Reads were getting dropped due to bad else clause.

bigdatagenomics · Mar 4, 2017 · e533c90 · e533c90
1 parent d00e7e2
commit e533c90
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 2 deletions.
diff --git a/adam-core/src/main/scala/org/bdgenomics/adam/rdd/read/realignment/RealignIndels.scala b/adam-core/src/main/scala/org/bdgenomics/adam/rdd/read/realignment/RealignIndels.scala
@@ -412,10 +412,10 @@ private[read] class RealignIndels(
           } else {
             log.info("On " + refRegion + ", skipping realignment due to insufficient LOD improvement (" +
               lodImprovement + "for consensus " + bestConsensus)
-            readsToClean.map(_._1) ++ realignedReads
+            reads
           }
         } else {
-          realignedReads
+          reads
         }
         // return all reads that we cleaned and all reads that were initially realigned
         val endTime = System.nanoTime()

diff --git a/adam-core/src/test/scala/org/bdgenomics/adam/rdd/read/realignment/RealignIndelsSuite.scala b/adam-core/src/test/scala/org/bdgenomics/adam/rdd/read/realignment/RealignIndelsSuite.scala
@@ -541,4 +541,10 @@ class RealignIndelsSuite extends ADAMFunSuite {
     assert(realignedStRead.getCigar === "2S5M")
     assert(realignedStRead.getMismatchingPositions === "5")
   }
+
+  sparkTest("if realigning a target doesn't improve the LOD, don't drop reads") {
+    val reads = sc.loadAlignments(testFile("NA12878.1_854950_855150.sam"))
+    val realignedReads = reads.realignIndels()
+    assert(reads.rdd.count === realignedReads.rdd.count)
+  }
 }