Protein–protein interaction network with machine learning models and multiomics data reveal potential neurodegenerative disease-related proteins

Research of protein–protein interaction in several model organisms is accumulating since the development of high-throughput experimental technologies and computational methods. The protein–protein interaction network (PPIN) is able to examine biological processes in a systematic manner and has already been used to predict potential disease-related proteins or drug targets. Based on the topological characteristics of the PPIN, we investigated the application of the random forest classification algorithm to predict proteins that may cause neurodegenerative disease, a set of pathological changes featured by protein malfunction. By integrating multiomics data, we further showed the validity of our machine learning model and narrowed down the prediction results to several hub proteins that play essential roles in the PPIN. The novel insights into neurodegeneration pathogenesis brought by this computational study can indicate promising directions for future experimental research.