Locally Adjust Networks Based on Connectivity and Semantic Similarities for Disease Module Detection.

ABSTRACT: For studying the pathogenesis of complex diseases, it is important to identify the disease modules in the system level. Since the protein-protein interaction (PPI) networks contain a number of incomplete and incorrect interactome, most existing methods often lead to many disease proteins isolating from disease modules. In this paper, we propose an effective disease module identification method IDMCSS, where the used human PPI networks are obtained by adding some potential missing interactions from existing PPI networks, as well as removing some potential incorrect interactions. In IDMCSS, a network adjustment strategy is developed to add or remove links around disease proteins based on both topological and semantic information. Next, neighboring proteins of disease proteins are prioritized

SUBMITTER: Liu J 

PROVIDER: S-EPMC8575408 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature