Objective:To explore the gene microarray of patients with ankylosing spondylitis in GEO database by using various bioinformatics methods, and to explore the possible targets and mechanisms of action.Methods:The GEO database was searched with "ankylosing spondylitis" the keyword, and the expression profile of genes related to AS was selected as the research object. Standard difference analysis, weighted co-expression analysis and gene set enrichment analysis were conducted to construct the disease set. GO and KEGG enrichment analysis were performed on the disease sets. The NCC algorithm identifies the first five key genes. THP-1 cells were implanted into RPMI-1640 culture medium containing 10% fetal bovine serum to multiply and construct the cell model of AS in vitro. The expression levels of 5 key genes were detected by qRT-PCR and Western blot. The experimental measurement data were expressed as mean± standard deviation, and the t test was used in comparison between the two groups. Results:One thousand six hundred and sixty seven disease genes were analyzed, functional annotation was mainly concentrated in 689 molecular components of cytoplasmic ribosomes, ribosomal subunits, ribosomes, cytoplasmic large ribosomal subunits, the structural composition of ribosomal REDOX enzyme activity, 1 002 molecular functions of NADH dehydrogenase activity, NADH dehydrogenase activity, and 5 764 molecular processes of mRNA catabolism and RNA catabolism The physical process involved 1 002 signaling pathways involved in Alzheimer′s disease, Prion disease, Parkinson′s disease, and the first 5 key genes were identified as RPS11, RPL4, RPL37A, RPS23, and RPS9. The experimental results were obtained by t test. The results showed that TNF-α mRNA ( t=5.59, P=0.001) and protein ( t=20.14, P<0.001) were significantly increased, indicating that LPS had induced inflammatory response in THP-1 cells, while RPL37AmRNA ( t=5.87, P=0.001), RPS11 mRNA ( t=3.88, P=0.008), RPS23 mRNA ( t=2.64, P=0.038), RPL37A protein ( t=3.18, P=0.030), RPS11 protein ( t=11.26, P<0.001), RPS23 protein ( t=5.64, P<0.001), increased, while RPS9 mRNA ( t=3.16, P=0.020), RPL4 mRNA ( t=2.54, P=0.044), RPS9 protein ( t=5.85, P<0.001) and RPL4 ( t=2.93, P=0.040) protein expressions decreased. RPL23 stimulated the joint synovial tissue to produce effect-T lymphocytes and release a large number of IL-2 and other inflammatory cytokines. RPS9 acts on the early stages of ribosomogenesis, and knocking down RPS9 reduced overall protein synthesis. RPL4 interacted with TTC22 protein to enhance the binding of WTAP mRNA to RPL4, which was associated with immune diseases. The nucleoprotein OGFOD1 catalyzed the hydroxylation of RPS23 and participated in the inflammatory process. The chromosome conformation confirmed the single nucleotide polymorphism function of IL23R genomic locus in AS disease. Conclusion:Ribosomal protein may be an important target for exploring the mechanism of AS inflammation.