Network resilience of the human interactome to pan-cancer mutations reveals conserved pathway vulnerabilities

Interactomes encode interdependencies among molecular components in the cell. Gene mutations are seen as node failures that affect the biological information flow and possibly lead to diseases. Using an information-based resilience measure, we study how the overall topology of the human interactome is affected by tumor-related (pan-cancer) mutations and linked to functionality, aiming to reveal common mechanisms. We found that cancer-associated mutations disrupt the interactome significantly more than random failures, leading to faster network fragmentation. The impact of the single gene on the resilience shows two well-separated groups. Most mutations are consistent with random occurrence, but a small subset (664), undetectable through standard metrics, plays a pivotal role in maintaining network integrity and, consequently, cellular functionality. This allows the identification and ranking of crucial genes. Enrichment analysis shows the involvement of those genes in apoptotic and other relevant biological processes that are not enriched in the larger group and are conserved in cancers, supporting the concept of cancer as a disease emerging from system-level dysregulation.