Colocalisation status of GWAS study GCST012464 (Mullins et al.)

Community Feedback Data issue
1

We had done some colocalisation analysis of our own on the GWAS study of GCST012464 (Bipolar I disorder, Mullins N et al.) and wanted to check it against what OT had for a particular top gene but found that while there was a relatively low “QTL coloc” score as part of the L2G section, the “Evidence of colocalisation” was “No”.

We think this is likely due to this GWAS study being run through the alternate approximate colocalisation pipeline using PICS and CLPP due to a lack of summary statistics availability, so that these scores are only used for the L2G part, so “Evidence of colocalisation” will always be “No”**, but wanted to get confirmation that this was actually true for this GWAS?

I thought it best to ask because at this point it looks like summary statistics are available for this GWAS study and so was confused as to why it might have not been run with those (maybe they weren’t available at the time?).



**based on digging into the forum Q&A here:​Why does the Open Targets Genetics portal display partial L2G scores on Locus pages when there is no evidence of colocalization?). Further, searching the output data for the latest coloc runs under gs://open-targets-genetics-releases/22.02.01/v2d_coloc seemed to confirm this since there was no coloc related info for GCST012464 there.

2

The Genetics Portal integrates GWAS summary statistics mainly from the GWAS Catalog, when they are made available/submitted by the authors.
In this specific case (GCST012464, Mulllin et al) the authors have not yet submitted the bipolar disorder summary statistics to the Catalog and the team cannot make them available due to a “do not cross-post” clause on the PGC website.
I hope this answers your query!

3

I was the original poster (via the helpdesk).
Thanks for the quick response!
That clears up the matter on our side.