bcftools consensus in V1.17 not applying any variants to reference

[rchapman2000]

Hello,

I found an issue with bcftools consenus that is very similar to #1918.

For some context, I am creating a new assembly pipeline for my laboratory, and I use bcftools consensus to apply variants to a reference genome. As I was checking my assembly, I noticed that it was identical to the reference sequence, but my VCF files contained variants that should have been applied (See example below):

#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	unknown
NC_063383.1	4952	.	C	T	0.512731	.	AB=0;ABP=0;AC=0;AF=0;AN=2;AO=1;CIGAR=1X;DP=1;DPB=1;DPRA=0;EPP=5.18177;EPPR=0;GTI=0;LEN=1;MEANALT=1;MQM=60;MQMR=0;NS=1;NUMALT=1;ODDS=2.07944;PAIRED=0;PAIREDR=0;PAO=0;PQA=0;PQR=0;PRO=0;QA=0;QR=0;RO=0;RPL=0;RPP=5.18177;RPPR=0;RPR=1;RUN=1;SAF=1;SAP=5.18177;SAR=0;SRF=0;SRP=0;SRR=0;TYPE=snp	GT:DP:AD:RO:QR:AO:QA:GL	0/0:1:0,1:0:0:1:0:0,-0.30103,-4.34294e-09

However, BCFTools consensus runs and outputs a message as if all of the variants were applied:

bcftools consensus -f NC_063383.fasta sample-variants.vcf.gz > corrected.fasta
Applied 53 variants

I was using BCFtools V1.17, and upon downgrading to V1.16, the issue seems to have been resolved.

Note as well, I installed BCFTools via conda.

Thank you!

[pd3]

There was a change in the default behavior, motivated by #1828. Newly the program considers information in FORMAT/GT first, in case it is a homozygous reference 0/0.

If you want to apply all variants from the ALT column regardless of the genotype, run with -s -

-s, --samples LIST             Comma-separated list of samples to include or
                                     "-" to ignore samples and use REF,ALT