Identification of immune-related key genes in the peripheral blood of ischaemic stroke patients using a weighted gene coexpression network analysis and machine learning

• DOI: 10.1186/s12967-022-03562-w
• Publication Date: 2022-08-12T07:03:07Z
• Authors: Peng-Fei Zheng; Lu-Zhu Chen; Peng Liu; Hong Wei Pan; Wen-Juan Fan; Zheng-Yu Liu
• Keywords: Stroke; Lasso

Abstract Background The immune system plays a vital role in the pathological process of ischaemic stroke. However, exact immune-related mechanism remains unclear. current research aimed to identify key genes associated with Methods CIBERSORT was utilized reveal cell infiltration pattern stroke patients. Meanwhile, weighted gene coexpression network analysis (WGCNA) meaningful modules significantly correlated characteristic were identified by following two machine learning methods: support vector machine-recursive feature elimination (SVM-RFE) algorithm and least absolute shrinkage selection operator (LASSO) logistic regression. Results results suggested that there decreased naive CD4 T cells, CD8 resting mast cells eosinophils an increased neutrophils, M0 macrophages activated memory Then, three significant (pink, brown cyan) be enrichment indicated 519 above mainly involved several inflammatory or signalling pathways biological processes. Eight hub ( ADM , ANXA3 CARD6 CPQ SLC22A4 UBE2S VIM ZFP36 ) revealed LASSO regression SVM-RFE algorithm. external validation combined RT‒qPCR expression levels patients these positively neutrophils negatively cells. mean AUC value 0.80, 0.87, 0.91 0.88 training set, 0.85, 0.77, 0.86 0.72 testing set 0.83, samples, respectively. Conclusions These suggest are reliable serum markers for diagnosis crucial occurrence development