Bridging the phenotype-target gap for molecular generation via multi-objective reinforcement learning
Bioinformatics · 01.07.2026 · 42 (Supplement_1)
DOI: 10.1093/bioinformatics/btag242 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 07.07.2026. Verfasst von Haotian Guo und 1 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Information Engineering, Nanjing Tech University, Nanjing, Jiangsu 211800, China. Hauptthemen laut MEDLINE: Drug Design, Computational Biology, Drug Discovery. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province.
Abstract (Originalsprache)
Motivation The generation of high-quality candidate molecules remains a central challenge in AI-driven drug design. Current phenotype-based and target-based strategies each suffer limitations, either incurring high experimental costs or overlooking system-level cellular responses. To bridge this gap, we propose XMolRL, a novel generative framework that synergistically integrates phenotypic and target-specific cues for de novo molecular generation. Results The phenotype-guided generator is first pretrained on expansive drug-induced transcriptional profiles and subsequently fine-tuned via multi-objective reinforcement learning (RL). Crucially, the reward function fuses docking affinity and drug-likeness scores, augmented with ranking loss, prior-likelihood regularization, and entropy maximization. The multi-objective RL steers the model toward chemotypes that are simultaneously potent, diverse, and aligned with the specified phenotypic effects. Extensive experiments demonstrate XMolRL’s superior performance over state-of-the-art phenotype-based and target-based models across multiple well-characterized targets. Our generated molecules exhibit favorable drug-like properties, high target affinity, and inhibitory potency (IC50) against cancer cells. This unified framework showcases the synergistic potential of combining phenotype-guided and target-aware strategies, offering a more effective solution for de novo drug discovery. Availability and implementation The source code and datasets are available at: https://github.com/hliulab/XMolRL. The archived version of the source code and test data can be found at: https://doi.org/10.5281/zenodo.19607680.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- College of Computer and Information Engineering, Nanjing Tech University, Nanjing, Jiangsu 211800, China.
ORCID-Profile: Hui Liu
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Jiangsu Province
Zitieren
APA
Guo, H., & Liu, H. (2026). Bridging the phenotype-target gap for molecular generation via multi-objective reinforcement learning. Bioinformatics, 42(Supplement_1). https://doi.org/10.1093/bioinformatics/btag242
Vancouver
Guo H, Liu H. Bridging the phenotype-target gap for molecular generation via multi-objective reinforcement learning. Bioinformatics. 2026;42(Supplement_1). doi:10.1093/bioinformatics/btag242
BibTeX
@article{Guo2026,
author = {Guo, Haotian and Liu, Hui},
title = {Bridging the phenotype-target gap for molecular generation via multi-objective reinforcement learning},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {Supplement_1},
doi = {10.1093/bioinformatics/btag242}
}
EUVIMED