Tandem repeats are nucleotide sequence motifs that are repeated two or more times in a row at a given location in a strand of DNA. Although they are present in all individuals, they can cause disease if they expand. While tandem repeat expansions in single genes are associated with more than 40 disorders, including fragile X syndrome and Huntington’s disease, their involvement in disorders with more complex genetic underpinnings remains largely unexplored.

In the current study, SFARI Investigator Melissa Gymrek and colleagues developed new computational methods that allowed them to assess the contribution of tandem repeat mutations to autism spectrum disorder (ASD). Their work was supported in part by a SFARI Research Award.

In the first part of their study, Gymrek and colleagues developed a method (known as MonSTR) for identifying de novo mutations that lead to an expansion or contraction of short tandem repeats (defined as motifs that are 1-20 base pairs in size). This method was used to perform a genome-wide analysis of de novo tandem repeat mutations in individuals with ASD and their unaffected family members, using whole-genome sequencing data from 1,637 Simons Simplex Collection families. The analysis revealed that tandem repeat mutations are significantly more common in individuals with ASD than in unaffected siblings, with more mutations causing the repeat to expand rather than contract.

Since only a subset of these mutations is likely to contribute to ASD risk, Gymrek and colleagues next sought to prioritize the mutations on the basis of their predicted deleterious effects. To do this, they developed a method (called SISTR) to measure negative selection at tandem repeats. Their results revealed 25 mutations in individuals with ASD (all noncoding) that were likely to be pathogenic. Some of these mutations occurred in genes that have previously been associated with ASD (e.g., FOXP1 and MED13L), while others implicate potential new risk genes for the condition.

Importantly, another recent study also found that tandem repeat expansions contribute to ASD risk (Trost et al., Nature, 2020). This study used a different bioinformatics tool than the one developed by Gymrek’s team — one that was optimized for the detection of larger and more complex tandem repeat expansions; thus, the identified mutations are largely non-overlapping. The combined results of these two studies suggest that tandem repeat mutations contribute to at least four percent of ASD cases, highlighting the importance of studying this class of mutations in further detail in future studies.

The tools developed by Gymrek’s lab are available online, providing a valuable resource for other researchers interested in applying them to the characterization of tandem repeat mutations in additional data sets for ASD or other disorders.