BACKGROUND
Peri-implantitis is the main cause of failure of implant treatment, and there is little research on its molecular mechanism. This study aimed to identify key biomarkers and immune infiltration of peri-implantitis using a bioinformatics method.
MATERIAL AND METHODS
Three Gene Ontology (GO) gene expression profiles were selected from the Gene Expression Omnibus. Differentially expressed genes (DEGs) were identified by the LIMMA package, and functional correlations of DEGs were analyzed by Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis. Information on immune-related genes was obtained from ImmPort (https://www.immport.org) and InnateDB (http://www.innatedb.com). Immune-related DEGs were screened by least absolute shrinkage and selection operator (LASSO) and support vector machine-recursive feature elimination (SVM-RFE). The single-sample Gene Set Enrichment Analysis algorithm was used to analyze immune cell infiltration in gingival tissue between peri-implantitis and normal controls. Finally, results of bioinformatics analysis were verified by qPCR.
RESULTS
A total of 398 DEGs were identified, of which 96 were immune-related. Enrichment analysis showed these genes were enriched in inflammatory response, leucocyte chemotaxis, immune response-regulating signaling pathway, and cell activation. Seven key genes were selected by LASSO and SVM-RFE. Receiver operating characteristic curve results showed these genes had excellent diagnostic efficacy. Results of qPCR showed significant differences in the expression of these genes.
CONCLUSIONS
Differences in key genes and immune infiltration between peri-implantitis and gingival tissues of normal controls may provide new insights into the development of peri-implantitis. Elucidating the difference in immune infiltration between peri-implantitis tissues and normal tissues will help to understand the development of peri-implantitis.

Keywords: Peri-Implantitis; Computational Biology