Table 1.
Sample size and number of selected gene family features.
| Training genes | Testing genes | Genes in both | Subjects | Metabolites | Metabolites (in pathways) | ||
|---|---|---|---|---|---|---|---|
| ZOE 2.0 | DNA (total 403k genes) | 1,355 | 1,276 | 1,214 | 289 | 503 | 149 |
| RNA (total 403k genes) | 1,805 | 1,826 | 1,667 | 287 | 503 | 149 | |
| Both (total 806k genes) | 3,158 | 3,183 | 2,948 | 287 | 503 | 149 | |
| Lloyd-Price | DNA (total 2,741k genes) | 726 | 712 | 633 | 359 | 522 | 125 |
| RNA (total 1,079k genes) | 726 | 704 | 600 | 282 | 522 | 125 | |
| Both (total 3,820k genes) | 1,424 | 1,508 | 1,211 | 269 | 522 | 125 | |
| Mallick | DNA (total 1,000k genes) | 811 | 811 | 811 | 220 | 466 | 251 (filter only) |
Testing genes: genes that can be used in the testing set. Training genes: genes that can be used in the training set. Genes in both: genes that are in both training and testing sets.