A diagnostic model, based on 9 key genes, can reliably distinguish between patients with and without osteoporosis, according to research published in the Journal of Orthopaedic Surgery and Research.
Osteoporosis is becoming increasingly common among the aging population of the world. Researchers aimed to identify diagnostic signatures for osteoporosis for the clinical application of a machine learning analysis of gene expression profile.
Researchers downloaded mRNA profiles from the GEO database (Number: GSE152073), which included 90 peripheral blood samples from 44 individuals with osteoporosis and 46 healthy participants. They used weighted gene co-expression network analysis (WGCNA) to elucidate the correlation among the various genes in all samples. The Gene Ontology (GO) term, which included biological process, molecular function, and cellular component, and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed via the clusterProfiler function package of R language to identify key pathways involved in the progression of osteoporosis.
The widely used database STRING was used to analyze and predict the functional connections and interactions of proteins found in human diseases. In the current analysis, the researchers applied STRING to evaluate the interaction pairs among the various proteins during the evolution of osteoporosis. The protein-protein interaction (PPI) network was visualized based on Cytoscape, and the key genes were screened with use of the cytoHubba plug-in.
Researchers identified a gene module that included 176 genes predictive of being linked to the occurrence of osteoporosis. Based on these genes, 16 significantly enriched GO terms and 1 significantly enriched KEGG pathway were collected. Following this, the top 50 genes in the PPI network were recognized, with 22 genes being screened according to stepwise regression analysis from the 50 key genes. Next, 9 of these genes were further screened by multivariate regression analysis using the significant threshold of P value less than 0.01. The 9 genes that were identified and still significant (P <.001) included LCK, LY9, CD5, P2RY8, KCTD7, MDN1, ITK, CAPN2, and HTT, which implied that these 9 genes might be significantly associated with the occurrence of osteoporosis.
Researchers concluded that in order to assess the reliability of the current predictive model, more samples are required. Although the role of several key genes has been well described in the progression of osteoporosis, other genes need to be explored in detail in the future. The diagnostic model described in this analysis might have the potential to assist in the clinical diagnosis of osteoporosis.
Chen X, Liu G, Wang S, Zhang H, Xue P. Machine learning analysis of gene expression profile reveals a novel diagnostic signature for osteoporosis. J Orthop Surg Res. 2021;16(1):189. doi:10.1186/s13018-021-02329-1