Table 3.
Mean values (and SD of 20 replicates) for predictive ability (rPA), relative increment of rPA (∆ rPA), regression coefficient (Bias) and mean squared error of predictions (MSEPs) from BLUP models using different relationship matrices for grain yield (GY), stay-green (SG), plant height (PH) and flowering time (FT) prediction within and among families. The best values for each evaluation criterion are boldfaced
| Trait | Quality criterion | A-BLUP | G-BLUP | AG-BLUP | K-BLUPa |
|---|---|---|---|---|---|
| Within-family prediction | |||||
| GY | r PA | 0.299 (0.011) | 0.323 (0.013) | 0.339 (0.014) | 0.345 (0.014) |
| ∆ rPA(%) | 0 | 8.1 | 13.4 | 15.3 | |
| Bias (b) | 0.963 (0.040) | 0.924 (0.045) | 0.924 (0.043) | 0.948 (0.045) | |
| MSEP | 0.217 (0.002) | 0.214 (0.002) | 0.211 (0.002) | 0.210 (0.002) | |
| SG | r PA | 0.437 (0.010) | 0.475 (0.007) | 0.490 (0.009) | 0.494 (0.009) |
| ∆ rPA(%) | 0 | 8.6 | 12.1 | 13.0 | |
| Bias (b) | 0.988 (0.030) | 0.953 (0.024) | 0.952 (0.025) | 0.963 (0.025) | |
| MSEP | 0.514 (0.005) | 0.388 (0.004) | 0.401 (0.005) | 0.400 (0.005) | |
| PH | r PA | 0.420 (0.011) | 0.574 (0.011) | 0.579 (0.010) | 0.581 (0.009) |
| ∆ rPA(%) | 0 | 36.6 | 37.8 | 38.3 | |
| Bias (b) | 0.997 (0.032) | 0.935 (0.024) | 0.944 (0.023) | 0.949 (0.021) | |
| MSEP | 30.4 (0.4) | 24.8 (0.5) | 24.6 (0.5) | 24.4 (0.4) | |
| FT | r PA | 0.394 (0.015) | 0.489 (0.011) | 0.497 (0.014) | 0.500 (0.013) |
| ∆ rPA(%) | 0 | 24.0 | 26.1 | 26.8 | |
| Bias (b) | 0.964 (0.045) | 0.933 (0.029) | 0.937 (0.030) | 0.944 (0.027) | |
| MSEP | 0.774 (0.011) | 0.697 (0.011) | 0.693 (0.013) | 0.690 (0.013) | |
| Among-family prediction | |||||
| GY | r PA | 0.184 (0.037) | 0.230 (0.027) | 0.243 (0.028) | 0.249 (0.030) |
| ∆ rPA(%) | 0 | 25.0 | 31.9 | 35.1 | |
| Bias (b) | 0.858 (0.181) | 0.788 (0.094) | 0.828 (0.094) | 0.853 (0.104) | |
| MSEP | 0.231 (0.004) | 0.227 (0.004) | 0.225 (0.004) | 0.224 (0.004) | |
| SG | r PA | 0.365 (0.030) | 0.413 (0.022) | 0.426 (0.019) | 0.437 (0.016) |
| ∆ rPA(%) | 0 | 12.9 | 16.7 | 19.5 | |
| Bias (b) | 1.007 (0.121) | 0.925 (0.055) | 0.958 (0.061) | 0.998 (0.050) | |
| MSEP | 0.552 (0.014) | 0.417 (0.010) | 0.432 (0.009) | 0.434 (0.007) | |
| PH | r PA | 0.235 (0.044) | 0.468 (0.022) | 0.469 (0.021) | 0.477 (0.022) |
| ∆ rPA(%) | 0 | 99.1 | 99.4 | 102.9 | |
| Bias (b) | 0.791 (0.155) | 0.865 (0.051) | 0.884 (0.056) | 0.902 (0.055) | |
| MSEP | 35.0 (1.0) | 29.0 (0.9) | 28.9 (0.9) | 28.6 (0.9) | |
| FT | r PA | 0.156 (0.064) | 0.374 (0.020) | 0.352 (0.031) | 0.376 (0.023) |
| ∆ rPA(%) | 0 | 139.6 | 125.2 | 140.6 | |
| Bias (b) | 0.448 (0.196) | 0.748 (0.047) | 0.708 (0.066) | 0.747 (0.051) | |
| MSEP | 0.929 (0.042) | 0.802 (0.019) | 0.823 (0.030) | 0.801 (0.021) | |
aUsing K matrices constructed with the specific optimal weights given in Table 2