Towards better modelling of drug-loading in solid lipid nanoparticles: Molecular dynamics, docking experiments and...

Area of Science:

• Computational chemistry and materials science.
• Application of artificial intelligence in pharmaceutical research.

Background:

• Solid lipid nanoparticles (SLNs) are crucial drug delivery systems.
• Accurate prediction of drug loading capacity in SLNs is essential for formulation development.

Purpose of the Study:

• To develop a computational model for predicting drug mass loaded into tripalmitin-based SLNs.
• To utilize machine learning to correlate drug properties with molecular docking binding energies.

Main Methods:

• Molecular docking simulations of various drugs onto tripalmitin matrices using MOE® and GROMACS®.
• Application of Gaussian processes, a supervised machine learning technique.
• Correlation of drug descriptors (molecular weight, xLogP, TPSA, fragment complexity) with binding energies.

Main Results:

• A novel computational approach for drug loading prediction in SLNs.
• Accurate estimation of loaded drug mass by analyzing drug chemical structures.
• Achieved lower percentage bias compared to existing methods.

Conclusions:

• Gaussian processes effectively model the relationship between drug descriptors and binding energies.
• This AI-driven method enables precise prediction of drug loading in SLNs.
• The approach offers a valuable tool for pharmaceutical formulation and drug delivery research.