Artificial Intelligence in Pharmaceutical in Drug Discovery: From Structure Prediction to Cadidate Generation of Novel Therapeutics
Keywords:
Artificial Intelligence (AI), Drug Discovery, Protein Structure Prediction, Machine Learning and Deep Learning Generative Drug Design, ADMET PredictionAbstract
The incorporation of artificial intelligence (AI) into drug discovery has fundamentally transformed the pharmaceutical landscape, creating exceptional prospects for expediting the development of novel therapeutics through the pathway from structure prediction to candidate generation. The flexibility and forecasting strength of AI have been effectively utilized throughout the drug discovery pipeline, encompassing target identification, accurate target structure prediction with Alpha Fold and Rose TTA Fold, virtual screening, compound screening, drug-target interaction analysis, and lead optimization for candidate generation. Through the application of machine learning, deep learning, and large-scale biological data, scientists can accelerate the identification of promising molecules, accurately predict three-dimensional protein structures, design novel therapeutic candidates via de novo and generative approaches, minimize human effort, and enhance therapeutic efficacy. This review presents a detailed overview of AI's contribution to drug discovery from structure prediction to generation of novel therapeutic candidates, emphasizing its applications in binding affinity prediction, ADMET evaluation, its benefits, and existing obstacles. It also outlines the present status of AI-driven drug discovery, its transformative potential, and key limitations to be resolved. Moreover, it examines forthcoming perspectives of AI in pharmaceutical innovation, including requirements for high-quality datasets, standardization, personalized medicine, and regulatory approval for novel therapeutic.

.