Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis
Abstract Background Branched-chain amino acids (BCAA) metabolism is significantly associated with osteoarthritis (OA), but the specific mechanism of BCAA related genes (BCAA-RGs) in OA is still unclear. Therefore, this research intended to identify potential biomarkers and mechanisms of action of BC...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-05-01
|
| Series: | BMC Musculoskeletal Disorders |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12891-025-08779-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849687759555395584 |
|---|---|
| author | Xiao-Zhi ZhaYang Yan-Xiong Chen Wen-Da Hua Zheng-Lin Bai Yun-Peng Jin Xing-Wen Zhao Quan-Fu Liu Zeng-Dong Meng |
| author_facet | Xiao-Zhi ZhaYang Yan-Xiong Chen Wen-Da Hua Zheng-Lin Bai Yun-Peng Jin Xing-Wen Zhao Quan-Fu Liu Zeng-Dong Meng |
| author_sort | Xiao-Zhi ZhaYang |
| collection | DOAJ |
| description | Abstract Background Branched-chain amino acids (BCAA) metabolism is significantly associated with osteoarthritis (OA), but the specific mechanism of BCAA related genes (BCAA-RGs) in OA is still unclear. Therefore, this research intended to identify potential biomarkers and mechanisms of action of BCAA-RGs in OA tissues. Methods Differential genes were obtained from the Gene Expression Omnibus (GEO) database and intersections were taken with BCAA-RGs to identify candidate genes. The underlying mechanisms were revealed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). Subsequently, by combining three machine learning algorithms to identify genes with highly correlated OA features. In addition, created diagnostic maps and subject Receiver operating characteristic curves (ROCs) to assess the ability of the signature genes to diagnose OA and to predict their possible roles in molecular regulatory network axes and molecular signaling pathways. Results Eight candidate genes were acquired by intersecting 4,178 DEGs and 14 BCAA-RGs. Subsequently, five candidate biomarkers were obtained, namely SLC3A2, SLC7A5, SLC43A2, SLC43A1, and SLC7A7. Importantly, SLC3A2 and SLC7A5 were validated by validation set and qRT-PCR. Furthermore, the nomogram constructed by SLC3A2 and SLC7A5 exhibited excellent accuracy in predicting the incidence of OA. The enrichment results demonstrated that SLC3A2 and SLC7A5 were significantly enriched in ribosome, insulin signaling pathway, olfactory transduction, etc. Meanwhile, we also found XIST regulated SLC7A5 through hsa-miR-30e-5p, and regulated SLC3A2 through hsa-miR-7-5p.OIP5-AS1 regulated SLC7A5 and SLC3A2 through hsa-miR-7-5p. By the way, 150 drugs were identified, including Acetaminophen and Acrylamide, which exhibited simultaneous targeting of these two biomarkers. Conclusion Based on bioinformatics, SLC3A2 and SLC7A5 were identified as biomarkers related to BCAA in OA, which may provide a new reference for the treatment and diagnosis of OA patients. |
| format | Article |
| id | doaj-art-908a69f390a64d7a88c78bc8c3a52d1c |
| institution | DOAJ |
| issn | 1471-2474 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Musculoskeletal Disorders |
| spelling | doaj-art-908a69f390a64d7a88c78bc8c3a52d1c2025-08-20T03:22:15ZengBMCBMC Musculoskeletal Disorders1471-24742025-05-0126111610.1186/s12891-025-08779-6Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritisXiao-Zhi ZhaYang0Yan-Xiong Chen1Wen-Da Hua2Zheng-Lin Bai3Yun-Peng Jin4Xing-Wen Zhao5Quan-Fu Liu6Zeng-Dong Meng7Faculty of Medical Science, Kunming University of Science and TechnologyFaculty of Medical Science, Kunming University of Science and TechnologyDepartment of Orthopedic Surgery, The First People’s Hospital of Yunnan Province, Affiliated Hospital of Kunming University of Science and TechnologyDepartment of Orthopedic Surgery, The First People’s Hospital of Yunnan Province, Affiliated Hospital of Kunming University of Science and TechnologyDepartment of Orthopedic Surgery, The First People’s Hospital of Yunnan Province, Affiliated Hospital of Kunming University of Science and TechnologyDepartment of Orthopedic Surgery, The First People’s Hospital of Yunnan Province, Affiliated Hospital of Kunming University of Science and TechnologyDepartment of Orthopedic Surgery, The First People’s Hospital of Yunnan Province, Affiliated Hospital of Kunming University of Science and TechnologyFaculty of Medical Science, Kunming University of Science and TechnologyAbstract Background Branched-chain amino acids (BCAA) metabolism is significantly associated with osteoarthritis (OA), but the specific mechanism of BCAA related genes (BCAA-RGs) in OA is still unclear. Therefore, this research intended to identify potential biomarkers and mechanisms of action of BCAA-RGs in OA tissues. Methods Differential genes were obtained from the Gene Expression Omnibus (GEO) database and intersections were taken with BCAA-RGs to identify candidate genes. The underlying mechanisms were revealed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). Subsequently, by combining three machine learning algorithms to identify genes with highly correlated OA features. In addition, created diagnostic maps and subject Receiver operating characteristic curves (ROCs) to assess the ability of the signature genes to diagnose OA and to predict their possible roles in molecular regulatory network axes and molecular signaling pathways. Results Eight candidate genes were acquired by intersecting 4,178 DEGs and 14 BCAA-RGs. Subsequently, five candidate biomarkers were obtained, namely SLC3A2, SLC7A5, SLC43A2, SLC43A1, and SLC7A7. Importantly, SLC3A2 and SLC7A5 were validated by validation set and qRT-PCR. Furthermore, the nomogram constructed by SLC3A2 and SLC7A5 exhibited excellent accuracy in predicting the incidence of OA. The enrichment results demonstrated that SLC3A2 and SLC7A5 were significantly enriched in ribosome, insulin signaling pathway, olfactory transduction, etc. Meanwhile, we also found XIST regulated SLC7A5 through hsa-miR-30e-5p, and regulated SLC3A2 through hsa-miR-7-5p.OIP5-AS1 regulated SLC7A5 and SLC3A2 through hsa-miR-7-5p. By the way, 150 drugs were identified, including Acetaminophen and Acrylamide, which exhibited simultaneous targeting of these two biomarkers. Conclusion Based on bioinformatics, SLC3A2 and SLC7A5 were identified as biomarkers related to BCAA in OA, which may provide a new reference for the treatment and diagnosis of OA patients.https://doi.org/10.1186/s12891-025-08779-6OsteoarthritisBranched-chain amino acidsMachine learningBioinformatic analysisDrug prediction |
| spellingShingle | Xiao-Zhi ZhaYang Yan-Xiong Chen Wen-Da Hua Zheng-Lin Bai Yun-Peng Jin Xing-Wen Zhao Quan-Fu Liu Zeng-Dong Meng Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis BMC Musculoskeletal Disorders Osteoarthritis Branched-chain amino acids Machine learning Bioinformatic analysis Drug prediction |
| title | Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis |
| title_full | Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis |
| title_fullStr | Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis |
| title_full_unstemmed | Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis |
| title_short | Integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis |
| title_sort | integrating bioinformatics and machine learning to identify biomarkers of branched chain amino acid related genes in osteoarthritis |
| topic | Osteoarthritis Branched-chain amino acids Machine learning Bioinformatic analysis Drug prediction |
| url | https://doi.org/10.1186/s12891-025-08779-6 |
| work_keys_str_mv | AT xiaozhizhayang integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT yanxiongchen integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT wendahua integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT zhenglinbai integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT yunpengjin integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT xingwenzhao integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT quanfuliu integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis AT zengdongmeng integratingbioinformaticsandmachinelearningtoidentifybiomarkersofbranchedchainaminoacidrelatedgenesinosteoarthritis |