Skip faulty CIGARs instead of crashing

[Pranav-Garg]

There is a known bug in minimap2 (lh3/minimap2#1090) where occasionally faulty CIGARs are outputted. Currently when running on bam's containing even one such string, COBALT and GRIDSS crash. It would be better if these programs instead skipped reads with such CIGARs without crashing.

Example of a faulty CIGAR: 21M46I46D2I81M
Error message produced:

20:37:42.500 [INFO ] calculating read depths from dupsFlagged.bam
21:13:46.879 [ERROR] error: java.util.concurrent.ExecutionException: htsjdk.samtools.SAMFormatException: SAM validation error: WARNING::ADJACENT_INDEL_IN_CIGAR:Read name NOVASEQ1_131:4:1407:29143:34961, No M or N operator between pair of D operators in CIGAR
java.util.concurrent.ExecutionException: htsjdk.samtools.SAMFormatException: SAM validation error: WARNING::ADJACENT_INDEL_IN_CIGAR:Read name NOVASEQ1_131:4:1407:29143:34961, No M or N operator between pair of D operators in CIGAR
	at java.base/java.util.concurrent.FutureTask.report(FutureTask.java:122)
	at java.base/java.util.concurrent.FutureTask.get(FutureTask.java:191)
	at com.hartwig.hmftools.cobalt.count.BamReadCounter.generateDepths(BamReadCounter.java:98)
	at com.hartwig.hmftools.cobalt.CobaltApplication.run(CobaltApplication.java:86)
	at com.hartwig.hmftools.cobalt.CobaltApplication.main(CobaltApplication.java:65)
Caused by: htsjdk.samtools.SAMFormatException: SAM validation error: WARNING::ADJACENT_INDEL_IN_CIGAR:Read name NOVASEQ1_131:4:1407:29143:34961, No M or N operator between pair of D operators in CIGAR
	at htsjdk.samtools.SAMUtils.processValidationErrors(SAMUtils.java:457)
	at htsjdk.samtools.BAMRecord.getCigar(BAMRecord.java:284)
	at htsjdk.samtools.SAMRecord.isValid(SAMRecord.java:2102)
	at htsjdk.samtools.BAMFileReader$BAMFileIterator.advance(BAMFileReader.java:848)
	at htsjdk.samtools.BAMFileReader$BAMFileIterator.next(BAMFileReader.java:834)
	at htsjdk.samtools.BAMFileReader$BAMFileIterator.next(BAMFileReader.java:802)
	at htsjdk.samtools.BAMFileReader$BAMQueryFilteringIterator.advance(BAMFileReader.java:1079)
	at htsjdk.samtools.BAMFileReader$BAMQueryFilteringIterator.next(BAMFileReader.java:1069)
	at htsjdk.samtools.BAMFileReader$BAMQueryFilteringIterator.next(BAMFileReader.java:1033)
	at htsjdk.samtools.SamReader$AssertingIterator.next(SamReader.java:574)
	at htsjdk.samtools.SamReader$AssertingIterator.next(SamReader.java:553)
	at com.hartwig.hmftools.common.samtools.BamSlicer.slice(BamSlicer.java:68)
	at com.hartwig.hmftools.common.samtools.BamSlicer.slice(BamSlicer.java:52)
	at com.hartwig.hmftools.cobalt.count.BamReadCounter.sliceRegionTask(BamReadCounter.java:154)
	at com.hartwig.hmftools.cobalt.count.BamReadCounter.lambda$createFutures$1(BamReadCounter.java:141)
	at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:539)
	at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
	at java.base/java.lang.Thread.run(Thread.java:833)

[Pranav-Garg]

(I realize that GRIDSS is not developed by HMF but I thought I should document anyway)

[p-priestley]

Thank you for the info. Can you give an example of a faulty CIGAR that caused this issue?

FYI - we are working on a new SV caller which will replace GRIDSS in our pipeline and will not be updating GRIDSS further

[Pranav-Garg]

Here is such a string: 21M46I46D2I81M (I mentioned it above as well)

[charlesshale]

To turn of this strict validation in Cobalt, pass in this argument:
-bam_validation SILENT

[Pranav-Garg]

For any future readers, I also note that GRIDSS has a similar option:
--picardoptions "VALIDATION_STRINGENCY=SILENT"
And LENIENT can be used in place of SILENT.