RetroGFN: Diverse and Feasible Retrosynthesis using GFlowNets

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gaiński, Piotr, Koziarski, Michał, Maziarz, Krzysztof, Segler, Marwin, Tabor, Jacek, Śmieja, Marek
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916686937980928
author Gaiński, Piotr
Koziarski, Michał
Maziarz, Krzysztof
Segler, Marwin
Tabor, Jacek
Śmieja, Marek
author_facet Gaiński, Piotr
Koziarski, Michał
Maziarz, Krzysztof
Segler, Marwin
Tabor, Jacek
Śmieja, Marek
contents Single-step retrosynthesis aims to predict a set of reactions that lead to the creation of a target molecule, which is a crucial task in molecular discovery. Although a target molecule can often be synthesized with multiple different reactions, it is not clear how to verify the feasibility of a reaction, because the available datasets cover only a tiny fraction of the possible solutions. Consequently, the existing models are not encouraged to explore the space of possible reactions sufficiently. In this paper, we propose a novel single-step retrosynthesis model, RetroGFN, that can explore outside the limited dataset and return a diverse set of feasible reactions by leveraging a feasibility proxy model during the training. We show that RetroGFN achieves competitive results on standard top-k accuracy while outperforming existing methods on round-trip accuracy. Moreover, we provide empirical arguments in favor of using round-trip accuracy, which expands the notion of feasibility with respect to the standard top-k accuracy metric.
format Preprint
id arxiv_https___arxiv_org_abs_2406_18739
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RetroGFN: Diverse and Feasible Retrosynthesis using GFlowNets
Gaiński, Piotr
Koziarski, Michał
Maziarz, Krzysztof
Segler, Marwin
Tabor, Jacek
Śmieja, Marek
Machine Learning
Single-step retrosynthesis aims to predict a set of reactions that lead to the creation of a target molecule, which is a crucial task in molecular discovery. Although a target molecule can often be synthesized with multiple different reactions, it is not clear how to verify the feasibility of a reaction, because the available datasets cover only a tiny fraction of the possible solutions. Consequently, the existing models are not encouraged to explore the space of possible reactions sufficiently. In this paper, we propose a novel single-step retrosynthesis model, RetroGFN, that can explore outside the limited dataset and return a diverse set of feasible reactions by leveraging a feasibility proxy model during the training. We show that RetroGFN achieves competitive results on standard top-k accuracy while outperforming existing methods on round-trip accuracy. Moreover, we provide empirical arguments in favor of using round-trip accuracy, which expands the notion of feasibility with respect to the standard top-k accuracy metric.
title RetroGFN: Diverse and Feasible Retrosynthesis using GFlowNets
topic Machine Learning
url https://arxiv.org/abs/2406.18739