Table 3.
Comparison of the number of significant markers (P ≤ 0.05) identified by multiple comparison methods including Bonferroni (Bon), false discovery rate (FDR), and positive false discovery rate (PFDR) using a simulated trait for soybean that had a heritability 60% and 20 QTLs (H60_Q20) in eight different association models eight including analysis of variance (ANOVA), general linear model (GLM), mixed linear model (MLM), compressed MLM (CMLM), enriched compressed MLM (ECMLM), settlement of MLM under progressively exclusive relationship (SUPER), multiple loci mixed linear model (MLMM), and fixed and random model circulating probability unification (FarmCPU).
| −Log10 P ≥ 3.5 | Bon | FDR | PFDR | |
|---|---|---|---|---|
| ANOVA | 2,465 | 411 | 7,204 | 9,760 |
| GLM | 520 | 38 | 1,336 | 1,966 |
| MLM | 24 | 0 | 0 | 0 |
| CMLM | 24 | 0 | 0 | 0 |
| ECMLM | 16 | 0 | 0 | 0 |
| SUPER | 229 | 5 | 327 | 521 |
| MLMM | 26 | 0 | 0 | 0 |
| FarmCPU | 19 | 4 | 10 | 10 |
For comparative purposes, a P-value threshold (–Log10 P ≥ 3.5) without any adjustment is included.