Department of Health and Intelligent Engineering, College of Health Management, China Medical University, Shenyang, Liaoning Province 110122, PR China..
Department of Pharmacology, Shenyang Medical College, Shenyang, Liaoning Province 110001, PR China..
Ecotoxicol Environ Saf. 2024 Apr 1;274:116168. doi: 10.1016/j.ecoenv.2024.116168. Epub 2024 Mar 9.
Dilated cardiomyopathy (DCM) is a primary cause of heart failure (HF), with the incidence of HF increasing consistently in recent years. DCM pathogenesis involves a combination of inherited predisposition and environmental factors. Endocrine-disrupting chemicals (EDCs) are exogenous chemicals that interfere with endogenous hormone action and are capable of targeting various organs, including the heart. However, the impact of these disruptors on heart disease through their effects on genes remains underexplored. In this study, we aimed to explore key DCM-related genes using machine learning (ML) and the construction of a predictive model. Using the Gene Expression Omnibus (GEO) database, we screened differentially expressed genes (DEGs) and performed enrichment analyses of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways related to DCM. Through ML techniques combining maximum relevance minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) logistic regression, we identified key genes for predicting DCM (IL1RL1, SEZ6L, SFRP4, COL22A1, RNASE2, HB). Based on these key genes, 79 EDCs with the potential to affect DCM were identified, among which 4 (3,4-dichloroaniline, fenitrothion, pyrene, and isoproturon) have not been previously associated with DCM. These findings establish a novel relationship between the EDCs mediated by key genes and the development of DCM.
扩张型心肌病（Dilated cardiomyopathy，DCM）是心力衰竭（Heart Failure，HF）的主要病因，近年来 HF 的发病率持续上升。DCM 的发病机制涉及遗传易感性和环境因素的综合作用。内分泌干扰化学物质（Endocrine-disrupting chemicals，EDCs）是外源性化学物质，它们干扰内源性激素作用，并能够针对包括心脏在内的各种器官。然而，这些干扰物通过对基因的影响对心脏病的影响仍未得到充分探索。在这项研究中，我们旨在使用机器学习（Machine Learning，ML）和构建预测模型来探索与 DCM 相关的关键基因。我们使用基因表达综合数据库（Gene Expression Omnibus，GEO）筛选差异表达基因（Differentially Expressed Genes，DEGs），并对与 DCM 相关的基因本体论（Gene Ontology，GO）和京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes，KEGG）通路进行富集分析。通过结合最大相关性最小冗余度（Maximum Relevance Minimum Redundancy，mRMR）和最小绝对收缩和选择算子（Least Absolute Shrinkage and Selection Operator，LASSO）逻辑回归的 ML 技术，我们确定了用于预测 DCM 的关键基因（IL1RL1、SEZ6L、SFRP4、COL22A1、RNASE2、HB）。基于这些关键基因，我们确定了 79 种可能影响 DCM 的 EDCs，其中 4 种（3,4-二氯苯胺、fenitrothion、pyrene 和 isoproturon）以前与 DCM 无关。这些发现建立了关键基因介导的 EDCs 与 DCM 发展之间的新关系。
