GuacaMol: Benchmarking Models for De Novo Molecular Design

Nathan Brown, Marco Fiscato, Marwin H. S. Segler, Alain C. Vaucher
11/22/2018
q-bio.QMcs.LGphysics.chem-ph

Official code: BenevolentAI/guacamol (535 stars, MIT, last commit 2024-02-11). 1 more implementation below.

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

View on arXivView PDF

Code Implementations(2)

BenevolentAI/guacamolOfficial100%
MIT

Benchmarks for generative chemistry

535101Python, DockerfileNov 16, 20182 years agoMIT
BenevolentAI/guacamol_baselinesOfficial100%
MIT

Baselines models for GuacaMol benchmarks

14539Shell, PythonNov 16, 20182 years agoMIT

Similar Papers

A Systematic Evaluation of Molecule Generation Models for De Novo Drug Design: From Benchmarks to Practical Insights

88% match1 repo12026

ChemTS: An Efficient Python Library for de novo Molecular Generation

87% match1 repo1722017

Improving Molecular Design by Stochastic Iterative Target Augmentation

86% match1 repo112020

How Evaluation Choices Distort the Outcome of Generative Drug Discovery

85% match2024

Molecule Design by Latent Prompt Transformer

85% match1 repo82024

Conditional molecular design with deep generative models

85% match1 repo512018

Cite this paper

@article{brown2018guacamol,
  title  = {GuacaMol: Benchmarking Models for De Novo Molecular Design},
  author = {Nathan Brown and Marco Fiscato and Marwin H. S. Segler and Alain C. Vaucher},
  year   = {2018},
  eprint = {1811.09621},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1811.09621}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1811.09621.svg)](https://codeofpaper.com/paper/1811.09621)

Discussion