Screening Key Genes for Rheumatoid Arthritis Based on Bioinformatics and Machine Learning
Rheumatoid arthritis(RA)is an organ-specific autoimmune disease characterized by chronic synovitis and bone erosion.Its high rate of disability has serious implications for society and individuals,and there is a need for effective and reliable diagnostic markers and therapeutic targets for RA.In this study,expression profile datasets and single cell sequencing datasets were download-ed from the GEO database to elucidate potentially essential candidate genes and pathways in RA.Firstly,we annotated 5 types of cell clusters using single-cell sequencing data,and 4109 RA-related genes were screened in each cell cluster.Then gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)pathway enrichment analysis were conducted.Secondly,a total of 677 RA-related genes were identified by bioinformatics analysis of expression data,and GSEA enrich-ment analysis of differentially expressed genes was performed.Then,the differential expression genes(DEGs)and the differential expression gene association of each cluster were analyzed,and the machine learning algorithm was used for further analysis.Six key genes were obtained,namely IGLL5,AIM1,NKG7,PSMB9,ANKRD11 and BIRC3.Finally,ROC curves of these 6 genes in train-ing set and validation set showed that they had good diagnostic performance in rheumatoid arthritis.
NETL
NSTL
万方数据
