Genome-wide meta-analysis of short-tandem repeats for Parkinson's disease risk using genotype imputation.

Idiopathic Parkinson's disease is determined by a combination of genetic and environmental factors. Recently, the first genome-wide association study on short-tandem repeats in Parkinson's disease reported on eight suggestive short-tandem repeat-based risk loci ( α = 5.3 × 10 -6 ), of which four were novel, i.e. they had not been implicated in Parkinson's disease risk by genome-wide association analyses of single-nucleotide polymorphisms before. Here, we tested these eight candidate short-tandem repeats in a large, independent Parkinson's disease case-control dataset ( n = 4757). Furthermore, we combined the results from both studies by meta-analysis resulting in the largest Parkinson's disease genome-wide association study of short-tandem repeats to date ( n = 43 844). Lastly, we investigated whether leading short-tandem repeat risk variants exert functional effects on gene expression regulation based on methylation quantitative trait locus data in human 'post-mortem' brain ( n = 142). None of the eight previously reported short-tandem repeats were significantly associated with Parkinson's disease in our independent dataset after multiple testing correction ( α = 6.25 × 10 -3 ). However, we observed modest support for short-tandem repeats near CCAR2 and NCOR1 in the updated meta-analyses of all available data. While the genome-wide meta-analysis did not reveal additional study-wide significant ( α = 6.3 × 10 -7 ) short-tandem repeat signals, we identified seven novel suggestive Parkinson's disease short-tandem repeat risk loci ( α = 5.3 × 10 -6 ). Of these, especially a short-tandem repeat near MEIOSIN showed consistent evidence for association across datasets. CCAR2 , NCOR1 and one novel suggestive locus identified here ( LINC01012 ) emerged from colocalization analyses showing evidence for a shared causal short-tandem repeat variant affecting both Parkinson's disease risk and cis DNA methylation in brain. Larger studies, ideally using short-tandem repeats called from whole-sequencing data, are needed to more fully investigate their role in Parkinson's disease.