Skip to main content

Table 5 Performance comparison with other methods on DIP dataset using CYC2008 as benchmark

From: Construction of dynamic probabilistic protein interaction networks for protein complex identification

  P R F Sn PPV Acc
Our method 0.483 0.471 0.477 0.373 0.694 0.509
ClusterONE 0.428 0.331 0.373 0.364 0.665 0.493
COAN 0.486 0.438 0.461 0.435 0.555 0.491
COACH 0.364 0.468 0.41 0.544 0.38 0.455
CMC 0.595 0.287 0.387 0.399 0.566 0.475
HUNTER 0.685 0.199 0.308 0.496 0.467 0.482
MCL 0.21 0.232 0.221 0.555 0.331 0.429
  1. F: F-score, P: precision, R: recall. The highest score of each row is shown in bold