Table 2.
Important features selected by random forests.
| Feature | #RFs | Raw score | Z-score | Correlation |
|---|---|---|---|---|
| UCC% | 10 | 2.461 | 14.909 | 0.294 |
| CAG% | 10 | 1.514 | 11.849 | -0.289 |
| GAG% | 10 | 1.652 | 11.674 | -0.305 |
| UC% | 10 | 1.988 | 11.255 | 0.281 |
| GCA% | 10 | 1.140 | 11.191 | -0.265 |
| G% | 10 | 1.672 | 9.483 | -0.266 |
| CG% | 10 | 1.235 | 8.460 | 0.133 |
| AUA% | 10 | 0.524 | 8.148 | -0.166 |
| AAG% | 9 | 0.848 | 7.851 | 0.102 |
| CUG% | 10 | 0.918 | 7.240 | -0.173 |
| U% | 9 | 1.201 | 7.170 | 0.127 |
| G/C% (first 5 bases) | 10 | 1.075 | 7.116 | -0.256 |
| AUC% | 8 | 0.632 | 6.565 | 0.201 |
| AG% | 8 | 0.910 | 6.557 | -0.277 |
| GG% | 9 | 0.831 | 6.478 | -0.190 |
| GCG% | 6 | 0.554 | 6.422 | 0.059 |
| G/C% (overall) | 5 | 0.959 | 6.414 | -0.147 |
| GGA% | 7 | 0.717 | 6.326 | -0.218 |
| AAC% | 10 | 0.409 | 6.326 | 0.162 |
| UUU% | 9 | 0.714 | 6.317 | 0.108 |
| GGC% | 9 | 0.595 | 6.304 | -0.134 |
| NT3 (C) | 5 | 0.901 | 6.258 | -0.199 |
| ACA% | 8 | 0.473 | 6.218 | 0.092 |
| UUC% | 7 | 0.542 | 5.897 | 0.125 |
| CC% | 7 | 0.704 | 5.807 | 0.004 |
| CAA% | 6 | 0.432 | 5.602 | 0.129 |