GuacaMol: Benchmarking Models for De Novo Molecular Design

Abstract

De novo design seeks to generate molecules with required property profiles by virtual design-make-test cycles. With the emergence of deep learning and neural generative models in many application areas, models for molecular design based on neural networks appeared recently and show promising results. However, the new models have not been profiled on consistent tasks, and comparative studies to well-established algorithms have only seldom been performed. To standardize the assessment of both classical and neural models for de novo molecular design, we propose an evaluation framework, GuacaMol, based on a suite of standardized benchmarks. The benchmark tasks encompass measuring the fidelity of the models to reproduce the property distribution of the training sets, the ability to generate novel molecules, the exploration and exploitation of chemical space, and a variety of single and multi-objective optimization tasks. The benchmarking open-source Python code, and a leaderboard can be found on https://benevolent.ai/guacamol

References in corpus (5)

• Deep Learning in Neural Networks: An Overview
• Automatic chemical design using a data-driven continuous representation of molecules
• Learning to Plan Chemical Syntheses
• Deep Reinforcement Learning for De-Novo Drug Design
• Population-based de novo molecule generation, using grammatical evolution

Cited by in corpus (55)

• Molecular Contrastive Learning of Representations via Graph Neural Networks
• Deep learning for molecular design - a review of the state of the art
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
• PoseBusters: AI-based docking methods fail to generate physically valid poses or generalise to novel sequences
• SELFIES and the future of molecular string representations
• Autonomous discovery in the chemical sciences part II: Outlook
• Learning to design drug-like molecules in three-dimensional space using deep generative models
• Exploring Chemical Space using Natural Language Processing Methodologies for Drug Discovery
• Artificial Intelligence and Machine Learning for Quantum Technologies
• Scaffold-based molecular design using graph generative model
• Deep Learning and Knowledge-Based Methods for Computer Aided Molecular Design -- Toward a Unified Approach: State-of-the-Art and Future Directions
• Diffusion Models in $\text{De Novo}$ Drug Design
• Generative Models for Automatic Chemical Design
• Exploring the Advantages of Quantum Generative Adversarial Networks in Generative Chemistry
• Graph Neural Networks for Molecules
• High-throughput property-driven generative design of functional organic molecules
• Optimizing Molecules using Efficient Queries from Property Evaluations
• Scaffold-constrained molecular generation
• Olympus: a benchmarking framework for noisy optimization and experiment planning
• JARVIS-Leaderboard: A Large Scale Benchmark of Materials Design Methods
• Hybrid quantum-classical machine learning for generative chemistry and drug design
• We Should at Least Be Able to Design Molecules That Dock Well
• El Agente: An Autonomous Agent for Quantum Chemistry
• Roughness of molecular property landscapes and its impact on modellability
• Docking-based generative approaches in the search for new drug candidates
• Accelerating Material Design with the Generative Toolkit for Scientific Discovery
• Utilizing Reinforcement Learning for de novo Drug Design
• Graph Machine Learning for Design of High-Octane Fuels
• Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning
• Toward Automated Simulation Research Workflow through LLM Prompt Engineering Design
• Curiosity in exploring chemical space: Intrinsic rewards for deep molecular reinforcement learning
• Faster and more diverse de novo molecular optimization with double-loop reinforcement learning using augmented SMILES
• Spiers Memorial Lecture: How to do impactful research in artificial intelligence for chemistry and materials science
• A generative model for molecule generation based on chemical reaction trees
• Application and Assessment of Deep Learning for the Generation of Potential NMDA Receptor Antagonists
• Machine Learning Scoring Functions for Drug Discoveries from Experimental and Computer-Generated Protein-Ligand Structures: Towards Per-Target Scoring Functions
• Language models in molecular discovery
• ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation
• Automated patent extraction powers generative modeling in focused chemical spaces
• Adversarial Learned Molecular Graph Inference and Generation
• Quantum-machine-assisted Drug Discovery
• Diffusion Models at the Drug Discovery Frontier: A Review on Generating Small Molecules versus Therapeutic Peptides
• Unraveling the Potential of Diffusion Models in Small Molecule Generation
• GraphXForm: Graph transformer for computer-aided molecular design
• PEMP: Leveraging Physics Properties to Enhance Molecular Property Prediction
• Hybrid Quantum Generative Adversarial Networks for Molecular Simulation and Drug Discovery
• Ranking over Regression for Bayesian Optimization and Molecule Selection
• pyscreener: A Python Wrapper for Computational Docking Software
• How Evaluation Choices Distort the Outcome of Generative Drug Discovery
• A Bayesian Flow Network Framework for Chemistry Tasks
• Evaluation Framework for AI-driven Molecular Design of Multi-target Drugs: Brain Diseases as a Case Study
• Evolutionary Multi-Objective Design of SARS-CoV-2 Protease Inhibitor Candidates
• Heuristics and Uncertainty Quantification in Rational and Inverse Compound and Catalyst Design
• SAGE-Amine: Generative Amine Design with Multi-Property Optimization for Efficient CO2 Capture
• CompanyKG: A Large-Scale Heterogeneous Graph for Company Similarity Quantification