Problem making UG work with ReducedReads

sasgari

Hi,

I'm trying to use GATK release2.0 with my nine exome-seq samples, following the steps on best practice I generated per-sample, ready-to-process .bam files and then used -T ReduceReads to generate .reduced.bam files for the next step (-T UnifiedGenotyper). When using these .reduced.bam files as UG input I receive this error message: "##### ERROR MESSAGE: Somehow the requested coordinate is not covered by the read. Too many deletions?" if I take my original .bam files as input things work smoothly. Do you have any idea what causes the problem?

Thanks a lot, Samira

Here is the command lines I use:

java -Xmx4g -jar $GATKv4 \
    -R $GATK_BUNDLE/ucsc.hg19.fasta \
    -T ReduceReads \
    -L $capture_library.bed \
    -I $i.recal_s.bam \
    -o $i.reduced.bam

 java -jar $GATKv4 \
    -T HaplotypeCaller \
    -R $GATK_BUNDLE/ucsc.hg19.fasta \
    -I InputReducedBams.list \
    -L $capture_library.bed \
    --dbsnp GATK_BUNDLE/dbsnp_135.hg19.vcf \
    -o raw.snp.indel.UnifiedGenotyper.rsv.vcf

Geraldine_VdAuwera

Hi there,

When you say version 2.0, do you mean literally 2.0, or a later version, 2.x? Some bugs have been fixed since 2.0 itself, so if you're not using the latest version I encourage you to try it. Currently we're on 2.1-8.

If that's still not working we'll take a look at your problem.

sasgari

Thanks a lot! Problem solved with upgrading to the latest version, btw, for future readers: in my original msg, 2nd part, I meant to write -T Unifiedgenotyper and not -T HaplotypeCaller.