Table 3.
Cumulative P-value thresholds for selecting SNPs |
Nonoverlapping P-value thresholds for selecting SNPs |
||||
---|---|---|---|---|---|
Ptraining1 threshold | Percentage of variance explained | P-value | Ptraining1 threshold | Percentage of variance explained | P-value |
p<0.00001 | 0.1 | 0.355 | 0–0.00001 | 0.1 | 0.355 |
p<0.0001 | 0 | 0.528 | 0.00001–0.0001 | 0 | 0.944 |
p<0.001 | 0.2 | 0.159 | 0.0001–0.001 | 0.2 | 0.118 |
p<0.01 | 0.1 | 0.524 | 0.001–0.01 | 0.1 | 0.977 |
p<0.1 | 0.1 | 0.043 | 0.01–0.1 | 0.1 | 0.013 |
p<0.2 | 0.1 | 0.002 | 0.1–0.2 | 0.1 | 0.003 |
p<0.3 | 0.2 | 0.003 | 0.2–0.3 | 0.1 | 0.832 |
p<0.4 | 0.1 | 0.002 | 0.3–0.4 | 0.1 | 0.269 |
p<0.5 | 0.1 | 0.004 | 0.4–0.5 | 0 | 0.996 |
NHS, Nurses' Health Study; SNP, single-nucleotide polymorphism.
Training set: remaining three NHS substudies except the testing sample.