Cancer genome studies have contributed to the analysis and discovery of somatic mutations that drive cancer growth. However, studying the genetic makeup of a tumor when it is already fully developed limits our ability to uncover how and which somatic mutations accumulate in normal tissues in the stages preceding cancer initiation. To address this challenge, recent studies performed deep sequencing in a limited number of tissue types and a small number of individuals, identifying a large number of microscopic clones carrying somatic mutations, some in known cancer genes. These findings emphasize the need to uncover the genomic events that occur in all normal tissues. Although efforts have begun to collect and analyze DNA from normal tissues, we still lack a comprehensive catalog of genetic events and clonal properties across a large number of tissues and individuals. By analyzing the information-rich content in RNA now available from recent advances in RNA sequencing methods, we may be able to substantially expand the scope and scale of these studies.

Some mutations found in the DNA can be detected in the corresponding RNA, depending on the mutation allele fraction and sequence coverage. We therefore hypothesized that a careful analysis of RNA sequences from normal bulk tissues could uncover somatic mutations reflecting macroscopic clones within the samples. In this work, we used the large collection of RNA sequences from the Genotype–Tissue Expression (GTEx) project, representing more than 6700 samples from ~500 individuals, spanning across 29 different normal tissues.

We developed a new method, called RNA-MuTect, to identify somatic mutations using a tissue-derived RNA sample and its matched-normal DNA. We validated RNA-MuTect on both tumor-adjacent and cancer samples from The Cancer Genome Atlas (TCGA), wherein DNA and RNA were coextracted from the same samples. Focusing on mutations contained within sufficiently covered sequences, RNA-MuTect achieved high sensitivity and precision, enabling the discovery of most driver events and mutational processes from TCGA tumor RNA data. When applied to the GTEx dataset of normal tissues, multiple somatic mutations were detected in almost all individuals and tissues studied here, including in known cancer genes. The three tissues with the largest number of somatic mutations were sun-exposed skin, esophagus mucosa, and lung; this finding suggests that environmental exposure can promote somatic mosaicism. Both the individuals’ age and tissue-specific proliferation rate were found to be associated with the number of detected mutations. A dN/dS (ratio of nonsynonymous to synonymous substitutions) analysis suggested that some of the mutations identified in cancer genes may confer a selective advantage. In addition, allelic imbalance events at the chromosome arm level were detected in normal tissues.

Genetic clones carrying somatic mutations are detected across normal tissues to different extents, and these differences depend on factors such as the tissue’s exposure to environmental mutagens, natural architecture, proliferation rate, and the microenvironment. Some of these clones may be the result of genetic drift. Others, however, may develop as a result of positive selection driven by certain somatic events, thus potentially representing the earliest stages of tumorigenesis. Higher-resolution studies of normal tissues and precancerous lesions are required if we are to advance our understanding of both aging and early cancer development.