From Generative Molecular Design to Agentic Artificial Intelligence in Drug Discovery
- 1Department of Life Science Informatics and Data Science B‐IT, LIMES Program Unit Chemical Biology and Medicinal Chemistry Rheinische Friedrich‐Wilhelms Universität Bonn Germany
- 2Lamarr Institute for Machine Learning and Artificial Intelligence Rheinische Friedrich‐Wilhelms‐Universität Bonn Bonn Germany
- 3Data Science Center Nara Institute of Science and Technology Ikoma Nara Japan
Abstract
Over the past decade, artificial intelligence (AI) has increasingly influenced the early stages of drug discovery, which focus primarily on target evaluation and the design or identification of new active small molecules or biologics. Deep neural networks adapted from areas such as natural language processing and computer vision have provided novel opportunities for generative modeling and human–machine interactions. State‐of‐the‐art AI models for generative molecular design have emerged, but remain only little understood so far. Furthermore, new computational systems are being devised rapidly that are expected to significantly transform the drug discovery landscape in the near future. These developments pose challenges to computer‐aided drug discovery in various ways. In the following, we discuss these opportunities and challenges, with a focus on molecular design.
Related
- Integrating artificial intelligence and molecular simulation for peptide discovery: limitations, validation and decision-centered workflows
- Accelerating Catalyst Discovery with Artificial Intelligence Using Molecular Dual-Atom Site Catalysts as a Case Study
- GEOMIND: a hybrid generative artificial intelligence model for geopolymer design and optimization
- Sampling a trillion-sized GDB-20 database of drug-like molecules by generative artificial intelligence.
- Generative and Foundation-Model Artificial Intelligence in Food Safety Analysis, Authenticity Identification, and Quality Control: A Systematic Review of Validation and Decision Readiness.
BibTeX
@article{mastropietro2026from,
title = {{From Generative Molecular Design to Agentic Artificial Intelligence in Drug Discovery}},
author = {Andrea Mastropietro and Jürgen Bajorath},
journal = {Chemmedchem},
year = {2026},
month = nov,
doi = {10.1002/cmdc.70564},
url = {https://doi.org/10.1002/cmdc.70564},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - From Generative Molecular Design to Agentic Artificial Intelligence in Drug Discovery
AU - Andrea Mastropietro
AU - Jürgen Bajorath
PY - 2026
DA - 2026/11/05
JO - Chemmedchem
DO - 10.1002/cmdc.70564
UR - https://doi.org/10.1002/cmdc.70564
AB - Over the past decade, artificial intelligence (AI) has increasingly influenced the early stages of drug discovery, which focus primarily on target evaluation and the design or identification of new active small molecules or biologics. Deep neural networks adapted from areas such as natural language processing and computer vision have provided novel opportunities for generative modeling and human–machine interactions. State‐of‐the‐art AI models for generative molecular design have emerged, but remain only little understood so far. Furthermore, new computational systems are being devised rapidly that are expected to significantly transform the drug discovery landscape in the near future. These developments pose challenges to computer‐aided drug discovery in various ways. In the following, we discuss these opportunities and challenges, with a focus on molecular design.
SN - 1860-7179
SN - 1860-7187
ER -