Issue with bcftools csq phasing

Question

[shubhra.bhattacharya]

Hi all!

I want to perform phasing on a .vcf file (Human mtDNA sample). For this purpose I run bcftools csq.

My command is as follows:

 bcftools-1.6/bcftools csq -f Human_mito.fasta -g sequence.gff3 -p s sample_fltd.vcf -o trial.out

My gff file is as follows:

##sequence-region NC_012920.1 1 16569
##species https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=9606
NC_012920.1 RefSeq  region  1   16569   .   +   .   ID=NC_012920.1:1..16569;Dbxref=taxon:9606;Is_circular=true;Name=MT;country=United Kingdom: Great Britain;gbkey=Src;genome=mitochondrion;isolation-source=caucasian;mol_type=genomic DNA;note=this is the rCRS;tissue-type=placenta
NC_012920.1 RefSeq  sequence_alteration 10  10  .   +   .   ID=id-NC_012920.1:10..10;Dbxref=dbSNP:879233578;gbkey=variation;replace=T

The error it prints is:

csq.c:715 gff_id_parse] Could not parse the line, "Parent=transcript:" not present: NC_012920.1 RefSeq  exon    577 647 .   +.  ID=exon-TRNF-1;Parent=rna-TRNF;Dbxref=GeneID:4558,HGNC:HGNC:7481,MIM:590070;Note=NAR: 1455;anticodon=(pos:611..613);codons=1;gbkey=tRNA;gene=TRNF;product=tRNA-Phe

Any idea how to solve this error? -Thanks

[Kevin Blighe]

How are the contigs in your VCF named? You should also bgzip and tab-index your VCF, as this can resolve some header issues.

[shubhra.bhattacharya]

No I have not done that. I'll give it a try!

[Kevin Blighe]

Okay. Please report back when you have time.

Answer 1

[rabio]

I was able to fix this by changing the gff file to one from the ensembl ftp: http://ftp.ensembl.org/pub/