Multi-omics Integration with GWAS Unveils Molecular Mechanistic Insights for Type 2 Diabetes.

Background: Type 2 diabetes (T2D) is a complex metabolic disorder driven by genetic and environmental factors. While genome-wide association studies (GWAS) have identified numerous T2D-associated variants, many remain functionally uncharacterized. Integration of GWAS with molecular phenotyping offers a path to revealing biological relevance. We investigated the influence of GWAS-variants, including sub-threshold T2D-associated variants (GWAS p-value ≤ 0.0001), on gene and protein expression to assign functional relevance. Methods: Genetic variants associated with T2D in the GWAS Catalog and present in our whole-genome sequencing (WGS) data were used to perform expression quantitative trait loci (eQTL) analysis in 242 whole-blood mRNA-sequenced samples. The same variants were used to perform protein quantitative trait loci (pQTL) analysis in a set of 362 plasma samples profiled on the Olink platform. For each analysis, the datasets were randomly split into discovery and validation subsets. Associations between variants and mRNA or protein levels were tested by multiple linear regression, and only QTLs that reached a false discovery rate adjusted p-value ≤ 0.05 in the discovery dataset and replicated in the validation dataset (p ≤ 0.05) with same direction of effect were carried forward. QTL-linked mRNAs and proteins were subsequently evaluated for their relationship with T2D status to connect them with T2D pathophysiology. Results: We identified 1,291 eQTLs linked to 97 mRNAs and 1,273 pQTLs linked to 22 proteins. Among these, 10 mRNAs and 5 proteins were differentially expressed between non-diabetic and diabetic individuals. Notably, LPL, APOBR, APOM (lipid metabolism), NOTCH2, TREH (β-cell/endocrine regulation), and HLA-A, OAS3 (immune response) converged on three biological axes central to T2D pathophysiology. The directionality of molecular effects was consistent with known disease mechanisms, including insulin resistance (LPL, APOBR), β-cell stress (TREH, NOTCH2), and chronic inflammation (OAS3). Conclusions: Our findings indicate that variants falling below conventional GWAS significance thresholds can have demonstrable effects on gene expression and protein levels. This underscores the importance of prioritizing biological relevance alongside statistical significance, rather than relying solely on rigid p-value cutoffs.