From 53a7d5cbae245296e2c112b8a1a441255823b27c Mon Sep 17 00:00:00 2001
From: Eric Banks <ebanks@broadinstitute.org>
Date: Thu, 20 Feb 2014 00:53:13 -0500
Subject: [PATCH] Fixing a bug in the GVCF writer.

The writer was never resetting the pointer to the end of the last non-ref VariantContext that it saw.
This was fine except when it jumped to a new contig - and a lower position on that contig - where it
thought that it was still part of that previous non-ref VariantContext so wouldn't emit a reference
block.  Therefore, ref blocks were missing from the beginnings of all chromosomes (except chr1).

Added unit test to cover this case.
---
 .../sting/utils/gvcf/GVCFWriter.java          | 15 +++++++--
 .../sting/utils/gvcf/GVCFWriterUnitTest.java  | 31 +++++++++++++++++++
 2 files changed, 43 insertions(+), 3 deletions(-)

diff --git a/protected/gatk-protected/src/main/java/org/broadinstitute/sting/utils/gvcf/GVCFWriter.java b/protected/gatk-protected/src/main/java/org/broadinstitute/sting/utils/gvcf/GVCFWriter.java
index aa269779b..71d61c920 100644
--- a/protected/gatk-protected/src/main/java/org/broadinstitute/sting/utils/gvcf/GVCFWriter.java
+++ b/protected/gatk-protected/src/main/java/org/broadinstitute/sting/utils/gvcf/GVCFWriter.java
@@ -84,6 +84,7 @@ public class GVCFWriter implements VariantContextWriter {
 
     /** fields updated on the fly during GVCFWriter operation */
     int nextAvailableStart = -1;
+    String contigOfNextAvailableStart = null;
     private String sampleName = null;
     private HomRefBlock currentBlock = null;
 
@@ -187,10 +188,17 @@ public class GVCFWriter implements VariantContextWriter {
      * @return a VariantContext to be emitted, or null if non is appropriate
      */
     protected VariantContext addHomRefSite(final VariantContext vc, final Genotype g) {
-        if ( nextAvailableStart != -1 && vc.getStart() <= nextAvailableStart ) {
+        if ( nextAvailableStart != -1 ) {
             // don't create blocks while the hom-ref site falls before nextAvailableStart (for deletions)
-            return null;
-        } else if ( currentBlock == null ) {
+            if ( vc.getStart() <= nextAvailableStart && vc.getChr().equals(contigOfNextAvailableStart) ) {
+                return null;
+            }
+            // otherwise, reset to non-relevant
+            nextAvailableStart = -1;
+            contigOfNextAvailableStart = null;
+        }
+
+        if ( currentBlock == null ) {
             currentBlock = createNewBlock(vc, g);
             return null;
         } else if ( currentBlock.withinBounds(g.getGQ()) ) {
@@ -302,6 +310,7 @@ public class GVCFWriter implements VariantContextWriter {
                 // g is variant, so flush the bands and emit vc
                 emitCurrentBlock();
                 nextAvailableStart = vc.getEnd();
+                contigOfNextAvailableStart = vc.getChr();
                 underlyingWriter.add(vc);
             }
         }
diff --git a/protected/gatk-protected/src/test/java/org/broadinstitute/sting/utils/gvcf/GVCFWriterUnitTest.java b/protected/gatk-protected/src/test/java/org/broadinstitute/sting/utils/gvcf/GVCFWriterUnitTest.java
index 77a9714fe..cda022ab8 100644
--- a/protected/gatk-protected/src/test/java/org/broadinstitute/sting/utils/gvcf/GVCFWriterUnitTest.java
+++ b/protected/gatk-protected/src/test/java/org/broadinstitute/sting/utils/gvcf/GVCFWriterUnitTest.java
@@ -202,6 +202,37 @@ public class GVCFWriterUnitTest extends BaseTest {
         assertGoodVC(mockWriter.emitted.get(1), "21", 3, 3, false);
     }
 
+    @Test
+    public void testCrossingContigBoundaryToLowerPositionsRef() {
+        final GVCFWriter writer = new GVCFWriter(mockWriter, standardPartition);
+
+        writer.add(makeHomRef("20", 30, 30));
+        writer.add(makeHomRef("20", 31, 30));
+        Assert.assertEquals(mockWriter.emitted.size(), 0);
+        writer.add(makeHomRef("21", 10, 30));
+        Assert.assertEquals(mockWriter.emitted.size(), 1);
+        assertGoodVC(mockWriter.emitted.get(0), "20", 30, 31, false);
+        writer.add(makeNonRef("21", 11, 30));
+        Assert.assertEquals(mockWriter.emitted.size(), 3);
+        assertGoodVC(mockWriter.emitted.get(1), "21", 10, 10, false);
+        assertGoodVC(mockWriter.emitted.get(2), "21", 11, 11, true);
+    }
+
+    @Test
+    public void testCrossingContigBoundaryFromNonRefToLowerPositionsRef() {
+        final GVCFWriter writer = new GVCFWriter(mockWriter, standardPartition);
+
+        writer.add(makeNonRef("20", 20, 30));
+        Assert.assertEquals(mockWriter.emitted.size(), 1);
+        writer.add(makeHomRef("21", 10, 30));
+        Assert.assertEquals(mockWriter.emitted.size(), 1);
+        assertGoodVC(mockWriter.emitted.get(0), "20", 20, 20, true);
+        writer.add(makeNonRef("21", 11, 30));
+        Assert.assertEquals(mockWriter.emitted.size(), 3);
+        assertGoodVC(mockWriter.emitted.get(1), "21", 10, 10, false);
+        assertGoodVC(mockWriter.emitted.get(2), "21", 11, 11, true);
+    }
+
     @Test
     public void testCrossingContigBoundaryNonRef() {
         final GVCFWriter writer = new GVCFWriter(mockWriter, standardPartition);