Classification prediction of early pulmonary nodes based on weighted gene correlation network analysis and machine...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Lung adenocarcinoma (LUAD) classification is crucial for early diagnosis and treatment.
• Identifying reliable biomarkers for early-stage LUAD remains a challenge.

Purpose of the Study:

• To develop a predictive model for early pulmonary nodule classification using gene expression data.
• To identify key gene signatures associated with LUAD classification.

Main Methods:

• Differential gene expression analysis (DEGs) using Dseq2, Limma, and EdgeR.
• Weighted Gene Correlation Network Analysis (WGCNA) to identify relevant gene modules and key genes.
• Machine learning algorithms, including LASSO regression and XGBoost, for predictive model development.

Main Results:

• 1306 differentially expressed genes (DEGs) were identified in LUAD.
• WGCNA highlighted 116 genes significantly related to LUAD classification, enriched in 14 KEGG pathways.
• Six gene signatures (HMGB3, ARHGAP6, TCF21, FCN3, COL6A6, GOLM1) were identified through integrated analysis for early pulmonary nodule classification.

Conclusions:

• A panel of six genes can assist clinicians in predicting early-stage LUAD classification.
• The integrated approach of DEGs, WGCNA, and machine learning provides a robust method for biomarker discovery.