zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Don't die when a readgroup is absent from the covariates table - it could 

happen when all reads are unmapped (or have MQ0); instead, just don't alter
the quals.


git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@1394 348d0f76-0448-11de-a6fe-93d51630548a
This commit is contained in:
ebanks 2009-08-10 03:10:33 +00:00
parent 089dab00e2
commit 02f1af0743
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
java/src/org/broadinstitute/sting/gatk/walkers/recalibration/TableRecalibrationWalker.java
View File

@ -292,7 +292,16 @@ public class TableRecalibrationWalker extends ReadWalker<SAMRecord, SAMFileWrite
byte[] recalQuals = new byte[quals.length];
RecalMapping mapper = cache.get(readGroup);
if ( mapper == null ) { throw new StingException(String.format("BUG: couldn't find RecalMapping for readgroup %s", readGroup)); }
if ( mapper == null ) {
// throw new StingException(String.format("BUG: couldn't find RecalMapping for readgroup %s", readGroup));
// Edited by ebanks on 8/9/09
// Actually, there is a valid case when the mapper == null: when all reads in a RG are
// unmapped or have mapping quality == 0. In this case, we don't want to die - nor do we
// want to lose these reads - so we just don't alter the quals
for ( int cycle = 0; cycle < bases.length; cycle++ )
recalQuals[cycle] = quals[cycle];
return recalQuals;
}
recalQuals[0] = quals[0]; // can't change the first -- no dinuc
for ( int cycle = 1; cycle < bases.length; cycle++ ) { // skip first and last base, qual already set because no dinuc