Table 1. Description of the KEGG benchmark dataset.
| KEGG Pathway | Proteins | Number of Positives | Number of Negatives |
| All pathways | 1393 | 53424 | 916104 |
| Amino Acid Metabolism | 200 | 4612 | 41421 |
| Carbohydrate Metabolism | 304 | 8266 | 37767 |
| Nucleotide Metabolism | 112 | 4993 | 41040 |
| Lipid Metabolism | 70 | 1262 | 44771 |
| Metabolism of Cofactors & Vitamins | 145 | 1974 | 44059 |
| Energy Metabolism | 135 | 3756 | 42277 |
| Glycan Biosynthesis & Metabolism | 54 | 536 | 45497 |
| Metabolism of Other Amino Acids | 67 | 1509 | 44524 |
| Signal Transduction | 129 | 8690 | 37343 |
| Membrane Transport | 253 | 17437 | 28596 |
| Cell Motility | 51 | 1962 | 44071 |
| Replication and Repair | 56 | 1222 | 44811 |
| Folding, Sorting and Degradation | 52 | 297 | 45736 |
| Translation | 191 | 2196 | 43837 |
E. coli pathways are based on the 2nd level KEGG orthology definition with 50 or more protein components only. Positive pairs are with proteins that share at least one KEGG pathway at the 3rd level of KEGG orthology definition.