Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes

Abstract Atrial fibrillation (AF) is a predominant cardiac arrhythmia with unclear etiology. This study used bioinformatics and machine learning to explore the relationship between mitochondrial energy metabolism-related genes (MEMRGs) and immune infiltration in AF. The datasets GSE31821, GSE41177,...

Full description

Saved in:
Bibliographic Details
Main Authors: Xunjie Yang, Weng Lan, Chunyi Lin, Chunyu Zhu, Zicong Ye, Zhishi Chen, Guian Zheng
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-91047-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850238255493021696
author Xunjie Yang
Weng Lan
Chunyi Lin
Chunyu Zhu
Zicong Ye
Zhishi Chen
Guian Zheng
author_facet Xunjie Yang
Weng Lan
Chunyi Lin
Chunyu Zhu
Zicong Ye
Zhishi Chen
Guian Zheng
author_sort Xunjie Yang
collection DOAJ
description Abstract Atrial fibrillation (AF) is a predominant cardiac arrhythmia with unclear etiology. This study used bioinformatics and machine learning to explore the relationship between mitochondrial energy metabolism-related genes (MEMRGs) and immune infiltration in AF. The datasets GSE31821, GSE41177, and GSE79768 were retrieved from the Gene Expression Omnibus (GEO) database, and differential expression analysis identified 59 mitochondrial energy metabolism-related differentially expressed genes (MEMRDEGs) associated with AF. Key MEMRDEGs were selected using the least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM) methods, and a diagnostic model was developed. Immune infiltration was assessed using single-sample gene set enrichment analysis (ssGSEA) and the microenvironment cell population counter (MCPcounter). The diagnostic model, based on the key genes ACAT1, ALDH1L2, HTT, OGDH, and SLC25A3, achieved an area under the curve (AUC) of 0.903. Significant differences in immune cell composition were observed between the AF and control groups. ALDH1L2 was positively correlated with most immune cells, while SLC25A3 showed a negatively correlated with the monocytic lineage. The findings indicate that MEMRGs interact with immune responses in AF, offering insights into the potential molecular mechanisms and therapeutic targets for AF.
format Article
id doaj-art-e1fabde76a2d4897b14977cc50bc0a4a
institution OA Journals
issn 2045-2322
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e1fabde76a2d4897b14977cc50bc0a4a2025-08-20T02:01:30ZengNature PortfolioScientific Reports2045-23222025-02-0115111910.1038/s41598-025-91047-3Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genesXunjie Yang0Weng Lan1Chunyi Lin2Chunyu Zhu3Zicong Ye4Zhishi Chen5Guian Zheng6Department of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityDepartment of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityDepartment of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityDepartment of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityDepartment of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityDepartment of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityDepartment of Cardiology, Zhangzhou Affiliated Hospital of Fujian Medical UniversityAbstract Atrial fibrillation (AF) is a predominant cardiac arrhythmia with unclear etiology. This study used bioinformatics and machine learning to explore the relationship between mitochondrial energy metabolism-related genes (MEMRGs) and immune infiltration in AF. The datasets GSE31821, GSE41177, and GSE79768 were retrieved from the Gene Expression Omnibus (GEO) database, and differential expression analysis identified 59 mitochondrial energy metabolism-related differentially expressed genes (MEMRDEGs) associated with AF. Key MEMRDEGs were selected using the least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM) methods, and a diagnostic model was developed. Immune infiltration was assessed using single-sample gene set enrichment analysis (ssGSEA) and the microenvironment cell population counter (MCPcounter). The diagnostic model, based on the key genes ACAT1, ALDH1L2, HTT, OGDH, and SLC25A3, achieved an area under the curve (AUC) of 0.903. Significant differences in immune cell composition were observed between the AF and control groups. ALDH1L2 was positively correlated with most immune cells, while SLC25A3 showed a negatively correlated with the monocytic lineage. The findings indicate that MEMRGs interact with immune responses in AF, offering insights into the potential molecular mechanisms and therapeutic targets for AF.https://doi.org/10.1038/s41598-025-91047-3AFMEMRGsImmune infiltrationMachine learning
spellingShingle Xunjie Yang
Weng Lan
Chunyi Lin
Chunyu Zhu
Zicong Ye
Zhishi Chen
Guian Zheng
Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes
Scientific Reports
AF
MEMRGs
Immune infiltration
Machine learning
title Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes
title_full Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes
title_fullStr Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes
title_full_unstemmed Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes
title_short Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes
title_sort atrial fibrillation risk model based on lasso and svm algorithms and immune infiltration of key mitochondrial energy metabolism genes
topic AF
MEMRGs
Immune infiltration
Machine learning
url https://doi.org/10.1038/s41598-025-91047-3
work_keys_str_mv AT xunjieyang atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes
AT wenglan atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes
AT chunyilin atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes
AT chunyuzhu atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes
AT zicongye atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes
AT zhishichen atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes
AT guianzheng atrialfibrillationriskmodelbasedonlassoandsvmalgorithmsandimmuneinfiltrationofkeymitochondrialenergymetabolismgenes