Reconstructing genotypes in trios from a Sardinian sample for fine mapping of SNPs in MYO1D-ASIC2 genes using the Knockoff filter: a candidate region for Multiple Sclerosis susceptibility

Introduction. Multiple Sclerosis (MS) is a progressive autoimmune disease that damages the myelin, the protective covering of the central nervous system (CNS). This damage leads to a wide range of physical and cognitive impairments. While the primary symptoms are neurological, MS can also affect multiple organs, creating both health and social challenges. MS is considered a complex trait disease, as it results from the interplay of various factors, including genetic, biological, and environmental influences, along with their interactions. Aim. Our work focuses on a fine mapping method, a statistical technique used to narrow down a selected region of chromosome 17 (30820506:32483270 bp). In fine mapping settings, typically a statistical method that takes advantage of the patterns of linkage disequilibrium (LD) between genetic variants to find the strongest significant signals among all the genetic variants in the considered window. The region under investigation is near the ASIC2 gene and has already been targeted for association of microRNA with MS, hence it is likely to contain associated variants. The study was conducted by using a sample of families from Sardinia, a genetically isolated population, with high prevalence of MS, making it suitable for this kind of investigation. Indeed, Sardinian MS prevalence is estimated to reach 330 per 100,000 inhabitants (95%CI: [321, 338]) compared to mainland Italy (176 per 100,000) and the Northern European one (100-190 per 1000). Methods. The employed sample consisted of 157 trios (father-mother-child for a total of 471), on which a multi-step analysis was applied to localise MS-associated variants at multiple resolutions using the knockoff filter method. A total of 2537 single nucleotide polymorphisms (SNPs) have been imputed and filtered according to standard quality controls, and haplotypes were reconstructed for subsequent analyses. The analysis continued with the augmentation of the genotype matrix, by creating knockoff variants, which have the same correlation structure as the original variants but are independent of the disease conditional on the observed genotypes. Afterwards, filter was applied by testing the conditional associations of genetic segments of decreasing width along the targeted window using a family-based association test (FBAT). Results. Scanning the imputed variants, two distinct regions of chromosome 17 exhibited a noteworthy effect on MS. The variant rs756787, at 3’UTR MYO1D, was found to be associated with an increased risk of MS (OR=1.57, 95%CI: [1.07, 2.29]), while rs56175840, in intronic region of ASIC2, was found associated with a decreased risk of MS (OR=0.17, 95%CI: [0.04, 0.74]). The alternative allele of variant rs756787 (involving CDK5R1 and the 3’UTR of MYO1D) had a false discovery rate (FDR) of 0.21. By studying pathways and LD-traits in GWAS catalogues, this study allowed us to prioritise the rs756787 variant which is also linked to neurodegeneration and EBV pathway. This comprehensive analysis enhances our understanding of the genetic architecture of MS, offering insights for further investigation on genetic MS susceptibility.