Identification of Cancer Hotspot Residues and Driver Mutations Using Machine Learning

Cancer is one of the most life-threatening diseases and mutations in selected genes are associated with tumorigenesis. Identification of driver mutations, which are responsible for the disease progression, is crucial for precision oncology. Experimentally available information on cancer-causing mutations is accumulated in several databases and the data is utilized to develop computational algorithms that predict the driver mutations. In this chapter, we survey literature to review the available databases and we summarize their key features. We also explore computational methods for identifying disease-causing mutations in specific genes and cancer types as well as more generic predictive methods. Furthermore, we discuss applications of these computational tools that are focused on large scale studies. The databases and methods for identifying driver mutations discussed in this review are helpful for the development of precision medicine and they advance the blueprint for biological and clinical endeavors.