Re-Deriving Ensemble Mutations w/ different numpass parameter

[ghost]

Hey,

I've used ensemble mutation calling to call mutations which appear in 2/5 individual callers. I'm now wanting to go back and re-call those which appear in 3/5, 4/5, 5/5, and 1/5. Is there a way of doing this without doing the whole mutation-calling step again, but using the individual called VCF files as input? What would the yaml for this look like? I've attached the basic format of the one I'm using.

details:

• algorithm: aligner: bwa mark_duplicates: true recalibrate: gatk realign: gatk effects: vep platform: illumina variantcaller: [mutect, mutect2, freebayes, vardict, varscan] ensemble: numpass: 2 analysis: variant2 description: p145-normal files:
  • p145_N_1.fastq.gz
  • p145_N_2.fastq.gz genome_build: hg38-noalt metadata: batch: p145 phenotype: normal
• algorithm: aligner: bwa mark_duplicates: true recalibrate: gatk realign: gatk effects: vep platform: illumina variantcaller: [mutect, mutect2, freebayes, vardict, varscan] ensemble: numpass: 2 analysis: variant2 description: p145-tumor files:
  • p145_T_1.fastq.gz
  • p145_T_2.fastq.gz genome_build: hg38-noalt metadata: batch: p145 phenotype: tumor fc_date: 'Jun-17' fc_name: p145_T-N_Calling upload: dir: ./

Best wishes,

Nick

[chapmanb]

Nick; Thanks for the question. We don't have a clean way to re-run the ensemble calling from within bcbio, unless you've saved the work directory. If you have the work directory, then you can manually remove the ensemble directory, adjust the YAML settings, and then re-run producing a new ensemble callset.

Alternatively, you could run with with the minimum numpass you want (1 in this case) and then do custom post-filtering for higher sets using the CALLERS attribute in the INFO field. That's probably the easiest way to avoid a lot of manual re-running and editing.

Hope this helps.

[ghost]

Hey,

Thanks a lot, I'll give it a go.