Saliva is an accessible biofluid with potential for non-invasive disease diagnostics. This study explores how genetic susceptibility to common diseases is reflected in DNA methylation (DNAm) and gene expression profiles in saliva. We constructed cis-mQTL (n = 345) and cis-eQTL (n = 277) datasets and examined correlations between DNAm and gene expression. Saliva QTLs were integrated with summary statistics from 36 genome-wide association studies (GWAS) using Summary-based Mendelian Randomization (SMR) to identify disease-associated molecular traits. We found 501 CpG sites and 24 genes as candidate biomarkers, as well as 56 eQTMs linked to conditions such as prostate cancer, squamous cell carcinoma, coronary artery disease, type 2 diabetes, and Parkinson’s disease. This work introduces a publicly available resource and suggests that saliva-based molecular signatures may capture systemic disease risk, supporting future exploration as diagnostic markers.