Table 4.
Clus-HMC random forest prediction results of metabolic class based on phylogenetic profile of gene families and singleton genes.
| KEGG Metabolism class | NGF/S | AUCPRC |
|---|---|---|
| Carbohydrate metabolism (CM) | 549 | 0.478 |
| Nucleotide metabolism (NM) | 301 | 0.318 |
| Energy metabolism (EM) | 204 | 0.293 |
| Amino acid metabolism (AAM) | 292 | 0.246 |
| Metabolism of cofactors and vitamins (MCV) | 229 | 0.212 |
| Lipid metabolism (LM) | 207 | 0.151 |
| Biosynthesis of other secondary metabolites (BSM) | 106 | 0.110 |
| Glycan biosynthesis and metabolism (GBM) | 114 | 0.082 |
| Metabolism of other amino acids (MOAA) | 97 | 0.076 |
| Metabolism of terpenoids and polyketides (MTP) | 94 | 0.059 |
Reported for the 10 metabolic classes are the number of gene families/ singletons (NGF/S) and the obtained areas under the prediction-recall curve (AUCPRC). The two classes corresponding to secondary metabolism processes are highlighted in bold-face font.