HaplotypeCaller 2.4

andrewseverin

I am getting the following error. What is the minimum read size to do assembly? 50 basepair too short?

ERROR stack trace

java.lang.IllegalStateException: Reads are too small for use in assembly. at org.broadinstitute.sting.gatk.walkers.haplotypecaller.DeBruijnAssembler.createDeBruijnGraphs(DeBruijnAssembler.java:139) at org.broadinstitute.sting.gatk.walkers.haplotypecaller.DeBruijnAssembler.runLocalAssembly(DeBruijnAssembler.java:123) at org.broadinstitute.sting.gatk.walkers.haplotypecaller.HaplotypeCaller.map(HaplotypeCaller.java:483) at org.broadinstitute.sting.gatk.walkers.haplotypecaller.HaplotypeCaller.map(HaplotypeCaller.java:132) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.processActiveRegion(TraverseActiveRegions.java:552) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.processActiveRegions(TraverseActiveRegions.java:512) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.traverse(TraverseActiveRegions.java:244) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.traverse(TraverseActiveRegions.ja at org.broadinstitute.sting.gatk.executive.LinearMicroScheduler.execute(LinearMicroScheduler.j at org.broadinstitute.sting.gatk.GenomeAnalysisEngine.execute(GenomeAnalysisEngine.java:283 at org.broadinstitute.sting.gatk.CommandLineExecutable.execute(CommandLineExecutable.java:1 at org.broadinstitute.sting.commandline.CommandLineProgram.start(CommandLineProgram.java:24 at org.broadinstitute.sting.commandline.CommandLineProgram.start(CommandLineProgram.java:15 at org.broadinstitute.sting.gatk.CommandLineGATK.main(CommandLineGATK.java:91)

ERROR ---------------------------------------------------------------------------------------

ERROR A GATK RUNTIME ERROR has occurred (version 2.4-3-g2a7af43):

ERROR

ERROR Please visit the wiki to see if this is a known problem

ERROR If not, please post the error, with stack trace, to the GATK forum

ERROR Visit our website and forum for extensive documentation and answers to

ERROR commonly asked questions http://www.broadinstitute.org/gatk

ERROR

ERROR MESSAGE: Reads are too small for use in assembly.

ERROR ---------------------------------------------------------------------------------------

::::::::::::::

Geraldine_VdAuwera

Ah right, you meant the second error. This is why we prefer to have different errors reported in different threads...

2.4-7 is the latest version. We weren't able to work on fixing that error since you didn't send us files to reproduce it. Please do so now and we will do our best to fix it asap.

Geraldine_VdAuwera

Hi Andrew,

Yes I saw, thank you. We believe that this bug is now also fixed in our development version. It will be available in the next public release. If you can't wait for the next release, you are welcome to try our nightly development builds (see downloads page).

andrewseverin

Actually the reads on this sample is 200 bases. It is from a MiSeq run and works fine with version 2.3.

Geraldine_VdAuwera

Ooh, first bug report for 2.4 already! That was fast.

We'll take a look at it. Could you upload a snippet of your bam where the error occurs? We'll probably need to reproduce this locally to figure out what's going on here.

andrewseverin

Let me check a few other things and let this run finish. I will be interested to see if the interval that I am running this on returns any snps/indels or if it exits out.

Speed has improved through by a factor of 4?

Geraldine_VdAuwera

Sure, no problem. Ryan tells me it's probably a region where lots of reads are getting clipped and resulting in very short leftovers. Before 2.4 the HC would simply have skipped through the region silently, but now it's flipping out unnecessarily. I expect if you look at the interval in question that may confirm his supposition.

andrewseverin

It would be nice if the stack trace would give me some idea as to which interval it is flipping out about.

andrewseverin

Any Idea what this one means?

java.lang.ArrayIndexOutOfBoundsException: -1 at org.broadinstitute.sting.gatk.walkers.haplotypecaller.GenotypingEngine.generateVCsFromAlignment(GenotypingEngine.java:675) at org.broadinstitute.sting.gatk.walkers.haplotypecaller.GenotypingEngine.assignGenotypeLikelihoods(GenotypingEngine.java:140) at org.broadinstitute.sting.gatk.walkers.haplotypecaller.HaplotypeCaller.map(HaplotypeCaller.java:500) at org.broadinstitute.sting.gatk.walkers.haplotypecaller.HaplotypeCaller.map(HaplotypeCaller.java:132) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.processActiveRegion(TraverseActiveRegions.java:552) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.processActiveRegions(TraverseActiveRegions.java:512) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.traverse(TraverseActiveRegions.java:244) at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions.traverse(TraverseActiveRegions.java:69) at org.broadinstitute.sting.gatk.executive.LinearMicroScheduler.execute(LinearMicroScheduler.java:100) at org.broadinstitute.sting.gatk.GenomeAnalysisEngine.execute(GenomeAnalysisEngine.java:283) at org.broadinstitute.sting.gatk.CommandLineExecutable.execute(CommandLineExecutable.java:113) at org.broadinstitute.sting.commandline.CommandLineProgram.start(CommandLineProgram.java:245) at org.broadinstitute.sting.commandline.CommandLineProgram.start(CommandLineProgram.java:152) at org.broadinstitute.sting.gatk.CommandLineGATK.main(CommandLineGATK.java:91)

ERROR ------------------------------------------------------------------------------------------

andrewseverin

How do I upload?

Geraldine_VdAuwera

Off-by-one error while taking down the variant context info. Sounds like another edge case that's not being handled properly.

To see which interval HC is choking on, use --debug -- it will print out a lot of info including the interval it's processing.

See the instructions on how to file a bug report here:
http://www.broadinstitute.org/gatk/guide/article?id=1894

getvictor

For off-by-one error, the following seems to bypass the issue. In GenotypingEngine.java:675, replace

final byte refByte = ref[refPos-1]; if( BaseUtils.isRegularBase(refByte) ) { insertionAlleles.add( Allele.create(refByte, true) ); }

with:

// Only create a ref allele if it exists and is a regular base if (refPos > 0) { final byte refByte = ref[refPos-1]; if( BaseUtils.isRegularBase(refByte) ) { insertionAlleles.add( Allele.create(refByte, true) ); } }

andrewseverin

Is this file in the jar zip file? GenotypingEngine.java or do I download a different source version of GATK? Also I was using the Haplotype caller is it still this file?

Geraldine_VdAuwera

This issue has been fixed in the latest version of GATK (2.4-7).

andrewseverin

hrmm. I am using (version 2.4-7-g5e89f01):

ERROR MESSAGE: -1

ERROR

But still get the error. Is this the most recent version?

I will work on getting an example for you guys. I did not have time in my last attempt unless you tell me this is an old version of 2.4-7

andrewseverin

I have uploaded the snippet of the problem with associated scripts that I was using. -rw-r--r-- 1 depristo wga 287606741 Mar 15 17:37 upload-minusoneError.zip

andrewseverin

Geraldine, Did you see I posted the snippet for the error above? Thanks