The Genetic Architecture of Vitamin D Deficiency among an Elderly Lebanese Middle Eastern Population: An Exome-Wide Association Study

Abstract

The Middle East region experiences a high prevalence of vitamin D deficiency, yet most genetic studies on vitamin D have focused on European populations. Furthermore, there is a lack of research on the genomic risk factors affecting elderly people, who are more susceptible to health burdens. We investigated the genetic determinants of 25-hydroxyvitamin D concentrations in elderly Lebanese individuals (n = 199) through a whole-exome-based genome-wide association study. Novel genomic loci displaying suggestive evidence of association with 25-hydroxyvitamin D levels were identified in our study, including rs141064014 in the MGAM (p-value of 4.40 × 10−6) and rs7036592 in PHF2 (p-value of 8.43 × 10−6). A meta-analysis of the Lebanese data and the largest European genome-wide association study confirmed consistency replication of numerous variants, including rs2725405 in SLC38A10 (p-value of 3.73 × 10−8). Although the polygenic risk score model derived from European populations exhibited lower performance than European estimations, it still effectively predicted vitamin D deficiency among our cohort. Our discoveries offer novel perspectives on the genetic mechanisms underlying vitamin D deficiency among elderly Middle Eastern populations, facilitating the development of personalized approaches for more effective management of vitamin D deficiency. Additionally, we demonstrated that whole-exome-based genome-wide association study is an effective method for identifying genetic components associated with phenotypes.

Genome-Wide Epistatic Network Analyses of Semantic Fluency in Older Adults

“…INT first maps the sample measurements onto a probability scale using the empirical cumulative distribution function where the observed values are replaced with fractional ranks, then transforms the observations into Z-scores using the probit function. Currently INT is one of the most popular approaches to achieve normally distributed traits (or normally distributed residuals) in genetic association studies [37][38][39].…”