Identification of Immune Infiltration and the Potential Biomarkers in Diabetic Peripheral Neuropathy through Bioinformatics and Machine Learning Methods

Abstract

Diabetic peripheral neuropathy (DPN) is one of the most common chronic complications in diabetes. Previous studies have shown that chronic neuroinflammation was associated with DPN. However, further research is needed to investigate the exact immune molecular mechanism underlying the pathogenesis of DPN. Expression profiles were downloaded from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were screened by R software. After functional enrichment analysis of DEGs, a protein–protein interaction (PPI) network analysis was performed. The CIBERSORT algorithm was used to evaluate the infiltration of immune cells in DPN. Next, the least absolute shrinkage and selection operator (LASSO) logistic regression and support vector machine-recursive feature elimination (SVM-RFE) algorithms were applied to identify potential DPN diagnostic markers. Finally, the results were further validated by qRT-PCR. A total of 1308 DEGs were screened in this study. Enrichment analysis identified that DEGs were significantly enriched in immune-related biological functions and pathways. Immune cell infiltration analysis found that M1 and M2 macrophages, monocytes, resting mast cells, resting CD4 memory T cells and follicular helper T cells were involved in the development of DPN. LTBP2 and GPNMB were identified as diagnostic markers of DPN. qRT-PCR results showed that 15 mRNAs, including LTBP2 and GPNMB, were differentially expressed, consistent with the microarray results. In conclusion, LTBP2 and GPNMB can be used as novel candidate molecular diagnostic markers for DPN. Furthermore, the infiltration of immune cells plays an important role in the progression of DPN.