Designing the Haystack: Programmable Chemical Space for Generative Molecular Discovery
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915825530699776 |
|---|---|
| author | Zhu, Yuchen Zhao, Donghai Zhang, Yangyang Li, Yitong Wang, Xiaorui Li, Shuwang Kong, Yue Zhang, Beichen Chen, Ricki Liu, Chang Zhang, Xingcai Hou, Tingjun Hsieh, Chang-Yu |
| author_facet | Zhu, Yuchen Zhao, Donghai Zhang, Yangyang Li, Yitong Wang, Xiaorui Li, Shuwang Kong, Yue Zhang, Beichen Chen, Ricki Liu, Chang Zhang, Xingcai Hou, Tingjun Hsieh, Chang-Yu |
| contents | Chemical space exploration underlies drug discovery, yet most generative models treat chemical space as a fixed, implicitly learned distribution, focusing on sampling molecules rather than deliberately designing the space itself. We introduce SpaceGFN, a generative framework that elevates chemical space to a programmable computational object: a controllable degree of freedom enabling explicit construction and adaptive traversal of structured molecular universes. SpaceGFN decouples space definition from exploration. Users specify building blocks and reaction rules to construct chemically and synthetically coherent spaces, while a GFlowNet performs efficient, property-biased sampling within them. In Discovery mode, we demonstrate programmable space design through two strategies. A pseudo-natural product space assembles natural product-like architectures. An evolution-inspired (Evo) space recombines endogenous metabolite fragments via enzyme-consistent transformations, introducing an evolutionary prior into chemical generation. This bias yields favorable shifts in predicted metabolic and toxicological profiles while preserving pharmacological diversity, supported by broad docking enrichment across therapeutic targets. In Editing mode, SpaceGFN enables reaction-consistent lead optimization through a curated toolkit of executable synthetic transformations, allowing local, synthesis-aware modification of existing compounds instead of unrestricted graph mutation. Across 96 drug targets, SpaceGFN achieves strong optimization performance while maintaining structural diversity under synthetic constraints. By integrating programmable chemical universe construction with flow-based exploration and reaction-level editing, SpaceGFN establishes a general paradigm for deliberate navigation of therapeutic chemical space. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_00614 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Designing the Haystack: Programmable Chemical Space for Generative Molecular Discovery Zhu, Yuchen Zhao, Donghai Zhang, Yangyang Li, Yitong Wang, Xiaorui Li, Shuwang Kong, Yue Zhang, Beichen Chen, Ricki Liu, Chang Zhang, Xingcai Hou, Tingjun Hsieh, Chang-Yu Chemical Physics Biomolecules Quantitative Methods Chemical space exploration underlies drug discovery, yet most generative models treat chemical space as a fixed, implicitly learned distribution, focusing on sampling molecules rather than deliberately designing the space itself. We introduce SpaceGFN, a generative framework that elevates chemical space to a programmable computational object: a controllable degree of freedom enabling explicit construction and adaptive traversal of structured molecular universes. SpaceGFN decouples space definition from exploration. Users specify building blocks and reaction rules to construct chemically and synthetically coherent spaces, while a GFlowNet performs efficient, property-biased sampling within them. In Discovery mode, we demonstrate programmable space design through two strategies. A pseudo-natural product space assembles natural product-like architectures. An evolution-inspired (Evo) space recombines endogenous metabolite fragments via enzyme-consistent transformations, introducing an evolutionary prior into chemical generation. This bias yields favorable shifts in predicted metabolic and toxicological profiles while preserving pharmacological diversity, supported by broad docking enrichment across therapeutic targets. In Editing mode, SpaceGFN enables reaction-consistent lead optimization through a curated toolkit of executable synthetic transformations, allowing local, synthesis-aware modification of existing compounds instead of unrestricted graph mutation. Across 96 drug targets, SpaceGFN achieves strong optimization performance while maintaining structural diversity under synthetic constraints. By integrating programmable chemical universe construction with flow-based exploration and reaction-level editing, SpaceGFN establishes a general paradigm for deliberate navigation of therapeutic chemical space. |
| title | Designing the Haystack: Programmable Chemical Space for Generative Molecular Discovery |
| topic | Chemical Physics Biomolecules Quantitative Methods |
| url | https://arxiv.org/abs/2603.00614 |