HaplotypeCaller's -bamout created BAM-file not showing reads even with --force-active and --output-mode EMIT_ALL_ACTIVE_SITES
AnsweredIf you are seeing an error, please provide(REQUIRED) :
a) GATK version used:
GATK 4.2.4.0
b) Exact command used:
gatk HaplotypeCaller -R $genomeDB -I $sample_arr -O $outputdir/$outputname --tmp-dir $sampledir -ERC GVCF --native-pair-hmm-threads 2 -pairHMM AVX_LOGLESS_CACHING_OMP -bamout $bamoutputdir/$bamoutname -L $vcf -ip 1000 --force-active true --output-mode EMIT_ALL_ACTIVE_SITES
If not an error, choose a category for your question(REQUIRED):
c) Why do I see (......)?
I am trying to look for de novo mutations on trios and for IGV have thus tried to create BAM-files after the initial joint-calling and mutation filtering to inspect manually those sites. I would like to have all the reads shown for both the parents and progeny for better inspection, but these locally realigned BAM files from HC keep giving me empty frames (look below for an example). I have tried with multiple different commands for the BAM files to contain all the reads and for IGV to show them but to no avail. For example on the picture below the middle sample (MUT-42-PARENT) on the VCFs putative DNM site LG10:13,292,108 it shows its sample info as 0/0:39,0:39:99:0,101,1469 but IGV shows nothing. These reads should be informative as you can see from the information. So my question is how can I get the BAM-files to show all the information? --disable-optimizations didn't work as it gave me an error message saying "There is no variation present."
-
Hi Vee Koo,
The bamout only shows sites where variants are present. If you want to see all your reads, you should stick to the input bam.
Let me know if you have further questions.
Best,
Genevieve
-
Okay, is it then safe to assume that those empty spaces contain only reads that are similar to the reference genome?
-
I think for what you are trying to do, I wouldn't recommend using the bamout. It's only meant for specific troubleshooting but not further analysis.
Please sign in to leave a comment.
3 comments