Exploring the relationship between cartilage-associated m6a gene and osteoarthritis development based on bioinformatics and machine learning

Abstract

Abstract Purpose:This study aimed to analyze the expression of n6-methyladenosine (m6a)-related genes in osteoarthritis (OA), the relationship between m6a signature genes and clinical morbidity, and the correlation between m6a gene immune cells by using bioinformatics and random forest tree methods. Methods:OA-related microarrays were obtained from the GEO database. The m6a-related genes were extracted, and their differential gene expression was analyzed using R software. Appropriate gene screening methods were selected to obtain m6a disease signature genes; m6a clinical prediction models were established; decision curve analysis (DCA) was applied to verify the model’s accuracy. Typing was performed according to m6a expression, and genetic differences between typing and differences in immune infiltration were analyzed. The correlation between the differential genes and immune cells was also analyzed. Finally, the m6a differential genes were analyzed using Metascape. Results:Random forest tree screening was used to obtain the following m6a disease signature genes for cartilage in OA: METTL3, CBLL1, YTHDC1, YTHDC2, YTHDF1, and FMR1. A strong correlation was found between the expression of disease-characterizing genes and clinical disease incidence, which was higher when the total score was between 200 and 230. Based on the m6a gene expression in cartilage, the samples were divided into groups A and B, and METTL3, FMR1, and YTHDC2 had significant genetic differences in the two groups. Among the immune cells, activated CD4T, activated dendritic, natural killer T, and plasma cells were significantly different in the two groups. A significant correlation was found between the high expression of immune cells and the three m6a genes in group B. Metascape functional pathway analysis revealed that OA is mainly related to cell development, differentiation, morphological changes, chemotaxis, and inflammatory response, mainly involving the FRA pathway. Conclusion: The expression of m6a disease-characterizing genes is significantly correlated with the clinical incidence of OA, and the abnormal expression of m6a-related genes in OA cartilage is an important factor that may cause cartilage damage mainly by affecting immune cells, thus releasing relevant pro-inflammatory factors causing damage to chondrocytes.