Identification of osteoporosis based on gene biomarkers using support vector machine

Area of Science:

• Genomics
• Biomarker Discovery
• Computational Biology

Background:

• Osteoporosis is a significant global health issue.
• Effective biomarkers are needed for early and accurate osteoporosis detection.
• Understanding the molecular mechanisms of osteoporosis is crucial.

Purpose of the Study:

• To identify novel biomarkers for osteoporosis detection.
• To investigate the molecular pathways involved in osteoporosis.
• To develop a computational model for diagnosing osteoporosis.

Main Methods:

• Differential gene expression analysis identified 559 DEGs.
• Weighted gene co-expression network analysis highlighted significant gene modules.
• Protein-protein interaction network and co-expression network analysis identified hub genes.
• Support Vector Machine (SVM) model was developed for classification.

Main Results:

• PI3K-Akt and Foxo signaling pathways were significantly enriched in osteoporosis.
• Six hub genes (VEGFA, DDX5, SOD2, HNRNPD, EIF5B, HSP90B1) were identified.
• The SVM model achieved 100% sensitivity and specificity with an AUC of 1.

Conclusions:

• The identified hub genes and pathways offer insights into osteoporosis pathogenesis.
• The SVM-based system demonstrates high potential for accurate osteoporosis diagnosis.
• This approach could aid in early detection and management of osteoporosis.