基于机器学习和生物信息学分析的铜死亡相关基因在牙周炎中的作用研究

Role of cuproptosis-related genes in periodontitis based on machine learning and bioinformatics analysis

目的铜死亡是一种新的细胞程序性死亡方式,本研究旨在分析铜死亡相关基因在牙周炎组织中的表达,探讨铜死亡与牙周炎的关系。方法通过下载GEO数据库中牙周炎基因芯片数据集GSE10334,分析12个铜死亡相关基因在牙周炎组织和正常组织中的表达差异。进一步用随机森林算法(RF)和支持向量机-递归特征消除(SVM-RFE)算法确定6个关键铜死亡基因,并用其构建诊断模型且进行校准。然后,使用CIBERSORT分析关键铜死亡基因和免疫细胞之间的相关性。结果筛选出6个关键的铜死亡相关基因与牙周炎有相关性,分别为SLC31A1、DLD、LIAS、ATP7A、DLST和DLAT。基于关键基因构建诊断模型具有较高的临床有效性(AUC=0.855)。此外,关键铜死亡基因与肥大细胞及NK细胞有显著相关性。结论通过机器学习和生物信息学分析构建了6个铜死亡相关基因(SLC31A1、DLD、LIAS、ATP7A、DLST和DLAT)的诊断模型,其与牙周炎之间存在一定关联。本研究不仅为牙周炎提供了新的可能病理机制,而且为牙周炎的诊断和治疗提供了新的方向。

Objective Cuproptosis is a new type of programmed cell death.The aim of this study was to analyze the expression of cuproptosis–related genes in periodontitis tissues and to investigate the relationship between cuproptosis and periodontitis.Method The expression differences of 12 cuproptosis–related genes in periodontitis tissues and normal tissues were analysed by downloading the dataset GSE10334 for periodontitis from the GEO database.Furthermore,random forest algorithm(RF)and support vector machine recursive feature elimination(SVM-RFE)algorithm were used to identify 6 key copper death genes,and the diagnostic models were constructed and calibrated.Correlations between key cuproptosis–related genes and immune cells were then analyzed using CIBERSORT.Result Six key cuproptosis–related genes were screened for association with periodontitis,namely SLC31A1,DLD,LIAS,ATP7A,DLST and DLAT.In addition,key cuproptosis–related genes were significantly correlated with mast cells and NK cells.Conclusion We constructed diagnostic models for six cuproptosis–related genes(SLC31A1,DLD,LIAS,ATP7A,DLST and DLAT)by machine learning and bioinformatics analysis,and there was an association between them and periodontitis.This study not only provides new possible pathological mechanisms for periodontitis,but also provides new directions for the diagnosis and treatment of periodontitis.