Proteomic and genomic signatures of repeat instability in cancer and adjacent normal tissues

Significance Repeats in proteins, including short tandem repeats, interspersed repeats, and repetitive domains, are hotspots of evolution. However, their role in tumor evolution is essentially unknown, being limited to microsatellites, a small subclass. Here, we develop a computational technique to measure the repeat content in bulk samples, beyond microsatellites, directly from proteomic and genomic sequence raw data of tumors and their matched normal tissues. Our analyses suggest that variations in the repeat content of tumors (repeat instability) are a compensatory, transient, and adaptive mechanism in tumor evolution, which is most pronounced in early stages of tumorigenesis. Repeat instability is also manifested in normal tissues adjacent to tumors and seems to be the first route of escape from stress toward neoplasia.