The set of somatic mutations present in a human tumor is a record of one or more mutational processes, each of which leaves distinct "signature" of mutation types. Mutation types can be classified in various ways, the most straightforward being the base change induced by a single-base substitution (e.g., C>A, T>G, etc.). The advent of high-throughput DNA sequencing has facilitated the comprehensive, genome-wide assessment of mutation types in human tumors. This has spurred the development of methodology to tease apart the relative contribution of each mutational process by decomposing the set of all mutations into individual signatures. Many mutational signatures have known etiologies. Therefore, mutational signature inference can shed light on the causes of cancer and inform patient treatment. To date, most studies in this area have been performed on solid tumors; consequently, the application of existing methods to hematological cancers has yielded limited results. In this review, we provide an overview of the history and methodology behind mutational signature inference. Here, we present the challenges inherent in its application to hematological cancers and survey the work performed thus far. We highlight how recent research analyzing mutational signatures in normal blood cells can elucidate the beginning of a continuum of mutational processes, from normal hematopoiesis through mature hematological malignancy. Accurate characterization of mutational signatures in cancer development may aid in clinical diagnosis, prognosis, and treatment decisions.
Sartori et al. (Thu,) studied this question.