UCLA

The gap between heritability estimates from genotype data and the heritability estimates from familial studies is known as missing heritability. For example, the heritability of autism spectrum disorder (ASD) is estimated to be 50-90% in twin studies but the disease genes and susceptibility loci identified in studies on the genetic variation of ASD only explain 11% of heritability. A part of the missing heritability of ASD may reside in repetitive regions of the human genome. Which are currently understudied in whole genome-association studies (GWAS) and large-scale sequencing efforts. A subclass of these repetitive regions consist of short tandem repeats (STRs), which are repetitive DNA sequence units composed of 2-6 nucleotides. There are some 40 monogenic disorders for which expansion of STRs are reported to be the cause of disease. With the increasing availability of large whole genome sequence (WGS) data sets, there have also been advances with the detection of computational tools to detect expanded alleles at STR loci. One of these methods is called ExpansionHunter, which aims to accurately detect and measure repeat expansions longer than the short sequencing read length itself. We used ExpansionHunter to identify FMR1 and C9orf72 repeat expansions in the WGS data of 20,576 samples from the largest available dataset of families affected by ASD, the Simons Simplex Collection. We also used WGS data from 1,359 healthy control subjects in the Australian Medical Reference Genome Bank to establish the repeat expansion prevalences for FMR1 and C9orf72 independently from ASD. We observed 180 samples with repeats expanded into the premutation range of FMR1 and 50 samples with repeats in the pathogenic range of C9orf72. We also submitted 7 pedigrees with at least one family member with an expanded repeat in the FMR1 or C9orf72 gene and validated the repeat lengths by PCR. The prevalence of the FMR1 premutation was observed to be significantly higher in ASD, however, no significant difference was observed in the prevalence of the C9orf72 pathogenic repeat. Furthermore, neither the FMR1 or C9orf72 repeat expansion was observed to have an effect on ASD susceptibility. We also used the familial structure of the ASD cohort to establish inheritance patterns of the alleles at the two repeat loci. For FMR1, we observed instances of both transmission of premutation alleles and the occurrence of de novo events leading to expanded alleles in the offspring. For the c9orf72 hexanucleotide repeat locus, we only observed parental transmission for expanded alleles of the C9orf72 repeats. In an analysis specific to the C9orf72 repeat, we identified risk haplotypes previously reported to be associated with the expanded C9orf72 repeat. We observed no evidence that expanded repeats contribute to ASD disease susceptibility. However, the repeat expansion prevalences that we observe in this dataset are higher than reports in the general population, which suggest repeat expansions may be more common than previously thought. Overall, our analyses provide further insight on the prevalence of pathogenic repeat expansions in large population samples and their contribution to neurological disorders.