Computational approaches for inferring tumor evolution from single-cell genomic data

Genomic heterogeneity in tumors results from mutations and selection of high-fitness single cells, the operational components of evolution. Precise knowledge about mutational heterogeneity and evolutionary trajectory of a tumor can provide useful insights into predicting cancer progression and designing personalized treatment. The rapidly advancing field of single-cell genomics provides an opportunity to study tumor heterogeneity and evolution at the ultimate level of resolution. In this review, we present an overview of the state-of-the-art single-cell DNA sequencing methods, technical errors that are inherent in the resulting large-scale datasets, and computational methods to overcome these errors. Finally, we discuss the computational and mathematical approaches for understanding intratumor heterogeneity and cancer evolution at the resolution of a single cell.