Classification of drug molecules for oxidative stress signalling pathway

Area of Science:

• Biochemistry and Molecular Biology
• Computational Biology and Bioinformatics
• Pharmacology and Toxicology

Background:

• Oxidative stress is implicated in numerous human diseases, including diabetes, cancer, hypertension, Alzheimer's disease, and heart failure.
• The Nrf2-antioxidant response element (ARE) signaling pathway is a key cellular defense mechanism against oxidative stress.

Purpose of the Study:

• To compute activity, efficacy, and potency scores for the ARE signaling pathway.
• To propose a multi-level prediction scheme for ARE pathway activity to improve understanding and management of oxidative stress.

Main Methods:

• Utilized knowledge discovery from data (KDD) principles.
• Applied machine learning techniques to develop a multi-level prediction scheme.
• Validated the proposed scheme using K-fold cross-validation.

Main Results:

• Achieved a 90% accuracy in predicting the activity score for ARE molecules.
• The prediction of activity scores indicates the molecules' potential to mitigate oxidative stress.

Conclusions:

• The developed multi-level prediction scheme effectively assesses ARE pathway activity.
• This approach offers a promising tool for identifying compounds that can combat oxidative stress and related diseases.