Table 2.
Number of genes with expression prediction models for each method after filtering by cross-validated R2 in the ALL cohort
Method | R2 > −0.1 | R2 > −0.01 | R2 > 0 | R2 > 0.01 | R2 > 0.05 | R2 > 0.1 | R2 > 0.5 |
---|---|---|---|---|---|---|---|
EN | 9,622 | 9,621 | 6,823 | 5,729 | 3,176 | 2,108 | 222 |
RF | 9,544 | 4,924 | 4,158 | 3,651 | 2,449 | 1,687 | 194 |
SVR | 9,622 | 8,929 | 5,355 | 3,772 | 2,185 | 1,454 | 141 |
KNN | 9,263 | 4,193 | 3,206 | 2,601 | 1,422 | 839 | 28 |
Total gene models before filtering; EN = 9,622, RF = 9,623, SVR = 9,623, KNN = 9,623. EN, elastic net; RF, random forest; SVR, support vector regression; KNN, K nearest neighbor.