GWAS-GWAS colocalisation

[Liuyh275]

Hi，

I noticed that some studies consider a situation where two traits have been measured in two distinct datasets of unrelated individuals. However, I wanna use this method to do colocalization with two distinct traits in the same population or individuals. Does it make sense either？

Thanks!

[chr1swallace]

We looked at this for quantitative traits and concluded that while mathematically incorrect, it was fine in practice as long as you had medium- large sample sizes https://academic.oup.com/nar/article/48/6/2866/5766659 Not published, but we saw similar results for case control.

…

-- https://chr1swallace.github.io

________________________________ From: Liuyh275 ***@***.***> Sent: Sunday, June 4, 2023 6:01:24 PM To: chr1swallace/coloc ***@***.***> Cc: Subscribed ***@***.***> Subject: [chr1swallace/coloc] GWAS-GWAS colocalisation (Issue #123) Hi， I noticed that some studies consider a situation where two traits have been measured in two distinct datasets of unrelated individuals. However, I wanna use this method to do colocalization with two distinct traits in the same population or individuals. Does it make sense either？ Thanks! — Reply to this email directly, view it on GitHub<#123>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAQWR2HKDOMPRIPU7JO7BKDXJS5OJANCNFSM6AAAAAAY2BYAEE>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[anbai106]

@chr1swallace If the samples totally overlap, we can tell the direction (opposite for different beta signs and the same for the same beta signs) based on the beta coefficient sign (quantitative traits) and the effect alleles (the same for both traits), right?

[anbai106]

@chr1swallace Do you have a paper in mind (or the one from your previous link is sufficient) that we can cite if we want to publish some colocalization results using GWAS from entirely overlapped populations between two traits?

Thanks

[chr1swallace]

This paper has results from quantitative traits, but we never published the case control studies https://academic.oup.com/nar/article/48/6/2866/5766659

…

-- https://chr1swallace.github.io

________________________________ From: Junhao WEN ***@***.***> Sent: Friday, June 23, 2023 4:21:00 PM To: chr1swallace/coloc ***@***.***> Cc: Chris Wallace ***@***.***>; Mention ***@***.***> Subject: Re: [chr1swallace/coloc] GWAS-GWAS colocalisation (Issue #123) @chr1swallace<https://github.com/chr1swallace> Do you have a paper in mind (or the one from your previous link is sufficient) that we can cite if we want to publish some colocalization results using GWAS from entirely overlapped populations between two traits? Thanks — Reply to this email directly, view it on GitHub<#123 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAQWR2FTR5JG7MFOP7VGBHTXMWX5ZANCNFSM6AAAAAAY2BYAEE>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[anbai106]

@chr1swallace Thanks! My GWAS are quantitative traits for overlapped populations

[woodylc]

Hi, we perform GWAS-GWAS colocalization and Two sample Mendelian randomization of two traits. So how to interpret the results that significant result of MR while non-significant result of colocalization? Is the result reasonable?