The Synthesizability of Molecules Proposed by Generative Models

By Wenhao Gao, Connor W. Coley. Journal of Chemical Information and Modeling, 2020

344 citations recorded by the source corpus.

The discovery of functional molecules is an expensive and time-consuming process, exemplified by the rising costs of small molecule therapeutic discovery. One class of techniques of growing interest for early-stage drug discovery is de novo molecular generation and optimization, catalyzed by the development of new deep learning approaches. These techniques can suggest novel molecular structures intended to maximize a multi-objective function, e.g., suitability as a therapeutic against a particular target, without relying on brute-force exploration of a chemical space. However, the utility of these approaches is stymied by ignorance of synthesizability. To highlight the severity of this issue, we use a data-driven computer-aided synthesis planning program to quantify how often molecules proposed by state-of-the-art generative models cannot be readily synthesized. Our analysis demonstrates that there are several tasks for which these models generate unrealistic molecular structures despite performing well on popular quantitative benchmarks. Synthetic complexity heuristics can successfully bias generation toward synthetically-tractable chemical space, although doing so necessarily detracts from the primary objective. This analysis suggests that to improve the utility of these models in real discovery workflows, new algorithm development is warranted.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Pharmacology

View Paper: https://www.semanticscholar.org/paper/555c0d308a1c9faa089430133566bf0dd6e4613d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Marinka Zitnik
2. Tianfan Fu
3. Jimeng Sun
4. Jure Leskovec
5. Shengchao Liu
6. Kexin Huang
7. Cao Xiao
8. Wenhao Gao

In Materials Chemistry

1. Tianfan Fu
2. Jimeng Sun
3. Shengchao Liu
4. Jian Tang
5. Yuanqi Du
6. Marinka Zitnik
7. Cao Xiao
8. Wenhao Gao

In Pharmacology

1. Jimeng Sun
2. Tianfan Fu
3. Wenhao Gao
4. Connor W. Coley
5. Cao Xiao
6. Jure Leskovec
7. Marinka Zitnik
8. Kexin Huang

Citations in the corpus

• MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design (2022, background): “In contrast, combinatorial optimization methods (COMs) have less requirement on the amount of training data, while the trade-off is that algorithm designers need to call the oracles during the exploration to the chemical space [103], [104], [105], [106], [107].”
• Differentiable Scaffolding Tree for Molecular Optimization (2021, positive): “Following [13], we use the normalize function for raw SA score, where µ = 2 .”
• Pre-training Molecular Graph Representation with 3D Geometry (2021, background): “(3) VRR, as a generative SSL method, is able to alleviate the potential issues in molecule generation [22, 93].”
• Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development (2021, background): “Further, generative models can generate molecules with structure outside of pre-defined chemical space, meaning that the generated molecule might have a valid chemical structure but could not be practically synthesized in a laboratory [40].”