Filter status mismatch between variant page and VCF in v4

[fedkon]

What I did:

Comparing this: https://gnomad.broadinstitute.org/variant/1-161518015-T-C?dataset=gnomad_r4 to the v4 exome and genome VCFs downloaded from gnomAD website, I've observed a discrepancy in the filter status. Some other variants have similar issue (not shown).

What was expected and what I've seen:

On the variant page, this variant is shown as PASS in both exomes and genomes (see screenshot).

In the genomes VCF, it's also good, but in the exomes VCF it has AS_VQSR status in FILTER field:

$ tabix gnomad.exomes.v4.0.sites.chr1.vcf.bgz chr1:161518015-161518015 | cut -f 1-7
chr1    161518015       rs382627        T       C       .       AS_VQSR
chr1    161518015       rs382627        T       G       .       PASS

$ tabix gnomad.genomes.v4.0.sites.chr1.vcf.bgz chr1:161518015-161518015 | cut -f 1-7
chr1    161518015       rs382627        T       C       .       PASS

[rileyhgrant]

Hiya, thanks for filing this issue! We should have a PR out for this relatively shortly that will fix this.

[fedkon]

Hi there! Wow, that was fast. Thanks a lot!

[rileyhgrant]

No prob, thanks for reporting it! Once the smol PR gets through review we have to re-deploy to prod, so the fix isn't live quite yet, but should be in by the end of the day.

[rileyhgrant]

This has been fixed to my eyes, and deployed to production.

https://gnomad.broadinstitute.org/variant/1-161518015-T-C?dataset=gnomad_r4