Identifying Candidate Disease Genes in Multilayer Heterogeneous Biological Networks

Most of existing random walk methods to identify candidate disease genes preferentially visit highly-connected genes, while unwell-known or poorly-connected genes probably relevant to known diseases are more easily ignored or complicated to identify. Moreover, these methods access only a single gene network or an aggregated network of various gene data, leading to bias and incompleteness. Therefore, it is a pressing challenge for controlling the motion direction of random walk and for integrating multiple data sources involving different information for disease-gene identification. To this end, we first construct a multilayer network and multilayer heterogeneous genetic network. Then, we propose a topologically biased random walk with restart (BRWR) algorithm applicable to multilayer and multilayer heterogeneous networks for the identification of candidate disease genes. Experimental results show that the BRWR algorithm to identify candidate disease genes outperforms the state-of-the-art ones on different types of networks. Finally, the BRWR algorithm on multilayer heterogeneous networks is used to predict disease genes implicated in the undiagnosed neonatal progeroid syndrome. © 2024 Science Press. All rights reserved.