Factor graph-aggregated heterogeneous network embedding for disease-gene association prediction

Abstract Background Exploring the relationship between disease and gene is of great significance for understanding pathogenesis developing corresponding therapeutic measures. The prediction disease-gene association by computational methods accelerates process. Results Many existing cannot fully utilize multi-dimensional biological entity to predict due multi-source heterogeneous data. This paper proposes FactorHNE, a factor graph-aggregated network embedding method prediction, which captures variety semantic relationships nodes factorization. It produces different graphs effectively aggregates relationships, using end-to-end multi-perspectives loss function optimize model. Then it good association. Conclusions Experimental verification analysis show FactorHNE has better performance scalability than models. also interpretability can be extended large-scale biomedical data analysis.