« Back to Farmacia Journal 1/2025

THE USE OF ARTIFICIAL INTELLIGENCE IN DRUG DISCOVERY: FROM CONCEPT TO CURRENT APPLICATIONS

ANDREEA PUȘCAȘU ^1,2, FLORENTINA GHERGHICEANU ³*, OCTAVIAN ANDRONIC 2

Download Full Article PDF

Integrating Artificial Intelligence (AI) in drug discovery has revolutionised traditional pharmaceutical development processes, offering significant improvements in efficiency, cost reduction, and success rates. Drug discovery, a traditionally labour-intensive and costly process, encompasses five stages: target identification, drug discovery, preclinical studies, clinical trials, and regulatory approval. With failure rates as high as 97% in clinical trials, particularly in oncology, innovation is critical. AI, including machine learning (ML) and deep learning (DL), has emerged as a transformative tool across all stages of drug development. AI applications include advanced data collection, molecular structure representation, target prediction, drug-target interaction analysis, and de novo drug design. Tools such as AlphaFold, DeepDTA, and DrugGPT demonstrate AI's capabilities in protein structure prediction, drug-binding affinity analysis, and ligand design. Moreover, AI's ability to predict drug-drug interactions, optimise pharmacokinetics (ADMET), and identify novel compounds accelerates drug discovery while reducing reliance on traditional experimental methods. Despite its promise, AI faces challenges such as ethical concerns, data quality issues, and algorithmic biases. Current applications in neurology, oncology, and antimicrobial resistance underscore AI's potential, exemplified by innovations like SyntheMol for antibiotic synthesis and AI-driven Alzheimer’s treatments. This review highlights AI's capacity to reshape drug discovery, emphasising its advantages, current implementations, and the need to address its limitations to fully leverage its transformative potential.