The Study of Disease Mechanisms Based on Cascading Failure

Abstract

Studying genes closely related to diseases from the perspective of system evolution is helpful to comprehensively understand the pathogenesis of diseases. Based on the cascading failure load-capacity model, this paper gives a method to screen the key fault nodes between two control groups by using the impact of failed nodes on other nodes, called the cascading failure key nodes method (CFKNM). Taking breast cancer (BC) (GSE15852) as an example, 28 genes with significant difference between control group and BC group are screened, among which 14 genes had been confirmed to be significantly correlated with BC, and they are significantly correlated with cell growth, apoptosis and metastasis, or biomarkers and therapeutic targets for breast cancer. This predicts that the method is effective. In addition, the method predicts that C2CD2, HSD11B1 and FMO2 are significantly correlated with breast cancer, although further laboratory validation is still needed.