Characterization of necroptosis and immune landscape in dermatomyositis by bioinformatics analysis and machine learning

Abstract

Abstract BackgroundDermatomyositis (DM) is a cell-mediated autoimmune disease of intricate aetiology. Necroptosis is a newly identified form of programmed cell death. This research aimed to explore the value of necroptosis-related genes in DM. Methods DM datasets were obtained from Gene Expression Omnibus (GEO) database. Necroptosis-related differentially expressed genes (NRDEGs) of DM were identified by intersecting differentially expressed genes (DEGs) with necroptosis gene set. Then, signature genes of NRDEGs were determined by the machine learning method of random forest (RF), support vector machine-recursive feature elimination (SVM-RFE), and the least absolute shrinkage and selection operator regression (LASSO). Moreover, immune microenvironment of DM and its correlation with signature genes were created to assess immune dysregulation. Besides, functional enrichment analysis, protein-protein interaction (PPI) co-expression network construction, transcription factor (TF)-miRNA network analysis were collectively performed on signature genes. In addition, the Mfuzz expression pattern clustering and functional enrichment based on the optimal signature was conducted. Results A total of 2524 DEGs in GSE143323 were obtained, including BAX, BIRC3, JAK3, SPATA2L and TNFSF10. Through the intersection with necroptosis gene set, 28 NRDEGs were examined. Furthermore, five signature genes were identified via machine learning and were verified in GSE1551. In immune landscape evaluation, signature genes were positively correlated with most immunocytes, human leukocyte antigen (HLA) genes, and immune checkpoints. Among them, TNFSF10 was the best diagnostic signature of DM. The most highly associated module genes with TNFSF10 by Mfuzz expression pattern clustering mainly enriched in immunity and immunoregulation. Conclusions Necroptosis occurs in DM, and is closely related to DM immune microenvironment, which merits further investigations in the necroptosis of DM pathogenesis.