Generative AI for drug discovery: Accelerating molecular design with deep learning using Nigerian local content

Abstract

This research explores how Generative Artificial Intelligence (AI) can be used to accelerate drug discovery, especially in developing nations like Nigeria. By integrating various generative models; including GANs, VAEs, and Transformer-based architectures—the study aims to rapidly create new molecular structures with therapeutic potential. A unique aspect of this research is its use of local Nigerian resources, such as indigenous medicinal plants and traditional knowledge, to create a specialized dataset. By combining this local data with global molecular databases, the framework is designed to find candidate molecules with better drug-likeness, lower toxicity, and higher binding affinity to target proteins. This approach not only speeds up the preclinical phase of drug discovery but also promotes sustainable healthcare innovation by utilizing Nigeria’s own resources. The study highlights its potential application in finding treatments for malaria, sickle cell disease, and antimicrobial resistance—all major health concerns in Nigeria.