AI in Drug Discovery

Artificial intelligence (AI) is revolutionizing the fields of bioinformatics, biotechnology, and drug discovery. Drug discovery is a time-consuming and costly process, often hampered by high failure rates at various stages of development. AI technologies, including machine learning, deep learning, and natural language processing, promise to address these challenges in many different ways, from enabling the analysis of vast datasets to designing potential drug candidates —both small molecules and biologics — more efficiently than ever before.

AI-powered bioinformatics tools have the potential to predict molecular properties, optimize drug combinations, identify novel targets, generate new molecules and proteins, and personalize therapeutic interventions. By leveraging AI, researchers can explore large chemical spaces, analyse genomic data for target identification, and model complex biological systems to simulate drug responses at an unprecedented scale and speed. These advancements are not only accelerating the drug discovery process but also significantly reducing costs and improving the likelihood of successful outcomes.

The application of AI in drug discovery is beginning to lead to breakthroughs in understanding disease mechanisms, discovering biomarkers, and developing precision medicine strategies.

The aim of this research topic is to explore the latest developments and applications of AI in drug discovery, identify best practices, and identify their limitations. We seek to highlight innovative original research, methodologies, technology and code, case studies, reviews, and perspectives that utilize AI to enhance the drug discovery pipeline, ultimately seeking to improve the efficacy, efficiency, and success rates of drug development and carry out an honest appraisal of the state of AI in drug discovery.

We invite researchers, practitioners, and industry professionals to contribute articles which might explore but are not limited to:
1. AI-Driven Drug Target Identification: Techniques and applications of AI for identifying novel drug targets and understanding their biological roles using genomic and proteomic data, as well as scientific literature.
2. AI in Virtual Screening and Molecular Docking: Innovations in computational methods for screening large libraries of compounds and understanding their interactions with biological macromolecules; active learning; co-folding.
3. Deep Learning in Quantitative Structure-Activity Relationship (QSAR) and Quantitative Structure-Property Relationship (QSPR) Modelling: Applications of deep learning models to analyse and leverage molecular activity data, predict bioactivity and ADME-Tox (Absorption, Distribution, Metabolism, Excretion, & Toxicity), and guide molecular optimization.
4. AI for Drug Repurposing: Strategies that utilize AI and machine learning to discover new therapeutic uses for existing drugs by analysing genetic, pharmacological, and clinical data.
5. Biomarker Discovery through AI: Methods leveraging AI to identify predictive biomarkers for disease progression and drug response, aiding in precision medicine.
6. Integrative AI Approaches for Multi-Omics Data Analysis: Approaches that integrate various -omics datasets (e.g., genomics, transcriptomics, proteomics) using AI to gain holistic insights into complex biological systems for drug discovery.
7. Generative AI for Molecular Design: Structure-based and ligand-based methods to design novel small molecules, molecular glues, peptidomimetics, biologics, including antibodies and nanobodies, and other modalities, such as mRNA vaccine design and delivery.
8. AI for Medicinal Chemistry: including novel methods for reaction and yield prediction, retrosynthesis, synthetic route optimization, and automated synthesis.
9. AI Benchmarking for Drug Discovery: Best practices in the construction of benchmarks, including automated high-throughput experimentation; high-quality training data sets; best practices in model comparison, including statistically sound evaluation.
10. Advances in Synthetic Data Preparation: Advances in the development of new models using synthetic data generated by the application of computational chemistry, including quantum mechanics; biomolecular simulation, including molecular dynamics and free energy perturbation methods; and course-grained methods.
11. Agentic AI and Large Language Models: Advances in the application of Large Language Models (LLMs) combined with computational tools (“agents”), novel agentic frameworks, workflows, and applications.
12. Explainable AI: new methods and applications of explainable AI, interpretable models.

This research topic aims to bring together diverse perspectives and expertise to foster collaborations and advance the understanding of AI’s transformative role in drug discovery. We encourage interdisciplinary submissions that bridge computational sciences with biological and pharmacological research.

Topic Editor Nathan Brown is currently employed by Healx Ltd.

Keywords: Artificial Intelligence, Drug Discovery, Machine Learning, Deep Learning, Precision Medicine, Biomarkers, Molecular Docking, Structure-Activity, LLM, Medical Chemistry, Synthetic Data

Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.