The impact of protein interaction networks' characteristics on computational complex detection methods

Protein complexes of physically interacting proteins play an important role in organizing various biological processes in the cell. Therefore, correctly identifying complexes is useful for deciphering the cellular mechanisms underlying many biological processes. Since the existing high-throughput techniques have produced a large amount of protein interactions, computational methods are useful complements to the experimental methods for detecting protein complexes. In this paper, we analyze six protein interaction networks widely used for protein complex detection, and compare the performance of six classic computational methods on them in order to find the impacts of network characteristics on the performances of these complex detection methods. Furthermore, we change topological characteristics of six protein interaction networks, and verify the findings by testing performances of six methods on new ones. We hope our study will not only help recognize the relations between characteristics of protein interaction networks and computational complex detection methods, but also provide valuable insight to improve the performance in protein complex detection area.