Research Paper Volume 16, Issue 11 pp 9437—9459
Machine learning-based identification and immune characterization of ferroptosis-related molecular clusters in osteoarthritis and validation
- 1 Department of Orthopedics, Shengjing Hospital of China Medical University, Shenyang, China
- 2 Department of Cadre Wards, Liaoning University of Traditional Chinese Medicine Affiliated Orthopedic Hospital, Shenyang, China
Received: January 15, 2024 Accepted: April 18, 2024 Published: May 29, 2024
https://doi.org/10.18632/aging.205875How to Cite
Copyright: © 2024 Guo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Abstract
Osteoarthritis (OA), a degenerative joint disease, involves synovial inflammation, subchondral bone erosion, and cartilage degeneration. Ferroptosis, a regulated non-apoptotic programmed cell death, is associated with various diseases. This study investigates ferroptosis-related molecular subtypes in OA to comprehend underlying mechanisms. The Gene Expression Omnibus datasets GSE206848, GSE55457, GSE55235, GSE77298 and GSE82107 were used utilized. Unsupervised clustering identified the ferroptosis-related gene (FRG) subtypes, and their immune characteristics were assessed. FRG signatures were derived using LASSO and SVM-RFE algorithms, forming models to evaluate OA’s ferroptosis-related immune features. Three FRG clusters were found to be immunologically heterogeneous, with cluster 1 displaying robust immune response. Models identified nine key signature genes via algorithms, demonstrating strong diagnostic and prognostic performance. Finally, qRT-PCR and Western blot validated these genes, offering consistent results. In addition, some of these genes may have implications as new therapeutic targets and can be used to guide clinical applications.