OBJECTIVE: Efficient and comprehensive prioritization of cancer driver genes across individual patients, cancer cohorts, and pan-cancer is crucial for advancing cancer diagnosis and treatment. The existing methods are effective, but they seem to have reached a plateau in accuracy enhancement and lack broad-scale joint analysis, flexibility in adapting to cancer and interpretability. METHODS: Here, we introduce GenMorw, a heterogeneous network framework that discovers a novel association score between patients and their mutated genes, enabling the estimation of the likelihood of the mutated genes acting as drivers in patients. GenMorw flexibly integrates or fully utilize collected mutation, gene/miRNA expression, methylation data and PPI networks to classify patient groups based on data-specific characteristics and identify potential drivers at the individual, cancer and pan-cancer levels. RESULTS: GenMorw outperforms existing algorithms with an average cohort AUC improvement of 17.66% and higher overall accuracy by a cumulative ranking strategy in patient-gene heterogeneous networks. Except for AUC evaluation, other various comparative strategies consistently demonstrate the superior performance of GenMorw across multiple cancers, outperforming other algorithms. Some uniquely predicted genes, such as ANK3, CENPF, and COL7A1, which are absent from standard databases and not identified by other methods, were validated as highly cancer-related through literature review and survival analysis. Based on GenMorw-derived heterogeneous networks, the strongly connected components and cliques, which are extracted from them, capture most of the predicted or known driver genes to help predict driver genes. CONCLUSION: We conclude that GenMorw, with its novel gene-patient score mechanism, offers a significant advance in cancer driver gene discovery by capturing both population-wide and patient-specific network signals, thereby improving predictive power and enabling deeper insights into cancer heterogeneity.

MOTIVATION: Cancer is caused by the accumulation of somatic mutations in multiple pathways, in which driver mutations are typically of the properties of high coverage and high exclusivity in patients. Identifying cancer driver genes has a pivotal role in understanding the mechanisms of oncogenesis and treatment. RESULTS: Here, we introduced MaxCLK, an algorithm for identifying cancer driver genes, which was developed by an integrated analysis of somatic mutation data and protein-protein interaction (PPI) networks and further improved by an information entropy index. Tested on pancancer and single cancers, MaxCLK outperformed other existing methods with higher accuracy. About pancancer, we predicted 154 driver genes and 787 driver modules. The analysis of co-occurrence and exclusivity between modules and pathways reveals the correlation of their combinations. Overall, our study has deepened the understanding of driver mechanism in PPI topology and found novel driver genes. AVAILABILITY AND IMPLEMENTATION: The source codes for MaxCLK are freely available at https://github.com/ShandongUniversityMasterMa/MaxCLK-main.