Generative AI for the Design of Molecules: Advances and Challenges
Journal of Chemical Information and Modeling · 08.12.2025 · 65 (23) · 12668-12690
DOI: 10.1021/acs.jcim.5c02234 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 18.11.2025. Verfasst von Yan Sun und 16 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biochemistry, University of Western Ontario, London, Ontario N6A 5C1, Canada. Hauptthemen laut MEDLINE: Artificial Intelligence, Drug Design. Die Arbeit wurde bisher 27-mal zitiert (Crossref) und nennt 210 Literaturangaben. Gefördert von: Canadian Institutes of Health Research, Canada Foundation for Innovation, Canada Research Chairs.
Abstract (Originalsprache)
The design of novel molecules underpins advances in both drug discovery and biomaterials engineering. Traditional approaches, from natural product isolation to high-throughput screening, have delivered important therapeutics but remain costly, inefficient, and limited in exploring the chemical and biomolecular space. While predictive machine learning models improved aspects of discovery, they cannot fully address the complexity of modern precision medicine. Generative artificial intelligence (AI) offers a paradigm shift by enabling de novo molecular creation guided by data-driven optimization. Architectures such as variational autoencoders, generative adversarial networks, normalizing flows, and diffusion models now demonstrate unprecedented capabilities in designing small molecules and macromolecules that satisfy complex physicochemical and biological requirements. This review surveys the rapidly evolving field of generative AI for molecular design. We first introduce the development of generative architectures and optimization strategies, focusing on how sampling, training, and postgeneration techniques improve control over molecular design. We then examine applications across molecular representations, unconstrained and property-constrained design, conformation modeling, and the generation of large biomolecules such as proteins, antibodies, and peptides. Benchmarking datasets, evaluation metrics, and real-world case studies, such as the AI-driven discovery of novel antibiotics demonstrated in vivo efficacy against multidrug-resistant infections, illustrate the growing maturity and translational potential of generative molecular design approaches. Despite rapid advances, generative molecular design still faces critical challenges that point to key future directions. These include integrating physicochemical priors through differentiable physical models, overcoming data scarcity via synthetic augmentation and transfer learning, enabling multimodal fusion of structural, omics, and phenotypic data, deploying autonomous AI agents for adaptive decision-making, and optimizing multiple objectives with uncertainty-aware strategies. Addressing these challenges could lead to more robust, generalizable, and experimentally aligned molecular design systems.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Biochemistry, University of Western Ontario, London, Ontario N6A 5C1, Canada.
- Department of Computer Science, University of Western Ontario, London, Ontario N6A 3K7, Canada.
- Department of Mechanical & Materials Engineering, University of Western Ontario, London, Ontario N6A 5B9, Canada.
- Department of Statistical and Actuarial Sciences, University of Western Ontario, London, Ontario N6A 5B7, Canada.
- Department of Chemistry, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada.
- Department of Materials Science and Engineering, McMaster University, Hamilton, Ontario L8S 4L7, Canada.
ORCID-Profile: Anna Klinkova, Kyla Sask, Boyu Wang, Elizabeth R Gillies, HaoTian Harvey Shi, Pingzhao Hu
Förderung
- Canadian Institutes of Health Research
- Canada Foundation for Innovation
- Canada Research Chairs
- Natural Sciences and Engineering Research Council of Canada
Zitieren
APA
Sun, Y., Chen, L., Jing, Z., Li, Y. Y., Kim, D., Gao, J. Y., Noroozi, R., Yi, G. Y., Tetsassi Feugmo, C. G., Klinkova, A., Sask, K., Kristiadi, A., Wang, B., Gillies, E. R., Lu, K. P., Shi, H. H., & Hu, P. (2025). Generative AI for the Design of Molecules: Advances and Challenges. Journal of Chemical Information and Modeling, 65(23), 12668-12690. https://doi.org/10.1021/acs.jcim.5c02234
Vancouver
Sun Y, Chen L, Jing Z, Li YY, Kim D, Gao JY, et al. Generative AI for the Design of Molecules: Advances and Challenges. Journal of Chemical Information and Modeling. 2025;65(23):12668-12690. doi:10.1021/acs.jcim.5c02234
BibTeX
@article{Sun2025,
author = {Sun, Yan and Chen, Lianghong and Jing, Zihao and Li, Yan Yi and Kim, Dongkyu and Gao, Jing-Yan and Noroozi, Reza and Yi, Grace Y and Tetsassi Feugmo, Conrard Giresse and Klinkova, Anna and Sask, Kyla and Kristiadi, Agustinus and Wang, Boyu and Gillies, Elizabeth R and Lu, Kun Ping and Shi, HaoTian Harvey and Hu, Pingzhao},
title = {Generative AI for the Design of Molecules: Advances and Challenges},
journal = {Journal of Chemical Information and Modeling},
year = {2025},
volume = {65},
number = {23},
pages = {12668--12690},
doi = {10.1021/acs.jcim.5c02234}
}
EUVIMED