e-Driver

e-Driver is a computational method to identify cancer driver genes with a novel approach that goes beyond the conventional analysis, which treats genes as single entities impacting cancer. Recognizing that mutations in different gene regions can affect cancer relevance, e-Driver focuses on the internal distribution of somatic missense mutations within the functional regions of proteins, such as domains or intrinsically disordered regions. This approach aims to detect regions within proteins that exhibit a biased mutation rate compared to other regions of the same protein, thereby indicating evidence of positive selection and suggesting their potential role as cancer drivers.

This method represents an intermediate level of analysis that offers better statistics than analyzing individual mutation positions and higher resolution than examining entire genes. By exploiting this more granular approach, e-Driver provides insights into how specific regions within proteins may contribute to cancer development, offering a more nuanced understanding of the genetic underpinnings of cancer.

e-Driver was applied to a comprehensive cancer genome dataset from The Cancer Genome Atlas (TCGA), and its performance was compared with four other methods for identifying cancer driver genes. The results demonstrated that e-Driver can identify novel candidate cancer drivers and, thanks to its increased resolution, can provide deeper insights into the mechanisms of action of cancer driver genes identified by other methods.