Molecular De Novo Design through Deep Reinforcement Learning

2017/04/25 by Marcus Olivecrona, Thomas Blaschke, Olivecrona, Marcus +5 · 22 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Materials Science · #Artificial Intelligence (cs.AI) #Chemical Synthesis and Analysis #Computational Drug Discovery Methods #FOS: Computer and information sciences #Machine Learning in Materials Science

paper · pdf · doi:10.48550/arxiv.1704.07555

openalex publication_date 2017/04/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This work introduces a method to tune a sequence-based generative model for molecular de novo design that through augmented episodic likelihood can learn to generate structures with certain specified desirable properties. We demonstrate how this model can execute a range of tasks such as generating analogues to a query structure and generating compounds predicted to be active against a biological target. As a proof of principle, the model is first trained to generate molecules that do not contain sulphur. As a second example, the model is trained to generate analogues to the drug Celecoxib, a technique that could be used for scaffold hopping or library expansion starting from a single molecule. Finally, when tuning the model towards generating compounds predicted to be active against the dopamine receptor type 2, the model generates structures of which more than 95% are predicted to be active, including experimentally confirmed actives that have not been included in either the generative model nor the activity prediction model.

Citations

• Recurrent Neural Network Regularization
• SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient
• Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models

Cited by

• From In Silico to In Vitro: Evaluating Molecule Generative Models for Hit Generation
• ReACT-Drug: Reaction-Template Guided Reinforcement Learning for de novo Drug Design
• Generative Multi-Objective Bayesian Optimization with Scalable Batch Evaluations for Sample-Efficient De Novo Molecular Design
• Toward Closed-loop Molecular Discovery via Language Model, Property Alignment and Strategic Search
• Efficient and Programmable Exploration of Synthesizable Chemical Space
• Guiding Generative Models for Protein Design: Prompting, Steering and Aligning
• FRAGMENTA: End-to-end Fragmentation-based Generative Model with Agentic Tuning for Drug Lead Optimization
• Q-Steer: Action-Value Guidance for Molecular Policy Optimization
• SHA-256 Infused Embedding-Driven Generative Modeling of High-Energy Molecules in Low-Data Regimes
• Uncertainty-Aware Multi-Objective Reinforcement Learning-Guided Diffusion Models for 3D De Novo Molecular Design
• KL-Regularized Reinforcement Learning is Designed to Mode Collapse
• Coder as Editor: Code-driven Interpretable Molecular Optimization
• Augmenting generative models with biomedical knowledge graphs improves targeted drug discovery
• Guiding Evolutionary Molecular Design: Adding Reinforcement Learning for Mutation Selection
• Refine Drugs, Don't Complete Them: Uniform-Source Discrete Flows for Fragment-Based Drug Discovery
• POLO: Preference-Guided Multi-Turn Reinforcement Learning for Lead Optimization
• A Genetic Algorithm for Navigating Synthesizable Molecular Spaces
• Oracle-Budgeted Molecular Optimization with Short-Term Graph Memory
• AI-aided de novo molecular design of phosphamide for fire-retardant polylactide with enhanced integrated properties
• Valid Property-Enhanced Contrastive Learning for Targeted Optimization & Resampling for Novel Drug Design
• PepThink-R1: LLM for Interpretable Cyclic Peptide Optimization with CoT SFT and Reinforcement Learning
• NovoMolGen: Rethinking Molecular Language Model Pretraining

Related

• Molecular De Novo Design through Transformer-based Reinforcement Learning
• Seq2Mol: Automatic design of de novo molecules conditioned by the target protein sequences through deep neural networks
• Generating Focussed Molecule Libraries for Drug Discovery with Recurrent Neural Networks
• Widely Used and Fast De Novo Drug Design by a Protein Sequence-Based Reinforcement Learning Model
• Application of generative autoencoder in de novo molecular design