Table 2.
Type I error estimates using a Bayesian threshold log10θ, determined from π0 = 0.99 and varying R
| N\R (log10 θ) | 1 (1.996) | 2 (1.695) | 4 (1.394) | 5 (1.297) | 10 (0.996) | 12 (0.916) | 15 (0.820) | 20 (0.695) |
|---|---|---|---|---|---|---|---|---|
| 2,000 | 3.33 × 10–4 | 7.38 × 10–4 | 1.64 × 10–3 | 2.14 × 10–3 | 4.94 × 10–3 | 6.16 × 10–3 | 8.11 × 10–3 | 1.15 × 10–2 |
| 3,000 | 3.10 × 10–4 | 6.74 × 10–4 | 1.49 × 10–3 | 1.93 × 10–3 | 4.37 × 10–3 | 5.43 × 10–3 | 7.03 × 10–3 | 9.94 × 10–3 |
| 4,000 | 2.77 × 10–4 | 6.24 × 10–4 | 1.38 × 10–3 | 1.78 × 10–3 | 3.96 × 10–3 | 4.92 × 10–3 | 6.37 × 10–3 | 8.90 × 10–3 |
| 5,000 | 2.77 × 10–4 | 5.84 × 10–4 | 1.27 × 10–3 | 1.63 × 10–3 | 3.60 × 10–3 | 4.42 × 10–3 | 5.74 × 10–3 | 8.04 × 10–3 |
| 8,000 | 2.34 × 10–4 | 5.05 × 10–4 | 1.07 × 10–3 | 1.36 × 10–3 | 2.94 × 10–3 | 3.60 × 10–3 | 4.66 × 10–3 | 6.46 × 10–3 |
| 10,000 | 2.16 × 10–4 | 4.51 × 10–4 | 9.56 × 10–4 | 1.23 × 10–3 | 2.66 × 10–3 | 3.26 × 10–3 | 4.19 × 10–3 | 5.82 × 10–3 |
| 15,000* | 1.79 × 10–4 | 3.78 × 10–4 | 8.05 × 10–4 | 1.03 × 10–3 | 2.22 × 10–3 | 2.70 × 10–3 | 3.47 × 10–3 | 4.81 × 10–3 |
| 20,000 | 1.61 × 10–4 | 3.37 × 10–4 | 7.07 × 10–4 | 8.94 × 10–4 | 1.92 × 10–3 | 2.33 × 10–3 | 3.00 × 10–3 | 4.15 × 10–3 |
| 30,000 | 1.22 × 10–4 | 2.55 × 10–4 | 5.54 × 10–4 | 7.19 × 10–4 | 1.58 × 10–3 | 1.93 × 10–3 | 2.47 × 10–3 | 3.39 × 10–3 |
| 50,000 | 9.62 × 10–5 | 2.04 × 10–4 | 4.44 × 10–4 | 5.71 × 10–4 | 1.22 × 10–3 | 1.49 × 10–3 | 1.90 × 10–3 | 2.59 × 10–3 |
| 100,000 | 6.72 × 10–5 | 1.39 × 10–4 | 3.05 × 10–4 | 3.79 × 10–4 | 8.19 × 10–4 | 9.96 × 10–4 | 1.27 × 10–3 | 1.73 × 10–3 |
Estimates are based on 5 million independent SNPs from equal‐sized case‐control samples, each of size N. Estimates at N = 15,000 are the result of a regression at each R value of the PFP estimates against a quadratic of log10 N.