Related Experiment Video
Updated: Jun 24, 2025
06:08
A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
155
Prognosis and diagnosis of prostate cancer based on hypergraph regularization sparse least partial squares regression
Ruo-Hui Huang1, Zi-Lu Ge2, Gang Xu1
1Department of Urology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China.
Aging
|June 3, 2024
Summary
This study integrated gene expression and DNA methylation data to identify prostate cancer (PCa) biomarkers. A diagnostic model using 10 DNA methylation sites and a prognostic model using 7 mRNAs were developed, showing high accuracy in diagnosing PCa and predicting survival.
Area of Science:
- Oncology
- Bioinformatics
- Genomics
Background:
- Prostate cancer (PCa) incidence is rising globally.
- Identifying reliable diagnostic and prognostic biomarkers for PCa is crucial.
- Current diagnostic and prognostic methods require improvement.
Purpose of the Study:
- To identify novel candidate biomarkers for prostate cancer (PCa) diagnosis and prognosis.
- To integrate gene expression and DNA methylation data for enhanced biomarker discovery.
- To develop accurate diagnostic and prognostic models for PCa.
Main Methods:
- A sparse partial least squares regression algorithm with hypergraph regularization (HR-SPLS) was employed.
- Data integration and clustering of gene expression (mRNAs) and DNA methylation data.
- Machine learning methods were used to construct diagnostic and prognostic models.
Main Results:
- A diagnostic model comprising 10 DNA methylation sites demonstrated high accuracy in PCa detection.
- A prognostic model based on 7 mRNAs accurately predicted disease-free survival (AUC = 0.761).
- Gene Set Enrichment Analysis and immune analysis suggested a link between patient prognosis and immune cell infiltration.
Conclusions:
- The study successfully identified potential diagnostic and prognostic biomarkers for PCa.
- Integrating DNA methylation and gene expression data offers a promising approach for biomarker discovery.
- Findings may lead to improved PCa management strategies.
