Predicting oxidative stress induced by organic chemicals by using quantitative Structure-Activity relationship...

Area of Science:

• Toxicology and Cheminformatics
• Environmental Science
• Computational Chemistry

Background:

• Xenobiotic compounds can cause cellular oxidative stress, leading to DNA and other damages.
• Predicting which chemicals induce oxidative stress and their toxicity is crucial for safety assessments.

Purpose of the Study:

• To develop quantitative structure-activity relationship (QSAR) models for predicting chemical activation of the ARE pathway.
• To identify compounds that induce oxidative stress and exert toxic effects on cells.
• To provide a basis for environmental risk assessment of chemicals.

Main Methods:

• Investigated 4270 compounds using the ARE-bla assay.
• Employed recursive partitioning (RP) and binomial logistic regression for prediction.
• Developed QSAR models to correlate chemical structures with activity and toxicity.

Main Results:

• A logistic regression model predicted chemical activity with 69.1% accuracy (AUROC 0.762).
• Number of multiple bonds (nBM) and hydrogen percentage (H%) were key predictive parameters.
• A QSAR model for carbon chain compounds showed strong predictive performance (R²=0.722, R²ext=0.798).

Conclusions:

• Simple predictive models can identify compounds inducing oxidative stress.
• These models support the scientific basis for environmental risk assessment.
• Further development is needed for global QSAR toxicity prediction.