Related Experiment Video
Updated: Jul 17, 2025
03:37
Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
Published on: March 1, 2024
798
[Keloid nomogram prediction model based on weighted gene co-expression network analysis and machine learning].
Zhengyu Li1, Baohua Tian1, Haixia Liang1
1College of Biomedical Engineering, Taiyuan University of Technology, Taiyuan 030024, P. R. China.
Summary
This study identifies four key genes (HGF, SDC4, ENPP2, RND3) as diagnostic markers for keloids. A nomogram prediction model was developed for early keloid risk assessment in trauma patients.
Area of Science:
- Dermatology and Bioinformatics
- Oncology
- Molecular Biology
Context:
- Keloids, benign skin tumors from excessive connective tissue proliferation, pose challenges in management.
- Accurate keloid risk prediction and early diagnosis are crucial for effective patient care.
- Existing diagnostic methods require improvement for timely intervention.
Purpose:
- To identify reliable diagnostic markers for keloids.
- To develop a predictive nomogram model for keloid risk assessment.
- To explore potential biological pathways involved in keloid development.
Summary:
- Analyzed four keloid gene expression datasets using WGCNA, differential expression, and network analysis to identify 37 core genes.
- Applied LASSO and SVM-RFE machine learning algorithms to pinpoint four diagnostic markers: HGF, SDC4, ENPP2, and RND3.
- Constructed and validated a nomogram prediction model demonstrating high accuracy, calibration, and clinical utility.
Impact:
- The developed nomogram model shows significant potential for early clinical diagnosis of keloids.
- Identified key genes provide insights into keloid pathogenesis.
- The model offers an improved strategy for keloid risk stratification and management.
