Skip to main content

Advertisement

Table 5 Performance comparison with other protein complex prediction methods

From: A method for predicting protein complex in dynamic PPI networks

PPI data Methods #Complexes P R F Sn PPV Acc
Gavin data Our method 447 0.662 0.434 0.524 0.413 0.617 0.505
CSO 174 0.645 0.302 0.411 0.476 0.534 0.503
Cluster ONE 243 0.502 0.324 0.393 0.46 0.597 0.524
COACH 326 0.525 0.331 0.406 0.44 0.547 0.49
CMC 120 0.608 0.218 0.321 0.371 0.606 0.474
HUNTER 69 0.87 0.206 0.333 0.386 0.508 0.443
MCODE 66 0.727 0.142 0.238 0.277 0.513 0.377
Krogan data Our method 816 0.464 0.591 0.52 0.448 0.677 0.551
CSO 190 0.726 0.331 0.455 0.411 0.642 0.514
Cluster ONE 240 0.579 0.328 0.419 0.398 0.681 0.521
COACH 345 0.617 0.343 0.441 0.432 0.544 0.485
CMC 111 0.748 0.235 0.358 0.381 0.589 0.474
HUNTER 74 0.865 0.199 0.323 0.374 0.569 0.462
MCODE 76 0.724 0.157 0.258 0.255 0.583 0.385
MIPS data Our method 1145 0.274 0.576 0.372 0.382 0.61 0.483
CSO 192 0.495 0.289 0.365 0.286 0.568 0.403
Cluster ONE 256 0.359 0.23 0.281 0.243 0.668 0.403
COACH 448 0.301 0.289 0.295 0.336 0.311 0.323
CMC 168 0.429 0.211 0.283 0.389 0.318 0.352
HUNTER 52 0.654 0.11 0.189 0.296 0.286 0.291
MCODE 85 0.447 0.115 0.183 0.19 0.503 0.309
STRING data Our method 1240 0.324 0.586 0.417 0.836 0.404 0.581
Cluster ONE 893 0.151 0.245 0.187 0.846 0.459 0.623
COACH 1645 0.186 0.292 0.227 0.955 0.12 0.338
HUNTER 5 0.5 0.01 0.019 0.104 0.298 0.176
MCODE 393 0.092 0.09 0.091 0.675 0.242 0.405
  1. #Complexes refers to the number of predicted complexes. F: F-score, P: precision, R: recall. The highest score of each approach is shown in bold