Table 4.
Proportion of QTL detection rates across 500 simulations applying three different approaches: whole-genome nested association mapping (WGNAM), multi-locus mixed model (MLMM), and multi-parent population (MPP)
| QTL | Method | ||
|---|---|---|---|
| WGNAM | MLMM | MPP | |
| QTL1 | 0.962 | 0.968 | 0.798 |
| QTL2 | 1.000 | 1.000 | 0.984 |
| QTL3 | 0.986 | 0.808 | 0.922 |
| QTL4 | 0.646 | 0.004 | 0.532 |
| QTL5 | 0.378 | 0.106 | 0.072 |
| QTL6 | 0.620 | 0.440 | 0.104 |
| QTL7 | 0.556 | 0.240 | 0.074 |
| QTL8 | 0.058 | 0.040 | 0.014 |
| False QTL (Type I rate) | 0.600 | 0.232 | 0.092 |
| False QTL (Mean number) | 1.050 | 0.714 | 0.098 |
For the eight QTL, a QTL was considered “detected” if the position of the detected marker was within 10 cM from the true QTL position otherwise it was considered a false QTL and the false positive rate (Type I error) and the mean number of detections were provided