Table 2.

Replicability analysis for FDR control for the study of ref. 17 on GWAS of T2D

Chr.	Position	p.primary	p1	p2	p_meta	r value
7	27,953,796	1.55e-04	8.07e-05	1.34e-07	4.96e-14	0.0055
10	12,368,016	4.21e-04	5.40e-05	1.49e-04	1.21e-10	0.0055
12	69,949,369	1.80e-05	9.83e-03	4.35e-05	1.11e-09	0.1490
2	43,644,474	1.83e-04	1.62e-03	9.22e-05	1.12e-09	0.0441
3	64,686,944	5.44e-04	1.02e-04	3.47e-03	1.17e-08	0.0254
1	120,230,001	1.14e-04	2.89e-03	1.95e-03	4.10e-08	0.0604
12	53,385,263	3.18e-05	3.11e-03	8.81e-03	1.79e-07	0.0604
3	12,252,845	1.05e-05	4.50e-03	1.22e-02	1.97e-07	0.0765
1	120,149,926	1.35e-03	1.17e-03	7.84e-03	4.04e-07	0.0431
6	43,919,740	5.41e-05	1.46e-03	9.49e-02	4.03e-06	0.2090
2	60,581,582	3.38e-05	1.38e-03	6.54e-01	1.02e-04	1.0000

The number of SNPs in the first follow-up study was 68, and 11 were followed up to the second follow-up study. For these 11 SNPs, the positions (columns 1 and 2), the primary study P values and first and second follow-up studies P values (columns 3–5), the metaanalysis P values from all three studies (column 6), and the r values quantifying the evidence of replicability from the first to the second follow-up study (column 7) are shown. The lower bound for $f_{00}$ was $l_{00}=0$ for the r-value computation, because the set of SNPs in the first follow-up study is already believed to be associated with T2D.