Identification of potentially oncogenic alterations from tumor-only samples reveals Fanconi anemia pathway mutations in bladder carcinomas

Cancer is caused by germline and somatic mutations, which can share biological features such as amino acid change. However, integrated germline and somatic analysis remains uncommon. We present a framework that uses machine learning to learn features of recurrent somatic mutations to (1) predict somatic variants from tumor-only samples and (2) identify somatic-like germline variants for integrated analysis of tumor-normal DNA. Using data from 1769 patients from seven cancer types (bladder, glioblastoma, low-grade glioma, lung, melanoma, stomach, and pediatric glioma), we show that “somatic-like” germline variants are enriched for autosomal-dominant cancer-predisposition genes (p
Cancer: Analytic tool reveals inherited and non-inherited tumor-causing alterations Bladder cancer cells often harbor DNA mutations that occur after tumor development, including some mutations that affect DNA repair. Raul Rabadan, Jiguang Wang, and colleagues from Columbia University in New York, USA, developed an analytic framework for identifying genetic variants, both inherited and newly arisen, that contribute to tumor development. The machine-learning tool—known as Tumor-Only Boosting Identification, or TOBI—learns what’s a cancer-associated mutation from a small training set of tumor samples and matched healthy controls. The researchers then fed the algorithm data from the tumors of 1769 patients with cancers of the bladder, brain, lungs, stomach and skin. They found that TOBI pinpointed many inherited and non-inherited mutations known to contribute to cancer growth. In bladder cancer samples, the tool also revealed a previously unknown role for inherited mutations in BRCA2 and other DNA repair genes in the so-called Fanconi anemia pathway.