Closed naumenko-sa closed 5 years ago
Hi, I'm having trouble understanding the issue. Are you saying that
No, they (variants/gene and variant_impacts/gene) both are gene symbols (WASH7P), and ensembl_gene_id (Gene in terms of VEP) is absent.
can you share a VEP-annotated 1 line VCF + header so I can debug?
I see what you mean. So, given a Gene of ENSG00000223972 and a SYMBOL of DDX11L1 from VEP, vcf2db stores DDX11L1 as the gene and keeps no record of the ENSG.
Storing the SYMBOL into gene is intentional as that's what most people want. It's an oversight to not also store the ENSG id, and I'd be happy to accept a PR to fix this, but that's easily added either to the database with a look-up table or post-hoc after the query is done.
Hello @brentp !
Thanks for vcf2db tool!
I'm loading VEP-annotated vcf, i.e.
VEP annotates a variant with Gene (= Ensembl stable gene id) and Symbol https://useast.ensembl.org/info/docs/tools/vep/vep_formats.html#output
It looks like vcf2db writes Symbol CSQ record into variant_impacts/gene field rather than Gene CSQ record, and in the result there is no Ensembl Gene ID in the table for a variant.
Would it be possible to load Gene CSQ record as Gene field of variant_impacts table, and Symbol CSQ record as Symbol field of variant_impacts table?
Thanks! Sergey