Table 3. The top three genes with the highest total feature importance from all k-mers used in each model.
Results are based on ten separate models built from 100 randomly sampled non-overlapping core gene sets.
| Protein family | Average gene length | Cumulative feature importance | PATRIC annotation |
|---|---|---|---|
| K. pneumoniae | |||
| PLF_570_00002496 | 1422.3 | 818.8 | Uncharacterized 55.8 kDa protein in cps region (ORF3) |
| PLF_570_00001044 | 1419.6 | 669.2 | Alpha,alpha-trehalose-phosphate synthase [UDP-forming] (EC 2.4.1.15) |
| PLF_570_00003328 | 1247.9 | 533.3 | N-carbamoyl-L-amino acid hydrolase (EC 3.5.1.87) |
| M. tuberculosis | |||
| PLF_1763_00001229 | 909.9 | 1157.0 | OpcA, an allosteric effector of glucose-6-phosphate dehydrogenase, actinobacterial |
| PLF_1763_00001419 | 697.0 | 649.9 | putative phosphoglycerate mutase |
| PLF_1763_00001681 | 1580.9 | 518.2 | DNA methylase |
| S. enterica | |||
| PLF_590_00006292 | 11345.6 | 1828.8 | T1SS secreted agglutinin RTX |
| PLF_590_00001168 | 1593.0 | 616.6 | Bis-ABC ATPase YbiT |
| PLF_590_00001455 | 906.0 | 571.3 | GTP-binding protein Era |
| S. aureus | |||
| PLF_1279_00000411 | 477.3 | 484.8 | 23S rRNA (pseudouridine(1915)-N(3))-methyltransferase (EC 2.1.1.177) |
| PLF_1279_00002023 | 687.0 | 250.0 | Phage-encoded chromosome degrading nuclease YokF |
| PLF_1279_00001560 | 899.8 | 232.0 | Secretory antigen precursor SsaA |