I am running susie and I need to determine the sample size N for it. What confused me is that the GWAS summary statistics provide Nsum for each SNP and they are different. So how can I choose a single N for susie?
Here is a screenshot for my GWAS stat (Nsum column is in bold):
head(gwas_df2)
uniqID.a1a2 CHR BP A1 A2 SNP Z
rs74948764 19:45081532_C_T 19 45081532 C T rs74948764 0.9232057
rs12975006 19:45081764_T_C 19 45081764 T C rs12975006 -3.1290923
rs73048309 19:45082131_T_C 19 45082131 T C rs73048309 -0.1703050
rs73048310 19:45082284_C_G 19 45082284 C G rs73048310 -0.6837157
rs72626213 19:45082375_A_C 19 45082375 A C rs72626213 -1.7237723
rs113571662 19:45082701_T_G 19 45082701 T G rs113571662 -2.0618427
P Nsum Neff dir EAF BETA SE
rs74948764 0.355900000 17477 17477.0 ???+ 0.1024000 0.016287261 0.017642071
rs12975006 0.001753472 423393 416974.9 ?--- 0.1505540 -0.009581407 0.003062040
rs73048309 0.864770262 379838 378572.7 ??-+ 0.0201725 -0.001392141 0.008174396
rs73048310 0.494154719 377649 376387.5 ??-+ 0.0284570 -0.004739332 0.006931729
rs72626213 0.084748949 428812 422344.5 ?--- 0.0812442 -0.006864892 0.003982482
rs113571662 0.039222714 371523 370272.3 ??-- 0.0620419 -0.009932148 0.004817122
varbeta
rs74948764 3.112427e-04
rs12975006 9.376092e-06
rs73048309 6.682075e-05
rs73048310 4.804886e-05
rs72626213 1.586016e-05
rs113571662 2.320466e-05
Hi @chr1swallace,
I am running susie and I need to determine the sample size N for it. What confused me is that the GWAS summary statistics provide Nsum for each SNP and they are different. So how can I choose a single N for susie?
Here is a screenshot for my GWAS stat (Nsum column is in bold):