Identification and Analysis of a Four-Gene Set for Diagnosing SFTS Virus Infection Based on Machine Learning Methods and Its Association with Immune Cell Infiltration

Abstract

Severe Fever with thrombocytopenia syndrome (SFTS) is a highly fatal viral infectious disease that poses a significant threat to public health. Currently, the phase and pathogenesis of SFTS are not well understood, and there are no specific vaccines or effective treatment available. Therefore, it is crucial to identify biomarkers for diagnosing acute SFTS, which has a high mortality rate. In this study, we conducted differentially expressed genes (DEGs) analysis and WGCNA module analysis on the GSE144358 dataset, comparing the acute phase of SFTSV-infected patients with healthy individuals. Through the LASSO–Cox and random forest algorithms, a total of 2128 genes were analyzed, leading to the identification of four genes: ADIPOR1, CENPO, E2F2, and H2AC17. The GSEA analysis of these four genes demonstrated a significant correlation with immune cell function and cell cycle, aligning with the functional enrichment findings of DEGs. Furthermore, we also utilized CIBERSORT to analyze the immune cell infiltration and its correlation with characteristic genes. The results indicate that the combination of ADIPOR1, CENPO, E2F2, and H2AC17 genes has the potential as characteristic genes for diagnosing and studying the acute phase of SFTS virus (SFTSV) infection.

Keywords

• SFTS
• SFTS acute phase
• machine learning
• LASSO–Cox
• immune cells infiltration