I am trying to run Clairvoyante using snakemake for multiple chromosome. the script looks like this:
chr_list = ['chr1', 'chr2', 'chr3', 'chr4', 'chr5', 'chr6', 'chr7', 'chr8', 'chr9', 'chr10', 'chr11', 'chr12', 'chr13', 'chr14', 'chr15', 'chr16', 'chr17', 'chr18', 'chr19', 'chr20', 'chr21', 'chr22', 'chrX', 'chrY', 'chrM']
rule clairvoyante:
"""
Identify SNPs using Clairvoyante.
"""
input:
datain="{dataset}/{sample}.{ref}.{align}.{case}.bam"
output:
dataout= ["{{dataset}}/{{sample}}.{{ref}}.{{align}}.{{case}}.clarivoyante.{}.vcf".format(i) for i in chr_list]
params:
reference=my_ref,
training=TRAINING_DATA,
clairvoyante=CLARIVOYANTE,
minCoverage=1,
message:
"Running callvarbam from clairvoyante, sample is: {wildcards.sample}"
shell:
"""
clairvoyante.py callVarBam \
--chkpnt_fn "{params.training}" \
--bam_fn "{input.datain}" \
--ref_fn "{params.reference}" \
--call_fn "{output.dataout}" \
--minCoverage "{params.minCoverage}" \
--ctgName 1 \
"""
My issue is the tool calls variant on one chromosome only per time using --ctgName parameter:
the easy way is to loop through the chr_list in shell and uses parallel or xargs to call snp on all of them ()I think this is not a convenient way) or maybe write a python function and use map to run them in parallel.
but I think there should be a better way to do that using snakemake to pass the chromosome to --ctgName for each output.
any idea?
1 answer
FYI: AFAIK the preferred support forum for snakemake is on StackOverflow.
That said, you'll need another function which is going to cat the vcf's together, and which has as input all separate vcf's like:
expand("myvariants_{{sample}}-{chromosome}.vcf", chromosome=chr_list)
You would then adapt your clairvoyante rule to write output in the form of "myvariants_{{sample}}-{chromosome}.vcf".
Yeah I agree that it would make sense to use biostars, but see also the support section in the snakemake documentation.
Log in to answer this question.