[ensembl-dev] Filter variants by evidence

[Sarah Hunt]

Hi Fin,

Thanks for the suggestion; we'll look into it.

We take the evidence statuses and frequency data for different sources, 
which can lead to discrepancies in a small minority of cases. When we 
import data from dbSNP, we extract the names of the data submitters and 
use this to create the evidence status. The frequency and any genotype 
data is taken from VCF, often direct from the project. If there is more 
frequency data available direct from gnomAD say than has been submitted 
to, or clustered into a RefSeq record by dbSNP, you will see frequency 
data but no evidence. (We don't like to withhold frequency data just 
because it has not yet been processed by dbSNP.) That is the case for 
rs79866206:

https://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=rs79866206

Our live site shows rs1051736170 from dbSNP release 150, with an allele 
string of C/A (based on the Human Longevity submission). These alleles 
don't match those in the TOPMed and gnomAD VCF's so the frequency data 
is not reported. The database we will release next month uses dbSNP151 
and here rs1051736170 has an allele string of C/G:

https://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=rs1051736170

Our pre-release site is showing the frequency data from gnomAD and 
TOPMed as the alleles now match.


Best wishes,

Sarah


On 27/08/2018 06:24, Fin Swimmer wrote:
> Hello,
>
> I guess this is more a feature request. It would be fine to have the 
> possibility to show only variants that have a certain evidence 
> (gnomAD, ExAC) on the exons page.
>
> BTW: What's going on by this variants:
>
> rs79866206 - The evidence "gnomAD" is not set. But there are 
> population data for it.
> rs1051736170 - The evidences "gnomAD" and "TOPMed" are set. But there 
> are nor population data.
>
> fin swimmer
> _______________________________________________
> Dev mailing list Dev at ensembl.org
> Posting guidelines and subscribe/unsubscribe info: 
> http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20180828/61f59512/attachment.html>