Table 4.
Features used for machine learning
| Name | Abbreviation |
|---|---|
| Sequence based features | |
| Number of found short linear eukaryotic motifs in protein A | elmA |
| Number of found short linear eukaryotic motifs in protein B | elmB |
| Number of disordered binding regions in protein A | DisRegionsA |
| Number of disordered binding regions in protein B | DisRegionsB |
| Fraction of disordered Amino Acids in protein A | FracDisASA |
| Fraction of disordered Amino Acids in protein B | FracDisASB |
| Length of the longest disordered binding regions in both proteins | MaxDisLen |
| Network based features | |
| Degree | Degree |
| Betweeness of the interactions | Betweeness |
| Affinity page rank score for the interactions | APR |
| EGDV values for orbit n = {1,2,3, ..., 69} | 1, 2, 3, ..., 69 |
| Functional similarity based features | |
| Functional similarity based on cellular component GO terms | CC |
| Functional similarity based on biological process GO terms | BP |
| Functional similarity based on molecular function GO terms | MF |
| Mean of CC, BP, and MF values. | MeanSim |