A better understanding of the immunosuppression genes (ISGs) related to ischemic stroke and further genetically targeting these genes might be beneficial for reducing the risk the stroke-induced immunosuppression resulted complication. 

The datasets regarding ischemic stroke were downloaded from GEO database, and immunosuppression genes were obtained from three databases (i.e., DisGeNET, HisgAtlas, and Drugbank). The CIBERSORT algorithm was used to calculate the mean proportion of 22 immune-infiltrating cells in stroke samples. The differential expression analysis was performed to identify the differentially expressed genes (DEGs) of stroke. The crosstalk genes linking ischemic stroke and immunosuppression were obtained by overlapping the ischemic stroke-DEGs and ISGs and, as well as feature selection performed by the Boruta algorithm. The classification model was constructed to evaluate the prediction accuracy of the obtained crosstalk gene results by using support vector machines (SVM). 

Among the 22 Immune infiltrating cells, T cells CD4 memory resting was significantly downregulated, while macrophages M0 was significantly upregulated in ischemic stroke. By overlapping the 54 crosstalk genes obtained by feature selection and ischemic stroke-related genes obtained in the DisGenet database, a total of 17 crosstalk genes were obtained: ARG1, CD36, FCN1, GRN, IL7R, JAK2, MAFB, MMP9, PTEN, STAT3, STAT5A, THBS1, TLR2, TLR4, TLR7, TNFSF10, and VASP. The transcription factors (TFs) targeting crosstalk genes (e.g., STAT3, SPI1, CEPBD, SP1, TP53, NFIL3, STAT1, HIF1A, and JUN) and signaling pathways enriched by crosstalk genes (e.g., PD-L1 expression and PD-1 checkpoint pathway, NF-kappa B, IL-17, TNF, and NOD-like receptor signaling) were also obtained by performing TFs-crosstalk genes network analysis and functional enrichment analysis.