The writer was never resetting the pointer to the end of the last non-ref VariantContext that it saw. This was fine except when it jumped to a new contig - and a lower position on that contig - where it thought that it was still part of that previous non-ref VariantContext so wouldn't emit a reference block. Therefore, ref blocks were missing from the beginnings of all chromosomes (except chr1). Added unit test to cover this case. |
||
|---|---|---|
| .. | ||
| gatk-protected | ||
| scala/test/org/broadinstitute/sting/queue/pipeline/examples | ||
| pom.xml | ||