Skip to main content
Figure 4 | Algorithms for Molecular Biology

Figure 4

From: A novel method for identifying disease associated protein complexes based on functional similarity protein complex networks

Figure 4

Investigation of the performance of each method on different functional similarity protein complex networks extracted with different thresholds (t). (a), (c) and (e) are the performance of NBH, RWR and PRINCE for networks which are extracted from SharedGOTermComNet, respectively; (b), (d) and (f) are the performance of NBH, RWR and PRINCE for networks which are extracted from SharedPPIComNet, respectively. For RWR and PRINCE, back-probability and trade-off parameters were varied in a range of [0.1, 0.9], respectively. Vertical axis represents average AUC values over 270 disease phenotypes. “All” is the original functional similarity protein complex networks (i.e., SharedGOTermComNet in (a), (c) and (e); SharedPPIComNet in (b), (d) and (f)).

Back to article page

Algorithms for Molecular Biology

ISSN: 1748-7188