Table 2.
Geographic distribution models of the maize clusters inferred by Bayesian analysis. Evaluation and variable importance
| Genetic Cluster | k-fold cv | AUCtrain (mean ± SD) | AUCtest (mean ± SD) | Variable importance (percent contribution) | ||
|---|---|---|---|---|---|---|
| Var1 | Var2 | Var3 | ||||
| NEA Flours | 5 | 0.997 ± 0.001 | 0.994 ± 0.006 | BIO3 (23.24) | BIO18 (17.26) | BIO4 (13.42) |
| Tropical Lowland | 10 | 0.968 ± 0.002 | 0.938 ± 0.019 | BIO4 (40.3) | BIO3 (22.9) | BIO16 (9.49) |
| Andean | 10 | 0.992 ± 0.001 | 0.977 ± 0.018 | Alt (35.75) | BIO4 (32.38) | BIO13 (7.04) |
| HM-US | 10 | 0.968 ± 0.003 | 0.949 ± 0.038 | BIO2 (34.49) | Alt (17.89) | BIO1 (12.39) |
cv cross-validation, Alt altitude, BIO1 Annual mean temperature, BIO2 Mean diurnal range, BIO3 Isothermality; BIO4 Temperature seasonality, BIO13 Precipitation of wettest month, BIO16 Precipitation of wettest quarter, BIO18 Precipitation of warmest quarter, HM-US Highland Mexico and US