Compositional Flows for 3D Molecule and Synthesis Pathway Co-design

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shen, Tony, Seo, Seonghwan, Irwin, Ross, Didi, Kieran, Olsson, Simon, Kim, Woo Youn, Ester, Martin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915389649190912
author Shen, Tony
Seo, Seonghwan
Irwin, Ross
Didi, Kieran
Olsson, Simon
Kim, Woo Youn
Ester, Martin
author_facet Shen, Tony
Seo, Seonghwan
Irwin, Ross
Didi, Kieran
Olsson, Simon
Kim, Woo Youn
Ester, Martin
contents Many generative applications, such as synthesis-based 3D molecular design, involve constructing compositional objects with continuous features. Here, we introduce Compositional Generative Flows (CGFlow), a novel framework that extends flow matching to generate objects in compositional steps while modeling continuous states. Our key insight is that modeling compositional state transitions can be formulated as a straightforward extension of the flow matching interpolation process. We further build upon the theoretical foundations of generative flow networks (GFlowNets), enabling reward-guided sampling of compositional structures. We apply CGFlow to synthesizable drug design by jointly designing the molecule's synthetic pathway with its 3D binding pose. Our approach achieves state-of-the-art binding affinity on all 15 targets from the LIT-PCBA benchmark, and 5.8$\times$ improvement in sampling efficiency compared to 2D synthesis-based baseline. To our best knowledge, our method is also the first to achieve state of-art-performance in both Vina Dock (-9.38) and AiZynth success rate (62.2\%) on the CrossDocked benchmark.
format Preprint
id arxiv_https___arxiv_org_abs_2504_08051
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Compositional Flows for 3D Molecule and Synthesis Pathway Co-design
Shen, Tony
Seo, Seonghwan
Irwin, Ross
Didi, Kieran
Olsson, Simon
Kim, Woo Youn
Ester, Martin
Machine Learning
Artificial Intelligence
Many generative applications, such as synthesis-based 3D molecular design, involve constructing compositional objects with continuous features. Here, we introduce Compositional Generative Flows (CGFlow), a novel framework that extends flow matching to generate objects in compositional steps while modeling continuous states. Our key insight is that modeling compositional state transitions can be formulated as a straightforward extension of the flow matching interpolation process. We further build upon the theoretical foundations of generative flow networks (GFlowNets), enabling reward-guided sampling of compositional structures. We apply CGFlow to synthesizable drug design by jointly designing the molecule's synthetic pathway with its 3D binding pose. Our approach achieves state-of-the-art binding affinity on all 15 targets from the LIT-PCBA benchmark, and 5.8$\times$ improvement in sampling efficiency compared to 2D synthesis-based baseline. To our best knowledge, our method is also the first to achieve state of-art-performance in both Vina Dock (-9.38) and AiZynth success rate (62.2\%) on the CrossDocked benchmark.
title Compositional Flows for 3D Molecule and Synthesis Pathway Co-design
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2504.08051