Exploring Antiviral Drugs on Monolayer Black Phosphorene: Atomistic Theory and Explainable Machine Learning-Assisted...

Area of Science:

• Materials Science
• Computational Chemistry
• Drug Discovery

Background:

• Favipiravir (FP) and ebselen (EB) are antiviral drugs with broad efficacy.
• Understanding drug-material interactions is crucial for developing new antiviral therapies.
• Phosphorene, a 2D material, offers unique electronic properties for potential applications.

Purpose of the Study:

• To elucidate the binding properties of favipiravir and ebselen on a phosphorene single-layer.
• To investigate the interaction characteristics and thermodynamic properties of the antiviral-phosphorene complex.
• To assess the potential of this 2D complex for drug adsorption and antiviral applications.

Main Methods:

• Utilized molecular dynamics (MD) simulations and van der Waals density functional theory (DFT).
• Employed four machine learning models (Random Forest, Gradient Boosting, XGBoost, CatBoost) to train the Hamiltonian of antiviral molecules on phosphorene.
• Applied SHAP (SHapley Additive exPlanations) for model interpretability.

Main Results:

• Accurately elucidated the binding properties of favipiravir and ebselen on a phosphorene single-layer.
• The functionalized 2D phosphorene complex demonstrated robust thermostability.
• Variations in free energy indicated adsorption potential of FP and EB molecules under different conditions.

Conclusions:

• The antiviral-phosphorene complex exhibits significant thermostability and potential for drug adsorption.
• Machine learning models accurately approximate DFT calculations for drug design.
• This study provides insights into the development of novel antiviral drug delivery systems using 2D materials.