Added - a class to count functional annotations by sample (currently for the MAF annotation strings, soon to be migrated to genomic annotator once it is up and running)

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@3988 348d0f76-0448-11de-a6fe-93d51630548a
2010-08-09 20:09:13 +00:00 · 2010-08-09 20:09:13 +00:00 · d0ecb8875a
parent 5b0b9e79ba
commit d0ecb8875a
1 changed files with 79 additions and 0 deletions
--- a/java/src/org/broadinstitute/sting/oneoffprojects/walkers/varianteval/FunctionalClassBySample.java
+++ b/java/src/org/broadinstitute/sting/oneoffprojects/walkers/varianteval/FunctionalClassBySample.java
@ -0,0 +1,79 @@
+package org.broadinstitute.sting.gatk.walkers.varianteval;
+
+import org.broad.tribble.util.variantcontext.Genotype;
+import org.broad.tribble.util.variantcontext.VariantContext;
+import org.broadinstitute.sting.gatk.contexts.AlignmentContext;
+import org.broadinstitute.sting.gatk.contexts.ReferenceContext;
+import org.broadinstitute.sting.gatk.refdata.RefMetaDataTracker;
+import org.broadinstitute.sting.gatk.walkers.varianteval.VariantEvaluatorBySample;
+import org.broadinstitute.sting.playground.utils.report.tags.Analysis;
+import org.broadinstitute.sting.gatk.walkers.varianteval.SampleDataPoint;
+import org.broadinstitute.sting.gatk.walkers.varianteval.VariantEvaluatorBySample;
+import org.broadinstitute.sting.playground.utils.report.tags.Analysis;
+import org.broadinstitute.sting.playground.utils.report.tags.DataPoint;
+
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * Extends the Per-sample variant evaluator class and returns, for each sample, the number of variants, the Ti/Tv, and
+ * the comp overlap. It does this only on sites where the sample is identified as hom var, or het.
+ */
+@Analysis(name="FunctionalClassBySample",description="Count of SNPs by functional class by sample")
+public class FunctionalClassBySample extends VariantEvaluatorBySample {
+
+    public FunctionalClassBySample(VariantEvalWalker parent) { super(parent); }
+
+    public List<SampleDataPoint> getDataPoints() {
+        List<SampleDataPoint> points = new ArrayList<SampleDataPoint>(10);
+        points.add(new FCPoint("miRNA","miRNA"));
+        points.add(new FCPoint("3'-UTR","3'-UTR"));
+        points.add(new FCPoint("Intron","Intron"));
+        points.add(new FCPoint("Splice_site","Splice_site"));
+        points.add(new FCPoint("Read-through","Read-through"));
+        points.add(new FCPoint("Nonsense","Nonsense"));
+        points.add(new FCPoint("Missense","Missense"));
+        points.add(new FCPoint("Synonymous","Synonymous"));
+        points.add(new FCPoint("5'-UTR","5'-UTR"));
+        points.add(new FCPoint("Promoter","Promoter"));
+
+        return points;
+    }
+
+    public String getTableName() { return "Functional Class Counts by Sample"; }
+
+    public String getName() { return "Functional Class Counts by Sample"; }
+
+    public int getComparisonOrder() { return 1; }
+
+    public boolean enabled() { return true; }
+
+    public boolean includeGenotype(Genotype g) { return ( ! g.isFiltered() ) && ( g.isHet() || g.isHomVar() ); }
+
+}
+
+class FCPoint extends SampleDataPoint {
+    private String matchStr;
+    private int count;
+
+    public FCPoint(String fcName, String fcMatch) {
+        super(fcName);
+        matchStr = fcMatch;
+        count = 0;
+    }
+
+    public void update1(VariantContext vc, RefMetaDataTracker tracker, ReferenceContext ref, AlignmentContext context) {
+        if ( vc == null ) {
+            return;
+        } else {
+            String type = vc.getAttributeAsString("type","none");
+            if ( type.equalsIgnoreCase(matchStr) ) {
+                count++;
+            }
+        }
+    }
+
+    public String toString() {
+        return String.format("%d",count);
+    }
+}