Test-drive the GATK tools and Best Practices pipelines on Terra
Check out this blog post to learn how you can get started with GATK and try out the pipelines in preconfigured workspaces (with a user-friendly interface!) without having to install anything.
calling SNPs from RNAseq data without reference SNPset
Hi, i read the entry about how to do BQSR without a knownSNP file and have some uncertainties how to apply it to RNAseq data.
I am actually calling SNPs from RNA-seq data on a draft genome of a non-model organism and were wondering what best practice might be to do so (must sound like a nightmare to you working with human data ).
I can think of following workflow for each of the RNA-seq samples:
1. best practice SNPcalling for RNA reads with HaplotypeCaller 2. filter variants for "high quality" (-window 25 -cluster 3 --filterExpression "MQ < 30.0" --filterExpression "QD < 2.0" --filterExpression "DP < 5" 3. select for PASS SNPs and biallelic SNPs (as sample is diploid) 4. use the selected SNPs as knownSNPs to do BQSR 5. run Haplotypecaller again on the recalibrated bam 6. go nuts with the gained vcf file... =)
Should i include heterozygous SNPs to generate the BQSR-recalibration file?
Would you agree on that workflow or alter the filters ( i know filtering for depth is not a good thing to do but for RNA-seq i think its good to have some minimal coverage of a site)
Comments and recommendations are very welcome,