Topological Characterization of Cancer Driver Genes Using Reactome Super Pathways Networks.

Cancer is a complex disease caused by genetic mutations categorized into two groups: passenger and driver. Contrary to passenger, drivers mutations directly impact oncogenesis. Drivers identification is a challenge in cancer genomics, frequently supported by statistical and computational methods. These methods utilize the increasing volume of molecular data related to cancer, gene interactions networks, and pathways. Reactome recently defined 26 Super Pathways that group genes responsible for essential biological processes. Pathways networks carry topological information relative to their biological functions that emerge from genes interactions. Since some pathways are more associated with cancer than others and all have distinct structures, this work aims to characterize cancer driver genes' topological role in Super Pathways networks. We combine data from three different databases to create Super Pathways networks enriched with cancer driver genes information. Results show that Super Pathways networks have distinct topologies and particular roles for drivers. Drivers have significant differences in clustering, betweenness, and closeness centralities when compared to others genes. Attacks using random and intentional removal reveal a remarkable resilience in some Super Pathways networks. Attacks also reveal that drivers in the Programmed Cell Death pathway are more critical than hubs in keeping the network integrity. These distinguishable patterns associated with drivers can support the task of identifying and validate unknown drivers. In addition, recognize the topological role of drivers helps understand the impact mutations in these genes have on pathways structure.