Identification of Immune-Associated Genes in Diagnosing Polycystic Ovary Syndrome with Metabolic Syndrome by Weighted Gene Co-expression Network Analysis

Abstract

Background: Dysregulated immune system and metabolic dysfunction take an essential impart on the pathogenesis of polycystic ovary syndrome (PCOS) and metabolic syndrome (MS). The purpose of this study is to identify the key immune-Associated genes of PCOS patients with MS.

Methods: Three PCOS and one MS dataset from the Gene Expression Omnibus (GEO) database were downloaded. After the data sets were combined with the de-batch effect, analysis was conducted using methods such as differentially expressed genes (DEGs), weighted gene co-expression networks (WGCNA), functional enrichment, protein-protein interaction (PPI) network construction, and LASSO regression, which were used to identify candidate immune-related essential genes for diagnosing the combination of these two diseases. Evaluate the obtained genes through ROC and ultimately explore abnormalities in PCOS through immune cell infiltration analysis.

Results: The merged PCOS dataset contains 1513 differentially expressed genes (DEGs), with 691 identified by MS. DEGs of MS were primarily enriched in immune regulation and metabolic dysfunction. The crossed gene functions of DEGs in PCOS and module genes in MS are also mainly enriched in the immune system and metabolic dysfunction. We screened 35 node genes from 57 cross genes through the PPI network and selected six candidate genes using the lasso regression method. Five genes analyzed by the ROC curve had high diagnostic values (RAB4B, SFP91, ARHGAP33, LIME1, and SHMT2) (area under the curve from 0.72 to 0.89). Finally, an imbalance in the proportion of immune cells in PCOS was also observed.

Conclusion: Five candidate genes were identified as the final diagnostic markers. Our conclusion can provide potential diagnostic candidate genes for patients of MS with PCOS.