      options: options
     } : VizProps));
 
     return <div className='drag-wrapper'>{el}</div>;

diff --git a/src/main/data/vcf.js b/src/main/data/vcf.js
@@ -16,6 +16,13 @@ export type Variant = {
   position: number;
   ref: string;
   alt: string;
+  id: string;
+  //this is the bigest allel frequency for single vcf entry
+  //single vcf entry might contain more than one variant like the example below
+  //20 1110696 rs6040355 A G,T 67 PASS NS=2;DP=10;AF=0.333,0.667;AA=T;DB
+  majorFrequency: ?number;
+  //this is the smallest allel frequency for single vcf entry
+  minorFrequency: ?number;
   vcfLine: string;
 }
 
@@ -41,12 +48,33 @@ function extractLocusLine(vcfLine: string): LocusLine {
 
 function extractVariant(vcfLine: string): Variant {
   var parts = vcfLine.split('\t');
+  var maxFrequency = null;
+  var minFrequency = null;
+  if (parts.length>=7){
+    var params = parts[7].split(';');
+    for (var i=0;i<params.length;i++) {
+      var param = params[i];
+      if (param.startsWith("AF=")) {
+        maxFrequency = 0.0;
+        minFrequency = 1.0;
+        var frequenciesStrings = param.substr(3).split(",");
+        for (var j=0;j<frequenciesStrings.length;j++) {
+          var currentFrequency = parseFloat(frequenciesStrings[j]);
+          maxFrequency = Math.max(maxFrequency, currentFrequency);
+          minFrequency = Math.min(minFrequency, currentFrequency);
+        }
+      }
+    }
+  }
 
   return {
     contig: parts[0],
     position: Number(parts[1]),
+    id: parts[2],
     ref: parts[3],
     alt: parts[4],
+    majorFrequency: maxFrequency,
+    minorFrequency: minFrequency,
     vcfLine
   };
 }

diff --git a/src/main/pileup.js b/src/main/pileup.js
@@ -5,6 +5,7 @@
 'use strict';
 
 import type {Track, VisualizedTrack, VizWithOptions} from './types';
+import {AllelFrequencyStrategy} from './types';
 
 import _ from 'underscore';
 import React from 'react';
@@ -176,6 +177,11 @@ var pileup = {
     variants: makeVizObject(VariantTrack),
     pileup:   makeVizObject(PileupTrack)
   },
+  enum: {
+    variants: {
+      allelFrequencyStrategy: AllelFrequencyStrategy,
+    },
+  },
   version: '0.6.8'
 };
 

diff --git a/src/main/types.js b/src/main/types.js
@@ -13,6 +13,11 @@
 
 import type React from 'react';
 
+export const AllelFrequencyStrategy = {
+  Minor : {name: "Minor"},
+  Major : {name: "Major"},
+};
+
 export type VizWithOptions = {
   component: ReactClass;
   options: ?Object;
@@ -23,7 +28,8 @@ export type Track = {
   data: Object;  // This is a DataSource object
   name?: string;
   cssClass?: string;
-  isReference?: boolean
+  isReference?: boolean;
+  options?: Object
 }
 
 export type VisualizedTrack = {

diff --git a/src/main/viz/VariantTrack.js b/src/main/viz/VariantTrack.js
@@ -3,6 +3,8 @@
  * @flow
  */
 'use strict';
+import {AllelFrequencyStrategy} from '../types';
+
 
 import type {VcfDataSource} from '../sources/VcfDataSource';
 import type {Variant} from '../data/vcf';
@@ -20,17 +22,17 @@ import canvasUtils from './canvas-utils';
 import dataCanvas from 'data-canvas';
 import style from '../style';
 
-
 class VariantTrack extends React.Component {
   props: VizProps & {source: VcfDataSource};
-  state: void;  // no state
+
+  state: void;
 
   constructor(props: Object) {
     super(props);
   }
 
   render(): any {
-    return <canvas onClick={this.handleClick} />;
+    return <canvas onClick={this.handleClick.bind(this)} />;
   }
 
   componentDidMount() {
@@ -80,11 +82,31 @@ class VariantTrack extends React.Component {
     ctx.fillStyle = style.VARIANT_FILL;
     ctx.strokeStyle = style.VARIANT_STROKE;
     variants.forEach(variant => {
+      var variantHeightRatio = 1.0;
+      if (this.props.options.variantHeightByFrequency) {
+        var frequency = null;
+        if (this.props.options.allelFrequencyStrategy === undefined) { //default startegy
+          frequency = variant.majorFrequency;
+        } else if (this.props.options.allelFrequencyStrategy === AllelFrequencyStrategy.Major) {
+          frequency = variant.majorFrequency;
+        } else if (this.props.options.allelFrequencyStrategy === AllelFrequencyStrategy.Minor) {
+          frequency = variant.minorFrequency;
+        } else {
+          console.log("Unknown AllelFrequencyStrategy: ",this.props.options.allelFrequencyStrategy);
+        }
+        if (frequency !== null && frequency !== undefined) {
+          variantHeightRatio = frequency;
+        }
+      }
+      var height = style.VARIANT_HEIGHT*variantHeightRatio;
+      var variantY = y - 0.5 + style.VARIANT_HEIGHT - height;
+      var variantX = Math.round(scale(variant.position)) - 0.5;
+      var width = Math.round(scale(variant.position + 1)) - 0.5 - variantX;
+
       ctx.pushObject(variant);
-      var x = Math.round(scale(variant.position));
-      var width = Math.round(scale(variant.position + 1)) - 1 - x;
-      ctx.fillRect(x - 0.5, y - 0.5, width, style.VARIANT_HEIGHT);
-      ctx.strokeRect(x - 0.5, y - 0.5, width, style.VARIANT_HEIGHT);
+
+      ctx.fillRect(variantX, variantY, width, height);
+      ctx.strokeRect(variantX, variantY, width, height);
       ctx.popObject();
     });
 
@@ -99,10 +121,19 @@ class VariantTrack extends React.Component {
         ctx = canvasUtils.getContext(canvas),
         trackingCtx = new dataCanvas.ClickTrackingContext(ctx, x, y);
     this.renderScene(trackingCtx);
-    var variant = trackingCtx.hit && trackingCtx.hit[0];
-    var alert = window.alert || console.log;
-    if (variant) {
-      alert(JSON.stringify(variant));
+
+    var variants = trackingCtx.hit;
+    if (variants && variants.length>0) {
+      var data = [];
+      for (var i=0;i<variants.length;i++) {
+        data.push({id: variants[i].id,vcfLine:variants[i].vcfLine});
+      }
+      //user provided function for displaying popup
+      if (typeof this.props.options.onVariantClicked  === "function") {
+        this.props.options.onVariantClicked(data);
+      } else {
+        console.log("Variants clicked: ", data);
+      }
     }
   }
 }

diff --git a/src/test/data/vcf-test.js b/src/test/data/vcf-test.js
@@ -46,6 +46,28 @@ describe('VCF', function() {
     });
   });
 
+  it('should have frequency', function() {
+    var vcf = new VcfFile(new RemoteFile('/test-data/allelFrequency.vcf'));
+    var range = new ContigInterval('chr20', 61790, 61800);
+    return vcf.getFeaturesInRange(range).then(features => {
+      expect(features).to.have.length(1);
+      expect(features[0].contig).to.equal('20');
+      expect(features[0].majorFrequency).to.equal(0.7);
+      expect(features[0].minorFrequency).to.equal(0.7);
+    });
+  });
+
+  it('should have highest frequency', function() {
+    var vcf = new VcfFile(new RemoteFile('/test-data/allelFrequency.vcf'));
+    var range = new ContigInterval('chr20', 61730, 61740);
+    return vcf.getFeaturesInRange(range).then(features => {
+      expect(features).to.have.length(1);
+      expect(features[0].contig).to.equal('20');
+      expect(features[0].majorFrequency).to.equal(0.6);
+      expect(features[0].minorFrequency).to.equal(0.3);
+    });
+  });
+
   it('should add chr', function() {
     var vcf = new VcfFile(new RemoteFile('/test-data/snv.vcf'));
     var range = new ContigInterval('chr20', 63799, 69094);

diff --git a/src/test/viz/VariantTrack.js b/src/test/viz/VariantTrack.js
@@ -0,0 +1,75 @@
+/**
+ * @flow
+ */
+'use strict';
+
+import {expect} from 'chai';
+
+import pileup from '../../main/pileup';
+import dataCanvas from 'data-canvas';
+import {waitFor} from '../async';
+
+import ReactTestUtils from 'react-addons-test-utils';
+
+describe('VariantTrack', function() {
+  var testDiv = document.getElementById('testdiv');
+
+  beforeEach(() => {
+    testDiv.style.width = '700px';
+    dataCanvas.RecordingContext.recordAll();
+  });
+
+  afterEach(() => {
+    dataCanvas.RecordingContext.reset();
+    // avoid pollution between tests.
+    testDiv.innerHTML = '';
+  });
+  var {drawnObjects} = dataCanvas.RecordingContext;
+
+  function ready() {
+    return testDiv.getElementsByTagName('canvas').length > 0 &&
+        drawnObjects(testDiv, '.variants').length > 0;
+  }
+
+  it('should render variants', function() {
+    var variantClickedData = null;
+    var variantClicked = function (data) {
+      variantClickedData = data;
+    };
+    var p = pileup.create(testDiv, {
+      range: {contig: '17', start: 9386380, stop: 9537390},
+      tracks: [
+        {
+          viz: pileup.viz.genome(),
+          data: pileup.formats.twoBit({
+            url: '/test-data/test.2bit'
+          }),
+          isReference: true
+        },
+        {
+          data: pileup.formats.vcf({
+            url: '/test-data/test.vcf'
+          }),
+          viz: pileup.viz.variants(),
+          options: {onVariantClicked: variantClicked},
+        }
+      ]
+    });
+
+    return waitFor(ready, 2000)
+      .then(() => {
+        var variants = drawnObjects(testDiv, '.variants');
+        expect(variants.length).to.be.equal(1);
+        var canvasList =  testDiv.getElementsByTagName('canvas');
+        var canvas = canvasList[1];
+        expect(variantClickedData).to.be.null;
+
+        //check clicking on variant
+        ReactTestUtils.Simulate.click(canvas,{nativeEvent: {offsetX: -0.5, offsetY: -15.5}});
+
+        expect(variantClickedData).to.not.be.null;
+        p.destroy();
+      });
+  });
+
+});
diff --git a/test-data/allelFrequency.vcf b/test-data/allelFrequency.vcf
@@ -0,0 +1,21 @@
+##fileformat=VCFv4.1
+##source=VarScan2
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total depth of quality bases">
+##INFO=<ID=SOMATIC,Number=0,Type=Flag,Description="Indicates if record is a somatic mutation">
+##INFO=<ID=SS,Number=1,Type=String,Description="Somatic status of variant (0=Reference,1=Germline,2=Somatic,3=LOH, or 5=Unknown)">
+##INFO=<ID=SSC,Number=1,Type=String,Description="Somatic score in Phred scale (0-255) derived from somatic p-value">
+##INFO=<ID=GPV,Number=1,Type=Float,Description="Fisher's Exact Test P-value of tumor+normal versus no variant for Germline calls">
+##INFO=<ID=SPV,Number=1,Type=Float,Description="Fisher's Exact Test P-value of tumor versus normal for Somatic/LOH calls">
+##FILTER=<ID=str10,Description="Less than 10% or more than 90% of variant supporting reads on one strand">
+##FILTER=<ID=indelError,Description="Likely artifact due to indel reads at this position">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FORMAT=<ID=RD,Number=1,Type=Integer,Description="Depth of reference-supporting bases (reads1)">
+##FORMAT=<ID=AD,Number=1,Type=Integer,Description="Depth of variant-supporting bases (reads2)">
+##FORMAT=<ID=FREQ,Number=1,Type=String,Description="Variant allele frequency">
+##FORMAT=<ID=DP4,Number=4,Type=Integer,Description="Strand read counts: ref/fwd, ref/rev, var/fwd, var/rev">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	NORMAL	TUMOR
+20	61795	.	G	T	.	PASS	DP=81;SS=1;SSC=2;GPV=4.6768E-16;SPV=5.4057E-1;AF=0.7	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:44:22:22:50%:16,6,9,13	0/1:.:37:18:19:51.35%:10,8,10,9
+20	62731	.	C	A,G	.	PASS	DP=68;SS=1;SSC=1;GPV=1.4855E-11;SPV=7.5053E-1;AF=0.4,0.5	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:32:17:15:46.88%:9,8,9,6	0/1:.:36:21:15:41.67%:8,13,8,7
+20	61731	.	C	A,G,T	.	PASS	DP=68;SS=1;SSC=1;GPV=1.4855E-11;SPV=7.5053E-1;AF=0.4,0.6,0.3	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:32:17:15:46.88%:9,8,9,6	0/1:.:36:21:15:41.67%:8,13,8,7
diff --git a/test-data/snv.chr17.vcf b/test-data/snv.chr17.vcf
@@ -18,7 +18,7 @@
 #CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	NORMAL	TUMOR
 17	125	.	G	T	.	PASS	DP=81;SS=1;SSC=2;GPV=4.6768E-16;SPV=5.4057E-1	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:44:22:22:50%:16,6,9,13	0/1:.:37:18:19:51.35%:10,8,10,9
 17	7512444	.	G	T	.	PASS	DP=81;SS=1;SSC=2;GPV=4.6768E-16;SPV=5.4057E-1	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:44:22:22:50%:16,6,9,13	0/1:.:37:18:19:51.35%:10,8,10,9
-17	7512454	.	C	A	.	PASS	DP=68;SS=1;SSC=1;GPV=1.4855E-11;SPV=7.5053E-1	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:32:17:15:46.88%:9,8,9,6	0/1:.:36:21:15:41.67%:8,13,8,7
+17	7512454	.	C	A	.	PASS	DP=68;SS=1;SSC=1;GPV=1.4855E-11;SPV=7.5053E-1;AF=0.73	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:32:17:15:46.88%:9,8,9,6	0/1:.:36:21:15:41.67%:8,13,8,7
 17	7512544	.	C	T	.	PASS	DP=72;SS=1;SSC=7;GPV=3.6893E-16;SPV=1.8005E-1	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:39:19:19:50%:8,11,11,8	0/1:.:33:12:21:63.64%:5,7,8,13
 17	7512644	.	G	T	.	PASS	DP=35;SS=1;SSC=0;GPV=7.8434E-5;SPV=8.2705E-1	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:21:13:8:38.1%:4,9,0,8	0/1:.:14:10:4:28.57%:2,8,0,4
 17	7512244	.	G	A	.	PASS	DP=53;SS=1;SSC=0;GPV=1.5943E-31;SPV=1E0	GT:GQ:DP:RD:AD:FREQ:DP4	1/1:.:26:0:26:100%:0,0,12,14	1/1:.:27:0:27:100%:0,0,15,12

diff --git a/test-data/test.vcf b/test-data/test.vcf
@@ -0,0 +1,19 @@
+##fileformat=VCFv4.1
+##source=VarScan2
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total depth of quality bases">
+##INFO=<ID=SOMATIC,Number=0,Type=Flag,Description="Indicates if record is a somatic mutation">
+##INFO=<ID=SS,Number=1,Type=String,Description="Somatic status of variant (0=Reference,1=Germline,2=Somatic,3=LOH, or 5=Unknown)">
+##INFO=<ID=SSC,Number=1,Type=String,Description="Somatic score in Phred scale (0-255) derived from somatic p-value">
+##INFO=<ID=GPV,Number=1,Type=Float,Description="Fisher's Exact Test P-value of tumor+normal versus no variant for Germline calls">
+##INFO=<ID=SPV,Number=1,Type=Float,Description="Fisher's Exact Test P-value of tumor versus normal for Somatic/LOH calls">
+##FILTER=<ID=str10,Description="Less than 10% or more than 90% of variant supporting reads on one strand">
+##FILTER=<ID=indelError,Description="Likely artifact due to indel reads at this position">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FORMAT=<ID=RD,Number=1,Type=Integer,Description="Depth of reference-supporting bases (reads1)">
+##FORMAT=<ID=AD,Number=1,Type=Integer,Description="Depth of variant-supporting bases (reads2)">
+##FORMAT=<ID=FREQ,Number=1,Type=String,Description="Variant allele frequency">
+##FORMAT=<ID=DP4,Number=4,Type=Integer,Description="Strand read counts: ref/fwd, ref/rev, var/fwd, var/rev">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	NORMAL	TUMOR
+17	9386385	.	G	T	.	PASS	DP=81;SS=1;SSC=2;GPV=4.6768E-16;SPV=5.4057E-1;AF=0.7	GT:GQ:DP:RD:AD:FREQ:DP4	0/1:.:44:22:22:50%:16,6,9,13	0/1:.:37:18:19:51.35%:10,8,10,9