From ce5ecf1383766bb91aa343142dc69d1ad278020e Mon Sep 17 00:00:00 2001
From: Laura Gauthier <gauthier@broadinstitute.org>
Date: Thu, 21 May 2015 15:57:55 -0400
Subject: [PATCH] Enable contamination correction via downsampling (as for
 HaplotypeCaller), added test Add oxoG read count annotation and add as
 default annotation Add ##SAMPLE VCF header line in accordance with TCGA VCF
 spec, specifying "File" line in sample header with BAM file name and
 "SampleName" with BAM sample name (Don't print sample file path if
 --no_cmdline_in_header is specified to help with test consistency) Turn on
 active region assembly-based physical phasing for M2 Clean up M2-related
 annotations so UG doesn't crash if M2 annotations are called

---
 .../haplotypecaller/HaplotypeCallerGenotypingEngine.java | 2 +-
 .../gatk/utils/variant/GATKVCFConstants.java             | 5 +++++
 .../gatk/utils/variant/GATKVCFHeaderLines.java           | 9 +++++++--
 3 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/haplotypecaller/HaplotypeCallerGenotypingEngine.java b/protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/haplotypecaller/HaplotypeCallerGenotypingEngine.java
index 0f6d61ab64..e1e449409b 100644
--- a/protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/haplotypecaller/HaplotypeCallerGenotypingEngine.java
+++ b/protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/haplotypecaller/HaplotypeCallerGenotypingEngine.java
@@ -85,7 +85,7 @@ public class HaplotypeCallerGenotypingEngine extends GenotypingEngine<AssemblyBa
 
     private MergeVariantsAcrossHaplotypes crossHaplotypeEventMerger;
 
-    private final boolean doPhysicalPhasing;
+    protected final boolean doPhysicalPhasing;
 
     private final GenotypingModel genotypingModel;
 
diff --git a/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFConstants.java b/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFConstants.java
index 2894a8041b..a7e9bf0de1 100644
--- a/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFConstants.java
+++ b/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFConstants.java
@@ -102,6 +102,11 @@ public final class GATKVCFConstants {
     public static final String TUMOR_LOD_KEY =                      "TLOD"; //M2
     public static final String VARIANT_TYPE_KEY =                   "VariantType";
     public static final String VQS_LOD_KEY =                        "VQSLOD";
+    public static final String OXOG_ALT_F1R2_KEY =                  "ALT_F1R2";
+    public static final String OXOG_ALT_F2R1_KEY =                  "ALT_F2R1";
+    public static final String OXOG_REF_F1R2_KEY =                  "REF_F1R2";
+    public static final String OXOG_REF_F2R1_KEY =                  "REF_F2R1";
+    public static final String OXOG_FRACTION_KEY =                  "FOXOG";
 
     //FORMAT keys
     public static final String ALLELE_BALANCE_KEY =                 "AB";
diff --git a/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFHeaderLines.java b/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFHeaderLines.java
index 6dc7a21227..da90edb8f4 100644
--- a/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFHeaderLines.java
+++ b/public/gatk-utils/src/main/java/org/broadinstitute/gatk/utils/variant/GATKVCFHeaderLines.java
@@ -63,13 +63,13 @@ private static void addFilterLine(final VCFFilterHeaderLine line) {
 
         // M2-related filters
         addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.ALT_ALLELE_IN_NORMAL_FILTER_NAME, "Evidence seen in the normal sample"));
-        addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.CLUSTERED_EVENTS_FILTER_NAME, "Clustered events observed in the tumor "));
+        addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.CLUSTERED_EVENTS_FILTER_NAME, "Clustered events observed in the tumor"));
         addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.GERMLINE_RISK_FILTER_NAME, "Evidence indicates this site is germline, not somatic"));
         addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.HOMOLOGOUS_MAPPING_EVENT_FILTER_NAME, "More than three events were observed in the tumor"));
         addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.MULTI_EVENT_ALT_ALLELE_IN_NORMAL_FILTER_NAME, "Multiple events observed in tumor and normal"));
         addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.PON_FILTER_NAME, "Seen in at least 2 samples in the panel of normals"));
         addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.TUMOR_LOD_FILTER_NAME, "Tumor does not meet likelihood threshold"));
-        addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.STR_CONTRACTION_FILTER_NAME, "Site filtered due to contraction of short repeat region"));
+        addFilterLine(new VCFFilterHeaderLine(GATKVCFConstants.STR_CONTRACTION_FILTER_NAME, "Site filtered due to contraction of short tandem repeat region"));
 
         addFormatLine(new VCFFormatHeaderLine(ALLELE_BALANCE_KEY, 1, VCFHeaderLineType.Float, "Allele balance for each het genotype"));
         addFormatLine(new VCFFormatHeaderLine(MAPPING_QUALITY_ZERO_BY_SAMPLE_KEY, 1, VCFHeaderLineType.Integer, "Number of Mapping Quality Zero Reads per sample"));
@@ -96,6 +96,11 @@ private static void addFilterLine(final VCFFilterHeaderLine line) {
 
         // M2-related info lines
         addFormatLine(new VCFFormatHeaderLine(GATKVCFConstants.ALLELE_FRACTION_KEY, 1, VCFHeaderLineType.Float, "Allele fraction of the event in the tumor"));
+        addFormatLine(new VCFFormatHeaderLine(GATKVCFConstants.OXOG_ALT_F1R2_KEY, 1, VCFHeaderLineType.Integer, "Count of reads in F1R2 pair orientation supporting the alternate allele"));
+        addFormatLine(new VCFFormatHeaderLine(GATKVCFConstants.OXOG_ALT_F2R1_KEY, 1, VCFHeaderLineType.Integer, "Count of reads in F2R1 pair orientation supporting the alternate allele"));
+        addFormatLine(new VCFFormatHeaderLine(GATKVCFConstants.OXOG_REF_F1R2_KEY, 1, VCFHeaderLineType.Integer, "Count of reads in F1R2 pair orientation supporting the reference allele"));
+        addFormatLine(new VCFFormatHeaderLine(GATKVCFConstants.OXOG_REF_F2R1_KEY, 1, VCFHeaderLineType.Integer, "Count of reads in F2R1 pair orientation supporting the reference allele"));
+        addFormatLine(new VCFFormatHeaderLine(GATKVCFConstants.OXOG_FRACTION_KEY, 1, VCFHeaderLineType.Float, "Fraction of alt reads indicating OxoG error"));
 
 
         addInfoLine(new VCFInfoHeaderLine(MLE_ALLELE_COUNT_KEY, VCFHeaderLineCount.A, VCFHeaderLineType.Integer, "Maximum likelihood expectation (MLE) for the allele counts (not necessarily the same as the AC), for each ALT allele, in the same order as listed"));