Gradient boosting-based classification of interactome hub genes in periimplantitis with periodontitis - an integrated bioinformatic approach

Introduction: Peri-implantitis, a destructive inflammatory condition affecting the tissues surrounding dental implants, shares pathological similarities with periodontitis, a chronic inflammatory disease that impacts the supporting structures of natural teeth. This study utilizes a network-based approach to classify interactome hub genes associated with peri-implantitis and periodontitis, aiming to improve understanding of disease mechanisms and identify potential therapeutic targets.

Methods: We employed gradient boosting and Weighted Gene Co-expression Network Analysis (WGCNA) to predict and classify these interactome hub genes. Gene expression data related to these diseases were sourced from the NCBI GEO dataset GSE223924, and differential gene expression analysis was conducted using the NCBI GEO R tool. Through WGCNA, we constructed a co-expression network to identify key hub genes, while gradient boosting was used to predict these hub genes.

Results: Our analysis revealed a co-expression network comprising 216 genes, including prominent hub genes such as IL17RC, CCN2, BMP7, TPM1, and TIMP1, which are implicated in periodontal disease. The gradient boosting model achieved an 88.2% accuracy in classifying interactome hub genes in samples related to peri-implantitis and periodontitis.

Discussion: These identified genes play roles in inflammation, osteoclast genesis, angiogenesis, and immune response regulation. This study highlights essential hub genes and molecular pathways associated with peri-implantitis and periodontitis, suggesting potential therapeutic targets for developing innovative treatment strategies.