Machine learning analysis of FOSL2 and RHoBTB1 as central immunological regulators in knee osteoarthritis synovium

Area of Science:

• Genomics and Bioinformatics
• Immunology
• Molecular Biology

Background:

• Knee osteoarthritis (OA) is a significant cause of disability, with synovitis playing a crucial role in its development and progression.
• Understanding the molecular mechanisms of knee OA is essential for developing effective treatments.
• Gene expression analysis using machine learning offers a powerful approach to identify key molecular players in complex diseases like knee OA.

Purpose of the Study:

• To identify key genes involved in the pathogenesis of knee osteoarthritis using machine learning.
• To investigate the relationship between identified key genes, immune cell infiltration, and disease markers.
• To explore potential therapeutic strategies by screening for small-molecule compounds targeting the identified pathways.

Main Methods:

• Differential gene expression analysis of knee OA synovial tissues using the GSE55235 dataset.
• Application of machine learning algorithms (LASSO, SVM-RFE) for key gene identification.
• Validation using an external dataset (GSE51588), an in vitro OA animal model, CIBERSORT for immune cell analysis, and single-cell RNA sequencing.

Main Results:

• Identified 930 differentially expressed genes, with FOSL2 and RHoBTB1 selected as key genes by machine learning.
• Validated FOSL2 and RHoBTB1 expression in independent datasets and an in vitro OA mouse model.
• Found significant correlations between immune cell infiltration and the key genes, and identified potential small-molecule compounds for therapeutic intervention.

Conclusions:

• FOSL2 and RHoBTB1 are implicated as critical genes in knee osteoarthritis pathogenesis, linked to immune cell dynamics.
• These genes represent promising therapeutic targets for knee OA, warranting further investigation.
• The study provides a foundation for developing targeted therapies to combat knee osteoarthritis.