Evaluating Multiple Sclerosis Risk Prediction in Families Through Comprehensive Polygenic Risk Score Analysis

9th International Congress of the Molecular Biology Association of Turkey, İzmir, Türkiye, 12 - 14 Eylül 2024, ss.1-138, (Özet Bildiri)

Background: Multiple sclerosis (MS) is a chronic, immune-mediated, inflammatory neurodegener-

ative disease with a complex inheritance pattern. In our previous study, we calculated the polygenic

risk scores (PRS) for six families using the 233 variants recommended by the International Multiple

Sclerosis Genetics Consortium (IMSGC). While some families exhibited distinguishable PRS scores

between patients and healthy relatives, others did not.

Materials and Methods: To better evaluate this, we investigated the effectiveness of PRS in predict-

ing MS susceptibility across these families and aimed to refine current PRS models, which are pre-

dominantly based on European ancestry. We employed a comprehensive variant selection approach

using Markov Chain Monte Carlo (MCMC) methods and Information Content (IC) analysis with the

GINI index to re-evaluate the 233 genetic variants.

Genetic data from seven families was imputed using BEAGLE v4, with the 1000 Genomes Project

serving as a reference, achieving high accuracy with an allelic R-squared cutoff of 0.90. Subsequent

analyses identified variants with high predictive accuracy through a combination of Euclidean dis-

tance metrics and MCMC sampling, uncovering novel variants potentially overlooked in prior ge-

nome-wide association studies (GWAS). The methodology included a detailed imputation process,

variant scoring based on effect size and zygosity, and the selection of informative subsets using

MCMC Metropolis-Hasting algorithms.

Results: The results highlight eighteen unique variants, chosen from 180 shared variants that explain

the phenotypes of all individuals within six MS families. Moreover, 27 variations among the 233 sug-

gested by IMSGC explain all individual phenotypes in these families, demonstrating the importance

of population-specific variant identification and the need to incorporate diverse genetic backgrounds

in PRS development for improved MS risk stratification.