Problem with read IDs repeated in the same BAM file

[gymreklab]

We get the following error if there is more than one read pair in a given region with the same read ID:

"ERROR: Failed to extract passes filters tag from BAM alignment"

These reads were from supplementary alignments. Ignoring alignments marked as supplementary would likely solve this issue. This is similar to a previous issue when read ids were repeated across BAM files (rather than within).

CC @shubhamsaini

[tfwillems]

@gymreklab @shubhamsaini Could you send me a BAM file that on its own can reproduce this issue?

[gymreklab]

Using HipSTR version v.0.6.1

./HipSTR/HipSTR --version
HipSTR version v0.6.1

Attached files for the command below chr9.testregions.bed.zip hipstr_error_reads.bam.zip

Command:

HipSTR \
    --bams hipstr_error_reads.bam \
    --fasta /storage/resources/dbase/human/GRCh38/GRCh38_full_analysis_set_plus_decoy_hla.fa \
    --regions chr9.testregions.bed \
    --output-gls \
    --min-reads 1 \
    --def-stutter-model \
    --str-vcf test.vcf.gz

[kkapuria3]

I am getting the same error,

"ERROR: Failed to extract passes filters tag from BAM alignment"

I don't even know the reason why. I have 136 samples, how do I know which one is busted ?

[shadrinams]

Any updates / recommendations on the error? Thank you.

[douym]

We get the following error if there is more than one read pair in a given region with the same read ID:

"ERROR: Failed to extract passes filters tag from BAM alignment"

These reads were from supplementary alignments. Ignoring alignments marked as supplementary would likely solve this issue. This is similar to a previous issue when read ids were repeated across BAM files (rather than within).

CC @shubhamsaini

same problem here. To "ignore alignments marked as supplementary", do we need to change the bam file or is there a parameter for HipSTR to do that? THX