Speeds up MarkDuplicates on queryname input by using the in memory re…

…ad-ends map. (#1411)
broadinstitute · Oct 22, 2019 · 65f7283 · 65f7283
1 parent 8598f14
commit 65f7283
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 2 deletions.
diff --git a/src/main/java/picard/sam/markduplicates/MarkDuplicates.java b/src/main/java/picard/sam/markduplicates/MarkDuplicates.java
@@ -533,7 +533,8 @@ private void buildSortedReadEndLists(final boolean useBarcodes) {
         final SamHeaderAndIterator headerAndIterator = openInputs(true);
         final SAMFileHeader.SortOrder assumedSortOrder = headerAndIterator.header.getSortOrder();
         final SAMFileHeader header = headerAndIterator.header;
-        final ReadEndsForMarkDuplicatesMap tmp = new DiskBasedReadEndsForMarkDuplicatesMap(MAX_FILE_HANDLES_FOR_READ_ENDS_MAP, diskCodec);
+        final ReadEndsForMarkDuplicatesMap tmp = assumedSortOrder == SAMFileHeader.SortOrder.queryname ?
+                new MemoryBasedReadEndsForMarkDuplicatesMap() :  new DiskBasedReadEndsForMarkDuplicatesMap(MAX_FILE_HANDLES_FOR_READ_ENDS_MAP, diskCodec);
         long index = 0;
         final ProgressLogger progress = new ProgressLogger(log, (int) 1e6, "Read");
         final CloseableIterator<SAMRecord> iterator = headerAndIterator.iterator;

diff --git a/src/main/java/picard/sam/markduplicates/util/MemoryBasedReadEndsForMarkDuplicatesMap.java b/src/main/java/picard/sam/markduplicates/util/MemoryBasedReadEndsForMarkDuplicatesMap.java
@@ -33,7 +33,7 @@
  *
  * @author alecw@broadinstitute.org
  */
-class MemoryBasedReadEndsForMarkDuplicatesMap implements ReadEndsForMarkDuplicatesMap {
+public class MemoryBasedReadEndsForMarkDuplicatesMap implements ReadEndsForMarkDuplicatesMap {
 
     /**
      * Index of this list is sequence index.  Value is map from String {read group id:read name} to ReadEnds.