Modified logistic regression models using gene coexpression and clinical features to predict prostate cancer...

Area of Science:

• Oncology
• Bioinformatics
• Genomics

Background:

• Prostate cancer progression prediction remains a significant challenge.
• Integrating gene expression data with clinical features may enhance prediction accuracy.

Purpose of the Study:

• To develop and validate a logistic regression (LR) model combining clinical features and gene co-expression data for improved prostate cancer progression prediction.
• To evaluate the efficacy of the Top-Scoring Pair (TSP) method for gene selection in prognostic models.

Main Methods:

• Applied a logistic regression (LR) model incorporating clinical variables and gene co-expression data.
• Utilized the Top-Scoring Pair (TSP) method for selecting relevant genes.
• Employed iterative cross-validation for statistical inference and model validation.

Main Results:

• LR models incorporating TSP-selected genes demonstrated superior accuracy in predicting prostate cancer progression compared to models with only clinical variables or genes selected via a one-gene-at-a-time approach.
• The TSP method effectively selected informative genes for prognostic modeling.

Conclusions:

• The Top-Scoring Pair (TSP) selection method is a valuable tool for feature selection in prognostic models for prostate cancer.
• The proposed LR model offers a promising alternative for predicting prostate cancer progression.