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

Tony Shen, Seonghwan Seo, Ross Irwin, Kieran Didi, Simon Olsson, Woo Youn Kim, Martin Ester
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:54381-54409, 2025.

Abstract

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 and synthesizability on all 15 targets from the LIT-PCBA benchmark, and 4.2x 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 (-9.42) and AiZynth success rate (36.1%) on the CrossDocked2020 benchmark.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-shen25b, title = {Compositional Flows for 3{D} Molecule and Synthesis Pathway Co-design}, author = {Shen, Tony and Seo, Seonghwan and Irwin, Ross and Didi, Kieran and Olsson, Simon and Kim, Woo Youn and Ester, Martin}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {54381--54409}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/shen25b/shen25b.pdf}, url = {https://proceedings.mlr.press/v267/shen25b.html}, abstract = {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 and synthesizability on all 15 targets from the LIT-PCBA benchmark, and 4.2x 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 (-9.42) and AiZynth success rate (36.1%) on the CrossDocked2020 benchmark.} }
Endnote
%0 Conference Paper %T Compositional Flows for 3D Molecule and Synthesis Pathway Co-design %A Tony Shen %A Seonghwan Seo %A Ross Irwin %A Kieran Didi %A Simon Olsson %A Woo Youn Kim %A Martin Ester %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-shen25b %I PMLR %P 54381--54409 %U https://proceedings.mlr.press/v267/shen25b.html %V 267 %X 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 and synthesizability on all 15 targets from the LIT-PCBA benchmark, and 4.2x 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 (-9.42) and AiZynth success rate (36.1%) on the CrossDocked2020 benchmark.
APA
Shen, T., Seo, S., Irwin, R., Didi, K., Olsson, S., Kim, W.Y. & Ester, M.. (2025). Compositional Flows for 3D Molecule and Synthesis Pathway Co-design. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:54381-54409 Available from https://proceedings.mlr.press/v267/shen25b.html.

Related Material