A Genetic Algorithm for Navigating Synthesizable Molecular Spaces

2025/09/25 by Anthony Lo, Connor W. Coley, Lo, Alston +3
Chemistry · Computer Science · Environmental Science · #Chemistry and Chemical Engineering #Computational Drug Discovery Methods #FOS: Biological sciences #FOS: Computer and information sciences #Machine Learning (cs.LG) #Quantitative Methods (q-bio.QM) #Various Chemistry Research Topics

paper · pdf · doi:10.48550/arxiv.2509.20719

openalex publication_date 2025/09/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Inspired by the effectiveness of genetic algorithms and the importance of synthesizability in molecular design, we present SynGA, a simple genetic algorithm that operates directly over synthesis routes. Our method features custom crossover and mutation operators that explicitly constrain it to synthesizable molecular space. By modifying the fitness function, we demonstrate the effectiveness of SynGA on a variety of design tasks, including synthesizable analog search and sample-efficient property optimization, for both 2D and 3D objectives. Furthermore, by coupling SynGA with a machine learning-based filter that focuses the building block set, we boost SynGA to state-of-the-art performance. For property optimization, this manifests as a model-based variant SynGBO, which employs SynGA and block filtering in the inner loop of Bayesian optimization. Since SynGA is lightweight and enforces synthesizability by construction, our hope is that SynGA can not only serve as a strong standalone baseline but also as a versatile module that can be incorporated into larger synthesis-aware workflows in the future.

Citations

• LLM-Augmented Chemical Synthesis and Design Decision Programs
• Compositional Flows for 3D Molecule and Synthesis Pathway Co-design
• SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Models
• Neural Genetic Search in Discrete Spaces
• Beyond predefined ligand libraries: a genetic algorithm approach for <i>de novo</i> discovery of catalysts for the Suzuki coupling reactions
• Molecule Generation with Fragment Retrieval Augmentation
• Generative Flows on Synthetic Pathway for Drug Design
• Generative Artificial Intelligence for Navigating Synthesizable Chemical Space
• Procedural Synthesis of Synthesizable Molecules
• Directly Optimizing for Synthesizability in Generative Molecular Design using Retrosynthesis Models
• Diagnosing and fixing common problems in Bayesian optimization for molecule design
• Projecting Molecules into Synthesizable Chemical Spaces
• RGFN: Synthesizable Molecular Generation Using GFlowNets
• Saturn: Sample-efficient Generative Molecular Design using Memory Manipulation
• SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraints
• Genetic-guided GFlowNets for Sample Efficient Molecular Optimization
• Genetic algorithms are strong baselines for molecule generation
• ChemCrow: Augmenting large-language models with chemistry tools
• GAUCHE: A Library for Gaussian Processes in Chemistry
• Reinforced Genetic Algorithm for Structure-based Drug Design
• Tartarus: A Benchmarking Platform for Realistic And Practical Inverse Molecular Design
• Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization
• Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pockets
• Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design
• Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation
• JANUS: Parallel Tempered Genetic Algorithm Guided by Deep Neural Networks for Inverse Molecular Design
• Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development
• Barking up the right tree: an approach to search over molecule synthesis DAGs
• Contrastive Learning with Hard Negative Samples
• Guiding Deep Molecular Optimization with Genetic Exploration
• Neural Additive Models: Interpretable Machine Learning with Neural Nets
• Learning To Navigate The Synthetically Accessible Chemical Space Using\n Reinforcement Learning
• The Synthesizability of Molecules Proposed by Generative Models
• A Model to Search for Synthesizable Molecules
• GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration
• Multi-Objective De Novo Drug Design with Conditional Graph Generative Model
• Molecular De Novo Design through Deep Reinforcement Learning
• Los carnavales de La Bañeza: Aspectos antropológicos

Related

• SynTwins: A Retrosynthesis-Guided Framework for Synthesizable Molecular Analog Generation
• Genetic Algorithm for Constrained Molecular Inverse Design
• Generative Artificial Intelligence for Navigating Synthesizable Chemical Space
• Gradient GA: Gradient Genetic Algorithm for Drug Molecular Design
• JANUS: Parallel Tempered Genetic Algorithm Guided by Deep Neural Networks for Inverse Molecular Design