Control signature of nodes in biological networks

Controllability of underlying biological networks is studied to help understand biological systems. Although structural properties of driver and critical nodes have been characterized for identifying novel disease genes and potential drug targets, the hypothesis that critical genes interpose crucial functional outcomes of cells has been consistently supported, and ordinary genes have been largely ignored because of their lower enrichment in the context of diseases. Here, we characterize ordinary genes from four different control perspectives, and present a 4-tuple index control signature to precisely describe their control properties. Importantly, we observed a small-scale topologically ordinary genes that make extraordinary contributions to disease etiology, which we called extraordinary genes (EGs). Studying the contribution of EGs in the omnigenic genetic architecture, we verified their high regulatory effects on core genes of diseases by analyzing expression quantitative trait loci, drug targets, and heritability data. We proposed a novel graph pattern of disease in which EGs converge the effects of peripheral signals into core, and improve the heritability enrichment of disease neighborhoods. Together, our results show, for the first time, the ability to characterize extraordinary disease-related genes from ordinary nodes, and suggest a more comprehensive disease-perturbed neighborhoods that further explains the omnigenic model.