Department of periodontics, Saveetha Institute Of Medical And Technical Science (SIMATS), Saveetha Dental College and Hospital, Saveetha University, Chennai, India.
Department of Periodontics, Ragas Dental College and Hospital, Chennai, India.
BMC Oral Health. 2024 Mar 26;24(1):385. doi: 10.1186/s12903-024-04041-y.
In recent years, the complex interplay between systemic health and oral well-being has emerged as a focal point for researchers and healthcare practitioners. Among the several important connections, the convergence of Type 2 Diabetes Mellitus (T2DM), dyslipidemia, chronic periodontitis, and peripheral blood mononuclear cells (PBMCs) is a remarkable example. These components collectively contribute to a network of interactions that extends beyond their domains, underscoring the intricate nature of human health. In the current study, bioinformatics analysis was utilized to predict the interactomic hub genes involved in type 2 diabetes mellitus (T2DM), dyslipidemia, and periodontitis and their relationships to peripheral blood mononuclear cells (PBMC) by machine learning algorithms.
Gene Expression Omnibus datasets were utilized to identify the genes linked to type 2 diabetes mellitus(T2DM), dyslipidemia, and Periodontitis (GSE156993).Gene Ontology (G.O.) Enrichr, Genemania, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were used for analysis for identification and functionalities of hub genes. The expression of hub D.E.G.s was confirmed, and an orange machine learning tool was used to predict the hub genes.
The decision tree, AdaBoost, and Random Forest had an A.U.C. of 0.982, 1.000, and 0.991 in the R.O.C. curve. The AdaBoost model showed an accuracy of (1.000). The findings imply that the AdaBoost model showed a good predictive value and may support the clinical evaluation and assist in accurately detecting periodontitis associated with T2DM and dyslipidemia. Moreover, the genes with p-value < 0.05 and A.U.C.>0.90, which showed excellent predictive value, were thus considered hub genes.
The hub genes and the D.E.G.s identified in the present study contribute immensely to the fundamentals of the molecular mechanisms occurring in the PBMC associated with the progression of periodontitis in the presence of T2DM and dyslipidemia. They may be considered potential biomarkers and offer novel therapeutic strategies for chronic inflammatory diseases.
近年来,系统性健康与口腔健康之间的复杂相互作用已成为研究人员和医疗保健从业者关注的焦点。在几个重要的关联中,2 型糖尿病(T2DM)、血脂异常、慢性牙周炎和外周血单核细胞(PBMC)的融合是一个显著的例子。这些组成部分共同构成了一个相互作用的网络,超出了它们的范围,强调了人类健康的复杂性。在本研究中,我们利用生物信息学分析通过机器学习算法预测 2 型糖尿病(T2DM)、血脂异常和牙周炎相关的互作枢纽基因及其与外周血单核细胞(PBMC)的关系。
利用基因表达综合数据集识别与 2 型糖尿病(T2DM)、血脂异常和牙周炎相关的基因(GSE156993)。使用基因本体论(G.O.)Enrichr、Genemania 和京都基因与基因组百科全书(KEGG)通路进行分析,以确定枢纽基因的鉴定和功能。通过橙色机器学习工具对枢纽基因进行预测,并验证枢纽 D.E.G.s 的表达。
决策树、AdaBoost 和随机森林在 R.O.C.曲线上的 AUC 分别为 0.982、1.000 和 0.991。AdaBoost 模型的准确率为(1.000)。结果表明,AdaBoost 模型具有良好的预测价值,可能支持临床评估并有助于准确检测与 T2DM 和血脂异常相关的牙周炎。此外,具有 p 值<0.05 和 AUC>0.90 的基因,表现出极好的预测价值,因此被认为是枢纽基因。
本研究中鉴定的枢纽基因和 D.E.G.s 为 PBMC 中与 T2DM 和血脂异常相关的牙周炎进展过程中发生的分子机制的基础做出了巨大贡献。它们可能被视为潜在的生物标志物,并为慢性炎症性疾病提供新的治疗策略。