Table 4.
Results of Gavin PPI data and biomedical literature PPI data.
| COACH | CMC | ClusterONE | COAN | |||||
|---|---|---|---|---|---|---|---|---|
| F | σF | F | σF | F | σF | F | σF | |
| Gavin dataset | 0.406 | - | 0.321 | - | 0.418 | - | 0.404 | - |
| Gavin + random I | 0.402 | 0.003 | 0.311 | 0.009 | 0.408 | 0.012 | 0.402 | 0.004 |
| Gavin + random II | 0.398 | 0.005 | 0.298 | 0.005 | 0.389 | 0.013 | 0.401 | 0.005 |
| Gavin + random III | 0.395 | 0.005 | 0.283 | 0.012 | 0.366 | 0.013 | 0.393 | 0.008 |
| Gavin + Aimed | 0.42 | - | 0.359 | - | 0.429 | - | 0.428 | - |
| Gavin + BioInfer | 0.414 | - | 0.329 | - | 0.415 | - | 0.413 | - |
| Gavin + IEPA | 0.406 | - | 0.342 | - | 0.427 | - | 0.409 | - |
| Gavin +HPRD50 | 0.417 | - | 0.32 | - | 0.423 | - | 0.409 | - |
| Gavin + LLL | 0.41 | - | 0.337 | - | 0.411 | - | 0.419 | - |
Gavin + random I, Gavin + random II, and Gavin + random III show the results of randomly adding 1000, 2000 and 3000 interactions to the Gavin dataset, respectively. The highest F-score of each approach is shown in bold.