PharmAgent improves molecule design under multiple chemical constraints
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization
Artificial Intelligence
Summary
Designing new molecules that work well and meet safety or property rules is hard and costly. The researchers developed PharmAgent, a method that helps find better molecules by learning how to handle multiple rules and goals step by step. It uses a special controller to track and balance these rules while exploring new molecules efficiently. Their tests show PharmAgent performs better than other advanced methods in both finding good target molecules and respecting property limits.
What this means in practice
- •For pharmaceutical research teams: Optimize molecules faster while ensuring they meet multiple safety and effectiveness criteria within limited testing budgets.
- •For chemical product development teams: Improve chemical design processes by balancing desired active properties with manufacturability and regulatory constraints automatically.
Authors
Nihui Shao, Guanxing Chen, Jilong Shi, Zhengyang Bai, Haohuai He, Zhenchao Tang, Qiujie Lv, Yu-An Huang, Zhi-An Huang
Abstract
Molecular optimization must improve target activity and satisfy developability constraints within limited evaluation budgets. Classical methods require tailored rules or training to incorporate chemical instructions and property feedback. Frozen language models can condition edits on this information, but need explicit constraint control and relevant experience. We therefore present PharmAgent, a constraint-aware molecular search method driven by adaptive external state. Its Lagrangian controller translates violations in accepted states into accumulated constraint pressure, keeping this history separate from current property measurements. Structure-indexed replay complements this feedback with relevant evaluated transitions that guide subsequent proposals. As a curriculum progressively activates constraints, candidates and the incumbent are compared under the same current objective, and the accepted state determines the next multiplier update. We derive an exact identity that characterizes how accepted-state violations accumulate in the controller's multipliers. Across five tasks with five independent runs, PharmAgent achieves a summed area under the target-score curves (AUC) of 3.9208 in target-only search, improving over MOLLEO by 37.3%. With online constraints, it achieves a property-adjusted AUC of 0.7076, improving over the strongest online baseline, ExLLM, by 53.8%. These results rank first among all evaluated methods in both target-only and constraint-aware search. The online comparison covers all five baseline frameworks. The full system leads every ablation variant in target quality, property-adjusted performance, and Pareto hypervolume. All five molecular cases reach feasible final states, documenting target gains and trade-offs.