Results from coloc.abf

[Duhh56]

Hi,

Thanks for developing the package, I am using coloc.abf and have run the following function:

coloc_abf_AL <- coloc.abf(dataset1 = list(beta=coloc_df$BETA, varbeta=coloc_df$Varbeta, N=1597, type="quant", MAF=coloc_df$maf, pvalues=coloc_df$P, snp=coloc_df$ID), dataset2 = list(beta=coloc_df$slope, varbeta=coloc_df$eQTL_Varbeta, N=605, type="quant", MAF=coloc_df$maf, pvalues=coloc_df$pval_nominal, snp=coloc_df$ID))

I get the result

> print(coloc_abf_AL$summary)
       nsnps    PP.H0.abf    PP.H1.abf    PP.H2.abf    PP.H3.abf    PP.H4.abf 
3.830000e+03 6.043220e-07 8.186484e-07 4.134895e-01 5.601097e-01 2.639932e-02 

I am not sure what the result means? Is this saying all my SNPs (3830) have posterior probability of 0%? Therefore, no co-localization of GWAS signals and eQTLs was found?

[tss0222]

I also would like to ask how to interpret the result when PP.H4.abf is less than 0.95 but SNP.PP.H4 is larger than 0.95

[chr1swallace]

PP.H4.abf is the primary result. If this is large, then colocalisation is likely.

If​ you believe colocalisation exists, then SNP.PP.H4 tells you which SNP(s) are most likely to be the shared causal variant responsible. Only consider this if PP.H4.abf is large.

-- https://chr1swallace.github.io

---

From: tss0222 @.> Sent: Sunday, April 2, 2023 10:26 AM To: chr1swallace/coloc @.> Cc: Subscribed @.***> Subject: Re: [chr1swallace/coloc] Results from coloc.abf (Issue #110)

I also would like to ask how to interpret the result when PP.H4.abf is less than 0.95 but SNP.PP.H4 is larger than 0.95

— Reply to this email directly, view it on GitHubhttps://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fchr1swallace%2Fcoloc%2Fissues%2F110%23issuecomment-1493277694&data=05%7C01%7Ccew54%40universityofcambridgecloud.onmicrosoft.com%7C36834ab853294cc87f8608db335c5f7a%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638160244048988242%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=cxNaik2uTX2%2F88GgEsSMrVi3yUnZ3K1em6kLdlEhZCE%3D&reserved=0, or unsubscribehttps://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAAQWR2CT54RZRPIG4GEISH3W7FA5DANCNFSM6AAAAAAS6UF4YA&data=05%7C01%7Ccew54%40universityofcambridgecloud.onmicrosoft.com%7C36834ab853294cc87f8608db335c5f7a%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638160244048988242%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=ENzpvktxkvRojv9n%2F%2F53TsbCIE8V2zMLV4G54OQQ2nY%3D&reserved=0. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[chr1swallace]

This says there is a 0.026 probability of colocalisation. On the other hand, PP.H3.abf > 0.5, so my interpretation is that there are different causal variants for your two traits in this region.

-- https://chr1swallace.github.io

---

From: Duhh56 @.> Sent: Wednesday, December 14, 2022 3:38 PM To: chr1swallace/coloc @.> Cc: Subscribed @.***> Subject: [chr1swallace/coloc] Results from coloc.abf (Issue #110)

Hi,

Thanks for developing the package, I am using coloc.abf and have run the following function:

coloc_abf_AL <- coloc.abf(dataset1 = list(beta=coloc_df$BETA, varbeta=coloc_df$Varbeta, N=1597, type="quant", MAF=coloc_df$maf, pvalues=coloc_df$P, snp=coloc_df$ID), dataset2 = list(beta=coloc_df$slope, varbeta=coloc_df$eQTL_Varbeta, N=605, type="quant", MAF=coloc_df$maf, pvalues=coloc_df$pval_nominal, snp=coloc_df$ID))

I get the result

print(coloc_abf_AL$summary) nsnps PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf 3.830000e+03 6.043220e-07 8.186484e-07 4.134895e-01 5.601097e-01 2.639932e-02

I am not sure what the result means? Is this saying all my SNPs (3830) have posterior probability of 0%? Therefore, no co-localization of GWAS signals and eQTLs was found?

— Reply to this email directly, view it on GitHubhttps://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fchr1swallace%2Fcoloc%2Fissues%2F110&data=05%7C01%7Ccew54%40universityofcambridgecloud.onmicrosoft.com%7Cf241e8c7204847e4d07008dadde93e42%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638066291078469155%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=cTETL%2FESQmreh4MC6mI7sOf%2BMsfZPdA7mL59BqlCfUE%3D&reserved=0, or unsubscribehttps://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAAQWR2GDYACPSO2YZ6OY4B3WNHSXDANCNFSM6AAAAAAS6UF4YA&data=05%7C01%7Ccew54%40universityofcambridgecloud.onmicrosoft.com%7Cf241e8c7204847e4d07008dadde93e42%7C49a50445bdfa4b79ade3547b4f3986e9%7C1%7C0%7C638066291078469155%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=gb6GH%2F%2FdruIU%2FhW%2BAc95NcMIXDBKAB%2B6HM9PoZb43zM%3D&reserved=0. You are receiving this because you are subscribed to this thread.Message ID: @.***>