Artificial Intelligence: A Modern Approach towards Drug Discovery

This review offers an overview of how artificial intelligence (AI) is advancing and expediting the drug discovery process. AI methods utilize large data from different databases like PubChem, ChEMBL, DrugBank, ZINC, SIDER, and COCONUT, offering critical insights into chemical compounds and drug-target interactions. These datasets, enriched with experimental and high-throughput screening data, are used to create predictive models for various drug development stages. AI techniques such as deep learning (DL), machine learning (ML), and natural language processing (NLP) are proving valuable in tasks like virtual screening, de novo drug design, target identification, and toxicity prediction. Data representations like SMILES, SELFIES, InChI, and DeepSMILES are used to model chemical structures effectively. Furthermore, advanced AI models such as generative adversarial networks (GANs), Recurrent neural network (RNN) and graph neural networks (GNNs) are enhancing drug discovery by creating novel molecular structures and predicting complex molecular interactions. AI is revolutionizing drug discovery, enabling faster, more efficient, and more accurate development of new therapies.