Closed apastore closed 8 years ago
Alessandro; Thanks for asking about this. It's definitely annoying but not something we're likely to tackle in bcbio as it would require doing a lot of work re-writing VCF files. Generally it's best to assume nothing about the order of samples and to reference them based on sample names. Hopefully that works for your downstream processing.
Hi Brad, This is probably non a major Issues but, I have noticed for pair Tumor/Normal pair some caller first place the Normal and other Place the Tumor sample in the vcf.
Freebayes, Mutect2, vardict have Tumor sample at first position. GATK-HC, mutest, varscan have Normal sample as position. ensemble is inconsistent depending on samples.
Thanks! Alessandro