From 181f901126dba8a92539aebdaf111c4b1999c3c5 Mon Sep 17 00:00:00 2001
From: ebanks <ebanks@348d0f76-0448-11de-a6fe-93d51630548a>
Date: Fri, 22 Oct 2010 14:38:26 +0000
Subject: [PATCH] Fix for Ryan: don't pull reference sequence for the portions
 of reads that extend beyond the contig boundaries

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4551 348d0f76-0448-11de-a6fe-93d51630548a
---
 .../gatk/walkers/genotyper/UnifiedGenotyperEngine.java      | 4 ++--
 .../org/broadinstitute/sting/utils/sam/AlignmentUtils.java  | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/java/src/org/broadinstitute/sting/playground/gatk/walkers/genotyper/UnifiedGenotyperEngine.java b/java/src/org/broadinstitute/sting/playground/gatk/walkers/genotyper/UnifiedGenotyperEngine.java
index 641761670..76e4617a5 100755
--- a/java/src/org/broadinstitute/sting/playground/gatk/walkers/genotyper/UnifiedGenotyperEngine.java
+++ b/java/src/org/broadinstitute/sting/playground/gatk/walkers/genotyper/UnifiedGenotyperEngine.java
@@ -453,9 +453,9 @@ public class UnifiedGenotyperEngine {
                     bitset.set(i);
             }
 
-            // if a read is too long for the reference context, extend the context
+            // if a read is too long for the reference context, extend the context (being sure not to extend past the end of the chromosome)
             if ( record.getAlignmentEnd() > refContext.getWindow().getStop() ) {
-                GenomeLoc window = GenomeLocParser.createGenomeLoc(refContext.getLocus().getContig(), refContext.getWindow().getStart(), record.getAlignmentEnd());
+                GenomeLoc window = GenomeLocParser.createGenomeLoc(refContext.getLocus().getContig(), refContext.getWindow().getStart(), Math.min(record.getAlignmentEnd(), referenceReader.getSequenceDictionary().getSequence(refContext.getLocus().getContig()).getSequenceLength()));
                 byte[] bases = referenceReader.getSubsequenceAt(window.getContig(), window.getStart(), window.getStop()).getBases();
                 StringUtil.toUpperCase(bases);
                 refContext = new ReferenceContext(refContext.getLocus(), window, bases);
diff --git a/java/src/org/broadinstitute/sting/utils/sam/AlignmentUtils.java b/java/src/org/broadinstitute/sting/utils/sam/AlignmentUtils.java
index 45cca9eec..f04440368 100644
--- a/java/src/org/broadinstitute/sting/utils/sam/AlignmentUtils.java
+++ b/java/src/org/broadinstitute/sting/utils/sam/AlignmentUtils.java
@@ -285,9 +285,9 @@ public class AlignmentUtils {
                         if ( currentReadPos++ < readStartPos )
                             continue;
 
-                        if ( refIndex >= refBases.length ) {
-                            throw new IllegalStateException("When calculating mismatches, we somehow don't have enough trailing reference context for read " + read.getReadName() + " at position " + ref.getLocus());
-                        }
+                        // this is possible if reads extend beyond the contig end
+                        if ( refIndex >= refBases.length )
+                            break;
 
                         byte refChr = refBases[refIndex];
                         byte readChr = readBases[readIndex];