Identifying potential therapeutic targets in periodontitis: a multi-omics approach integrating bulk and single-cell RNA sequencing with Mendelian randomization.

Periodontal disease (PD) is a common and complex oral health problem that affects teeth and gums, leading to tooth loss, misalignment, and infection, with significant impact. Identifying the cause and developing new treatments is crucial. This study employed Mendelian randomization (MR), single-cell RNA sequencing (scRNA-seq), and integrated transcriptomics to identify key gene signatures associated with periodontitis. Three independent datasets related to periodontitis from the GEO database were analyzed using R software for data handling and normalization. The relationships between differentially expressed genes (DEGs) and periodontitis were evaluated through differential expression, eQTL, and MR analyses. Furthermore, scRNA-seq along with GO/KEGG enrichment analyses was performed to investigate the functional roles and pathways of these genes. We identified 488 highly expressed and 252 lowly expressed genes, both playing important roles in periodontitis. Colocalization and MR analyses identified 11 significantly co-expressed genes linked to periodontitis. Afterwards, we meticulously analyzed these genes using the LASSO and random forest algorithms, which ultimately led to the discovery of four hub genes through intersection (CXCR4, ARHGDIB, PLAT, and C19orf10). These genes are involved in crucial biological processes and pathways. Single-cell analysis also identified these key gene expressions in various cell types within periodontitis samples. Our findings may offer novel insights into the molecular basis of periodontitis, with a focus on specific molecular pathways for treatment, paving the way for future research.