Hybrid Computational Framework Integrating Ensemble Learning, Molecular Docking, and Dynamics for Predicting...

Area of Science:

• Computational chemistry and pharmacology
• Drug discovery and development
• Parasitology

Background:

• Drug-resistant Plasmodium falciparum strains necessitate new antimalarial therapies.
• Existing antimalarial drug discovery pipelines require optimization for efficiency.

Purpose of the Study:

• To develop and validate an integrative computational framework for predicting antimalarial activity.
• To identify potent antimalarial compounds from the Malaria Box database.
• To elucidate potential mechanisms of action for identified compounds.

Main Methods:

• Ensemble machine learning models (regression and classification) using topographical and quantum mechanical descriptors.
• Molecular docking against Plasmodium falciparum Cytochrome B.
• Molecular dynamics simulations and binding free energy calculations.

Main Results:

• Ensemble classifiers achieved robust performance in categorizing compounds as active or very active.
• Subsequent regression models within classes showed high predictive accuracy (Q² > 0.79).
• Molecular docking and dynamics identified strong binding affinities and stable complexes, highlighting key interactions with P. falciparum Cytochrome B.

Conclusions:

• The integrative computational framework effectively predicts antimalarial activity and aids in identifying promising drug candidates.
• The study provides mechanistic insights into compound-target interactions, supporting future experimental validation.
• This approach accelerates the discovery of novel antimalarial agents against drug-resistant malaria.