The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis
Abstract Background Endothelial cells are integral components of the tumor microenvironment and play a multifaceted role in tumor immunotherapy. Targeting endothelial cells and related signaling pathways can improve the effectiveness of immunotherapy by normalizing tumor blood vessels and promoting...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | Biology Direct |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13062-024-00591-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841544886594043904 |
|---|---|
| author | Gujun Cong Jingjing Shao Feng Xiao Haixia Zhu Peipei Kang |
| author_facet | Gujun Cong Jingjing Shao Feng Xiao Haixia Zhu Peipei Kang |
| author_sort | Gujun Cong |
| collection | DOAJ |
| description | Abstract Background Endothelial cells are integral components of the tumor microenvironment and play a multifaceted role in tumor immunotherapy. Targeting endothelial cells and related signaling pathways can improve the effectiveness of immunotherapy by normalizing tumor blood vessels and promoting immune cell infiltration. However, to date, there have been no comprehensive studies analyzing the role of endothelial cells in the diagnosis and treatment of prostate adenocarcinoma (PRAD). Method By integrating clinical and transcriptomic data from TCGA-PRAD, we initially identified key endothelial cell-related genes in PRAD samples through single-cell analysis. Subsequently, cluster analysis was employed to classify PRAD samples based on the expression of these endothelial cell-related genes, allowing us to explore their correlation with patient prognosis and immunotherapy outcomes. A diagnostic model was then constructed and validated using a combination of 108 machine learning algorithms. The XGBoost and Random Forest algorithms highlighted the significant role of COL1A1, and we further analyzed the expression and correlation of COL1A1, AR, and EGFR through multiplex immunofluorescence staining. In vitro experimental analysis of the impact of COL1A1 on the progression of PRAD. Results Single-cell analysis identified 12 differential prognostic genes associated with endothelial cells. Cluster analysis confirmed a strong correlation between endothelial cell-related genes and both prostate cancer prognosis and immunotherapy responses. Diagnostic models developed using various machine learning techniques demonstrated the significant predictive capability of these 12 genes in the diagnosis of prostate cancer. Furthermore, based on patients’ prognostic information, multiple machine learning analyses highlighted the critical role of COL1A1. Immunofluorescence analysis results confirmed that COL1A1 is highly expressed in prostate cancer and is positively correlated with both AR and EGFR. In vitro experiments confirm that reducing COL1A1 expression levels can inhibit PRAD progression. Conclusion This study provides a comprehensive analysis of the role of endothelial cell-related genes in the diagnosis, prognosis, and immunotherapy of prostate cancer. The findings, supported by various machine learning algorithms and experimental results, highlight COL1A1 as a significant target for the diagnosis and immunotherapy of PRAD. |
| format | Article |
| id | doaj-art-586c092b1b1948ac91bfff06224eeeb7 |
| institution | Kabale University |
| issn | 1745-6150 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | Biology Direct |
| spelling | doaj-art-586c092b1b1948ac91bfff06224eeeb72025-01-12T12:11:25ZengBMCBiology Direct1745-61502025-01-0120111910.1186/s13062-024-00591-xThe role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysisGujun Cong0Jingjing Shao1Feng Xiao2Haixia Zhu3Peipei Kang4Clinical Laboratory Center, Affiliated Mental Health Center of Nantong University (Nantong Fourth People’s HospitalCancer Research Center Nantong, Affiliated Tumor Hospital of Nantong University & Nantong Tumor HospitalDepartment of Pathology, Affiliated Nantong Hospital 3 of Nantong University (Nantong Third People’s Hospital)Cancer Research Center Nantong, Affiliated Tumor Hospital of Nantong University & Nantong Tumor HospitalCancer Research Center Nantong, Affiliated Tumor Hospital of Nantong University & Nantong Tumor HospitalAbstract Background Endothelial cells are integral components of the tumor microenvironment and play a multifaceted role in tumor immunotherapy. Targeting endothelial cells and related signaling pathways can improve the effectiveness of immunotherapy by normalizing tumor blood vessels and promoting immune cell infiltration. However, to date, there have been no comprehensive studies analyzing the role of endothelial cells in the diagnosis and treatment of prostate adenocarcinoma (PRAD). Method By integrating clinical and transcriptomic data from TCGA-PRAD, we initially identified key endothelial cell-related genes in PRAD samples through single-cell analysis. Subsequently, cluster analysis was employed to classify PRAD samples based on the expression of these endothelial cell-related genes, allowing us to explore their correlation with patient prognosis and immunotherapy outcomes. A diagnostic model was then constructed and validated using a combination of 108 machine learning algorithms. The XGBoost and Random Forest algorithms highlighted the significant role of COL1A1, and we further analyzed the expression and correlation of COL1A1, AR, and EGFR through multiplex immunofluorescence staining. In vitro experimental analysis of the impact of COL1A1 on the progression of PRAD. Results Single-cell analysis identified 12 differential prognostic genes associated with endothelial cells. Cluster analysis confirmed a strong correlation between endothelial cell-related genes and both prostate cancer prognosis and immunotherapy responses. Diagnostic models developed using various machine learning techniques demonstrated the significant predictive capability of these 12 genes in the diagnosis of prostate cancer. Furthermore, based on patients’ prognostic information, multiple machine learning analyses highlighted the critical role of COL1A1. Immunofluorescence analysis results confirmed that COL1A1 is highly expressed in prostate cancer and is positively correlated with both AR and EGFR. In vitro experiments confirm that reducing COL1A1 expression levels can inhibit PRAD progression. Conclusion This study provides a comprehensive analysis of the role of endothelial cell-related genes in the diagnosis, prognosis, and immunotherapy of prostate cancer. The findings, supported by various machine learning algorithms and experimental results, highlight COL1A1 as a significant target for the diagnosis and immunotherapy of PRAD.https://doi.org/10.1186/s13062-024-00591-xSingle cell analysisMachine learningPRADEndothelial cellCOL1A1 |
| spellingShingle | Gujun Cong Jingjing Shao Feng Xiao Haixia Zhu Peipei Kang The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis Biology Direct Single cell analysis Machine learning PRAD Endothelial cell COL1A1 |
| title | The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis |
| title_full | The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis |
| title_fullStr | The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis |
| title_full_unstemmed | The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis |
| title_short | The role of endothelial cell-related gene COL1A1 in prostate cancer diagnosis and immunotherapy: insights from machine learning and single-cell analysis |
| title_sort | role of endothelial cell related gene col1a1 in prostate cancer diagnosis and immunotherapy insights from machine learning and single cell analysis |
| topic | Single cell analysis Machine learning PRAD Endothelial cell COL1A1 |
| url | https://doi.org/10.1186/s13062-024-00591-x |
| work_keys_str_mv | AT gujuncong theroleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT jingjingshao theroleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT fengxiao theroleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT haixiazhu theroleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT peipeikang theroleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT gujuncong roleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT jingjingshao roleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT fengxiao roleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT haixiazhu roleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis AT peipeikang roleofendothelialcellrelatedgenecol1a1inprostatecancerdiagnosisandimmunotherapyinsightsfrommachinelearningandsinglecellanalysis |