Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms
Abstract Background Osteoporosis significantly increases the risk of vertebral fractures, particularly among postmenopausal women, decreasing their quality of life. These fractures, often undiagnosed, can lead to severe health consequences and are influenced by bone mineral density and abnormal load...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-02-01
|
| Series: | BioData Mining |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13040-025-00427-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825197598229659648 |
|---|---|
| author | Yi-Chou Chen Hui-Chen Su Shih-Ming Huang Ching-Hsiao Yu Jen-Huei Chang Yi-Lin Chiu |
| author_facet | Yi-Chou Chen Hui-Chen Su Shih-Ming Huang Ching-Hsiao Yu Jen-Huei Chang Yi-Lin Chiu |
| author_sort | Yi-Chou Chen |
| collection | DOAJ |
| description | Abstract Background Osteoporosis significantly increases the risk of vertebral fractures, particularly among postmenopausal women, decreasing their quality of life. These fractures, often undiagnosed, can lead to severe health consequences and are influenced by bone mineral density and abnormal loads. Management strategies range from non-surgical interventions to surgical treatments. Moreover, the interaction between immune cells and bone cells plays a crucial role in bone repair processes, highlighting the importance of osteoimmunology in understanding and treating bone pathologies. Methods This study aims to investigate the xCell signature-based immune cell profiles in osteoporotic patients with and without vertebral fractures, utilizing advanced predictive modeling through the XGBoost algorithm. Results Our findings reveal an increased presence of CD4 + naïve T cells and central memory T cells in VF patients, indicating distinct adaptive immune responses. The XGBoost model identified Th1 cells, CD4 memory T cells, and hematopoietic stem cells as key predictors of VF. Notably, VF patients exhibited a reduction in Th1 cells and an enrichment of Th17 cells, which promote osteoclastogenesis and bone resorption. Gene expression analysis further highlighted an upregulation of osteoclast-related genes and a downregulation of osteoblast-related genes in VF patients, emphasizing the disrupted balance between bone formation and resorption. These findings underscore the critical role of immune cells in the pathogenesis of osteoporotic fractures and highlight the potential of XGBoost in identifying key biomarkers and therapeutic targets for mitigating fracture risk in osteoporotic patients. |
| format | Article |
| id | doaj-art-48e06f517e854f31bb53df0c4f2af802 |
| institution | Kabale University |
| issn | 1756-0381 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | BMC |
| record_format | Article |
| series | BioData Mining |
| spelling | doaj-art-48e06f517e854f31bb53df0c4f2af8022025-02-09T12:15:55ZengBMCBioData Mining1756-03812025-02-0118112010.1186/s13040-025-00427-yImmune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithmsYi-Chou Chen0Hui-Chen Su1Shih-Ming Huang2Ching-Hsiao Yu3Jen-Huei Chang4Yi-Lin Chiu5Department of Orthopedics, Taoyuan General Hospital, Ministry of Health and WelfareDepartment of Pharmacy, Chi-Mei Medical CenterDepartment of Biochemistry, National Defense Medical CenterDepartment of Orthopedics, Taoyuan General Hospital, Ministry of Health and WelfareOrthopedic Department, Cardinal Tien HospitalDepartment of Biochemistry, National Defense Medical CenterAbstract Background Osteoporosis significantly increases the risk of vertebral fractures, particularly among postmenopausal women, decreasing their quality of life. These fractures, often undiagnosed, can lead to severe health consequences and are influenced by bone mineral density and abnormal loads. Management strategies range from non-surgical interventions to surgical treatments. Moreover, the interaction between immune cells and bone cells plays a crucial role in bone repair processes, highlighting the importance of osteoimmunology in understanding and treating bone pathologies. Methods This study aims to investigate the xCell signature-based immune cell profiles in osteoporotic patients with and without vertebral fractures, utilizing advanced predictive modeling through the XGBoost algorithm. Results Our findings reveal an increased presence of CD4 + naïve T cells and central memory T cells in VF patients, indicating distinct adaptive immune responses. The XGBoost model identified Th1 cells, CD4 memory T cells, and hematopoietic stem cells as key predictors of VF. Notably, VF patients exhibited a reduction in Th1 cells and an enrichment of Th17 cells, which promote osteoclastogenesis and bone resorption. Gene expression analysis further highlighted an upregulation of osteoclast-related genes and a downregulation of osteoblast-related genes in VF patients, emphasizing the disrupted balance between bone formation and resorption. These findings underscore the critical role of immune cells in the pathogenesis of osteoporotic fractures and highlight the potential of XGBoost in identifying key biomarkers and therapeutic targets for mitigating fracture risk in osteoporotic patients.https://doi.org/10.1186/s13040-025-00427-yOsteoporosisVertebral fracturesXGBoostTh17 cell differentiation |
| spellingShingle | Yi-Chou Chen Hui-Chen Su Shih-Ming Huang Ching-Hsiao Yu Jen-Huei Chang Yi-Lin Chiu Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms BioData Mining Osteoporosis Vertebral fractures XGBoost Th17 cell differentiation |
| title | Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms |
| title_full | Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms |
| title_fullStr | Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms |
| title_full_unstemmed | Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms |
| title_short | Immune cell profiles and predictive modeling in osteoporotic vertebral fractures using XGBoost machine learning algorithms |
| title_sort | immune cell profiles and predictive modeling in osteoporotic vertebral fractures using xgboost machine learning algorithms |
| topic | Osteoporosis Vertebral fractures XGBoost Th17 cell differentiation |
| url | https://doi.org/10.1186/s13040-025-00427-y |
| work_keys_str_mv | AT yichouchen immunecellprofilesandpredictivemodelinginosteoporoticvertebralfracturesusingxgboostmachinelearningalgorithms AT huichensu immunecellprofilesandpredictivemodelinginosteoporoticvertebralfracturesusingxgboostmachinelearningalgorithms AT shihminghuang immunecellprofilesandpredictivemodelinginosteoporoticvertebralfracturesusingxgboostmachinelearningalgorithms AT chinghsiaoyu immunecellprofilesandpredictivemodelinginosteoporoticvertebralfracturesusingxgboostmachinelearningalgorithms AT jenhueichang immunecellprofilesandpredictivemodelinginosteoporoticvertebralfracturesusingxgboostmachinelearningalgorithms AT yilinchiu immunecellprofilesandpredictivemodelinginosteoporoticvertebralfracturesusingxgboostmachinelearningalgorithms |