Integrating machine learning and structural analysis to decipher benzo[a]pyrene-induced bladder cancer networks

Area of Science:

• Environmental Health
• Oncology
• Toxicology

Background:

• Benzo[a]pyrene (BaP) is a polycyclic aromatic hydrocarbon found in tobacco smoke and pollutants.
• BaP exposure is a known risk factor for bladder cancer (BLCA).

Purpose of the Study:

• To identify genes differentially expressed due to BaP exposure in bladder cancer.
• To elucidate the molecular mechanisms underlying BaP-induced bladder carcinogenesis.

Main Methods:

• Systematic analysis of Gene Expression Omnibus (GEO) datasets.
• Integration of network toxicology, machine learning (SHAP analysis), molecular docking, molecular dynamics, and single-cell transcriptomics.
• Prioritization of key BaP-related differentially expressed genes (DEBRGs).

Main Results:

• 19 significant DEBRGs were identified, with 7 key genes prioritized: GSK3B, SKP2, AURKB, EPHB4, KIT, NR3C2, and CA2.
• GSK3B and SKP2 were highlighted by SHAP analysis as crucial for predictive modeling of BaP-mediated BLCA.
• Single-cell transcriptomics revealed cell-type specific expression of these genes within the tumor microenvironment.
• Molecular simulations provided insights into BaP-protein interactions.

Conclusions:

• This study identifies critical genes involved in BaP-induced bladder cancer.
• The findings offer potential therapeutic targets and deepen the understanding of BaP's molecular carcinogenesis pathways.