Interpretation on SNP.PP.H4

[zl2860]

Hello, I am new to this amazing R package and just got confused about some outputs from the function coloc.abf().

I know that PP.H4 is the posterior probability of sharing the causal variant, but when PP. H4 is > 95%, how should I interpret the individual SNP.PP.H4 when its value is 0.01? I

I saw on the tutorial that individual SNPs with SNP.PP.H4 > 0.01 were selected as more likely to be the shared causal variants. How should I understand or interpret the value 0.01, when putting the PP.H4>0.95 and the SNP.PP.H4=0.01 together? Does it mean that given H4 is >95% true, the probability of the individual SNP to be the shared causal variant is 0.01? is this value (0.01) too low for further exploration?

Thanks!

[chr1swallace]

Does it mean that given H4 is >95% true, the probability of the individual SNP to be the shared causal variant is 0.01? Yes, almost exactly. If​ H4 is true (it can't be 95% true, it is either true or false, and PP.H4=95% suggests there is a 95% chance it is true), then this snp has an 0.01 chance to be the shared causal variant.

-- https://chr1swallace.github.io

---

From: zl2860 @.> Sent: 13 January 2023 08:46 To: chr1swallace/coloc @.> Cc: Subscribed @.***> Subject: [chr1swallace/coloc] Interpretation on SNP.PP.H4 (Issue #112)

Hello, I am new to this amazing R package and just got confused about some outputs from the function coloc.abf().

I know that PP.H4 is the posterior probability of sharing the causal variant, but when PP. H4 is > 95%, how should I interpret the individual SNP.PP.H4 when its value is 0.01? I

I saw on the tutorial that individual SNPs with SNP.PP.H4 > 0.01 were selected as more likely to be the shared causal variants. How should I understand or interpret the value 0.01, when putting the PP.H4>0.95 and the SNP.PP.H4=0.01 together? Does it mean that given H4 is >95% true, the probability of the individual SNP to be the shared causal variant is 0.01? is this value (0.01) too low for further exploration?

Thanks!

— Reply to this email directly, view it on GitHubhttps://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fchr1swallace%2Fcoloc%2Fissues%2F112&data=05%7C01%7Ccew54%40universityofcambridgecloud.onmicrosoft.com%7C90e08b7d560247f9ee4908daf542a663%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638091963845533672%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=AfIztr0hHvEbrZZdLMFCQee6js%2B1xE2jWXEcEDk2ejw%3D&reserved=0, or unsubscribehttps://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAAQWR2HYRTFJDDJSG4DIMLDWSEI53ANCNFSM6AAAAAAT2E4HWM&data=05%7C01%7Ccew54%40universityofcambridgecloud.onmicrosoft.com%7C90e08b7d560247f9ee4908daf542a663%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638091963845533672%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=DsizA4owmAWZdoDwbtc6%2FcSi2LcSHvn7SoIG6WSEYN8%3D&reserved=0. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[zl2860]

Thanks for your reply! Now I have one more question about defining the shared causal variant. Say if H4 is true, then is there any recommended cut-off value for SNP.PP.H4 to define the shared causal variant? In the tutorial, it seems that 0.01 acts as such a cut-off value, but is it too low or it is treated as ana acceptable threshold in genetic studies?