WGCNA combined with machine learning to explore potential biomarkers and treatment strategies for acute liver failure,with experimental validation  

作　　者：Xinyan Wu Xiaomei Zheng Gang Ye 

机构地区：[1]College of Veterinary Medicine,Sichuan Agricultural University,No.211 Huimin Road,Wenjiang District,Chengdu 611130,China

出　　处：《iLIVER》2024年第4期44-55,共12页国际肝胆健康(英文)

摘　　要：Background and aims:To identify biomarkers to predict acute liver failure and investigate the mechanisms and immune-related pathways linked to its onset and progression.Methods:We analyzed gene expression differences between patients with acute liver failure(ALF)and controls in the GSE14668 dataset.Clinically relevant modules and key ALF-associated genes were identified using weighted gene co-expression network analysis(WGCNA)in conjunction with differential gene expression(DEG)analysis.Enrichment analysis was carried out and protein–protein interaction networks were constructed to understand the functions and pathways.Six potential diagnostic biomarkers were identified using machine learning algorithms.Diagnostic performance was assessed via column charts and area under the curve calculations.Single-sample gene set enrichment analysis evaluated the relationship between known marker gene sets and potential biomarker expression.We also examined diagnostic biomarker mRNA levels in ALF models in vivo and in vitro.We estimated the relative infiltration levels of 22 immune cell subpopulations in ALF samples,and explored the link between diagnostic biomarkers and infiltrating immune cells.Result:We found 352 DEGs associated with ALF.WGCNA analysis and intersecting DEGs identified 191 significant ALF-related genes.Machine learning identified HORMAD2,WNT10A,ATP6V1E2,CMBL,ARRDC4,and LPIN2 as potential diagnostic biomarkers.Cell experiments and quantitative real-time polymerase chain reaction supported the therapeutic potential of eriodictyol for ALF.Immune infiltration analysis suggested that plasma cells,CD4 memory resting and activated T cells,macrophages,and neutrophils might play roles in the progression of ALF.Conclusion:We identified HORMAD2,WNT10A,ATP6V1E2,CMBL,ARRDC4,and LPIN2,as diagnostic biomarkers for ALF and demonstrated the effectiveness of eriodictyol for treating ALF.Immune cell infiltration may play a significant role in the pathogenesis and progression of ALF.

关 键 词：Acute liver failure Machine learning Eriodictyol Potential biomarker Weighted gene co-expression network analysis Differential gene expression analysis 

分 类 号：R57[医药卫生—消化系统]