Mendelian randomized integration of GEO and eQTL data revealed genes associated with periodontitis

Objective With the development of genetics and epigenetics, there is increasing evidence that genetic changes influence susceptibility to periodontitis. In this study, Mendelian randomization was used in combination with multi-omics data to determine the causal effect of genes in periodontitis. Methods In this study, the RNA-seq dataset of periodontitis was downloaded from GEO database, and the differentially expressed genes in periodontitis were screened by differential analysis. The genes with significant causal relationship with periodontitis were screened by MR Analysis. Further, the differential expression of periodontitis genes and periodontitis-related genes were intersected to obtain the hub genes related to periodontitis. In addition, ROC curve, pearson correlation analysis and CIBERSORT algorithm were also used in this study to explore the efficiency of hub gene diagnosis and immune cell correlation analysis. Results Four hub genes (SELL, PLXDC2, C4A, and HSPG2) were obtained by differential expression analysis and MR Analysis in this study. Where HSPG2 (OR = 1.279, 95%CI: 1.087 to 1.504, P = 0.003) and PLXDC2 (OR = 1.259, 95%CI: 1.013 to 1.565, P = 0.038) were positively correlated with periodontitis. However, C4A (OR = 0.843 95%CI: 0.741 to 0.959, P = 0.009) and SELL (OR = 0.903, 95%CI: 0.825 to 0.989, P = 0.028) were negatively correlated with periodontitis. Pearson correlation analysis showed that hub genes were significantly correlated with immune cell infiltration. Conclusion This multi-omics integration study identifies a causal relationship between four hub genes and periodontitis and provides potential new therapeutic targets for clinical practice.