Table 2. Linear regression model that uses significant associations in genome-wide association.
| Trait | NSM*1 | NUM*2 | r2 |
|---|---|---|---|
| Average fruit weight | 2 | 1 | 0.047 |
| Average marketable fruit weight | 1 | 1 | 0.067 |
| Soluble solids content | 3 | 3 | 0.535 |
| Pericarp colour | 3 | 3 | 0.648 |
| Style scar | 1 | 1 | 0.220 |
| Percentage of blossom-end rot fruits | 1 | 1 | 0.322 |
| Percentage of irregular-shaped fruits | 4 | 3 | 0.368 |
| Percentage of cracked fruits | 6 | 4 | 0.378 |
| Leaf length | 3 | 3 | 0.386 |
| Leaf width | 3 | 3 | 0.199 |
| Stem width | 1 | 1 | 0.021 |
| Height to the first truss | 1 | 1 | 0.020 |
| Number of flowers | 1 | 1 | 0.075 |
| Days to flowering | 3 | 3 | 0.443 |
| Number of leaves under the first truss | 2 | 1 | 0.137 |
*1NSM, Number of significant markers in the GWAS.
*2NUM, Number of markers used in the linear regression models. The variable selection was conducted using Akaike’s Information Criterion. See Methods for the details.