Objective: This study aimed to exploration of glycosylation-related genes and immune infiltration analysis of IgA nephropathy. Methods: IgAN datasets were obtained from the GEO database. We then identified differentially expressed glycosylation-related genes and performed functional enrichment analyses. Next, optimal feature genes (OFGs) were selected using LASSO, SVM-RFE, and Random Forest algorithms. The expression of OFGs in IgAN were validated by immunohistochemistry staining, western blot and the Nephroseq v5 database. OFGs were further used to create a nomogram model, compare immune cell infiltration and construct a ceRNA network. Results: After screening, three OFGs (ST8SIA1, CHSY1 and PIGH) were first reported. The nomogram model showed that OFGs had good predictive value for IgAN occurrence. Compared to normal samples, IgAN showed increased infiltration of CD8+^{T cells}, na?ve CD4+ ^{T cell}, memory activated CD4+ ^{T cells}, resting dendritic cells, and resting mast cells, while na?ve B cells, plasma cells, activated mast cells, and neutrophils were reduced. OFGs were associated with memory activated CD4+ ^{T cells}, memory resting CD4+ ^{T cells}, na?ve CD4+ ^{T cell, na?ve B cells} among others. The validation experiments also revealed that the expression levels of CHSY1 and PIGH were significantly decreased, while the expression level of ST8SIA1 was significantly increased in IgAN compared with minimal change nephropathy. Of note, the expression levels of OFGs in diabetic nephropathy and minimal change nephropathy were not statistically different. A ceRNA network consisting of 117 lncRNAs, 67 miRNAs, and 3 OFGs was constructed. Conclusion: ST8SIA1, CHSY1, and PIGH were identified as potential targets for diagnosis and treatment of IgAN. In conjunction with immune cell infiltration and ceRNA network, these results offer a novel perspective for future research on IgAN.