Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2407.06334 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912099312074752 |
|---|---|
| author | Yu, Kevin Roh, Jihye Li, Ziang Gao, Wenhao Wang, Runzhong Coley, Connor W. |
| author_facet | Yu, Kevin Roh, Jihye Li, Ziang Gao, Wenhao Wang, Runzhong Coley, Connor W. |
| contents | Computer-aided synthesis planning (CASP) algorithms have demonstrated expert-level abilities in planning retrosynthetic routes to molecules of low to moderate complexity. However, current search methods assume the sufficiency of reaching arbitrary building blocks, failing to address the common real-world constraint where using specific molecules is desired. To this end, we present a formulation of synthesis planning with starting material constraints. Under this formulation, we propose Double-Ended Synthesis Planning (DESP), a novel CASP algorithm under a bidirectional graph search scheme that interleaves expansions from the target and from the goal starting materials to ensure constraint satisfiability. The search algorithm is guided by a goal-conditioned cost network learned offline from a partially observed hypergraph of valid chemical reactions. We demonstrate the utility of DESP in improving solve rates and reducing the number of search expansions by biasing synthesis planning towards expert goals on multiple new benchmarks. DESP can make use of existing one-step retrosynthesis models, and we anticipate its performance to scale as these one-step model capabilities improve. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_06334 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Double-Ended Synthesis Planning with Goal-Constrained Bidirectional Search Yu, Kevin Roh, Jihye Li, Ziang Gao, Wenhao Wang, Runzhong Coley, Connor W. Artificial Intelligence Quantitative Methods Computer-aided synthesis planning (CASP) algorithms have demonstrated expert-level abilities in planning retrosynthetic routes to molecules of low to moderate complexity. However, current search methods assume the sufficiency of reaching arbitrary building blocks, failing to address the common real-world constraint where using specific molecules is desired. To this end, we present a formulation of synthesis planning with starting material constraints. Under this formulation, we propose Double-Ended Synthesis Planning (DESP), a novel CASP algorithm under a bidirectional graph search scheme that interleaves expansions from the target and from the goal starting materials to ensure constraint satisfiability. The search algorithm is guided by a goal-conditioned cost network learned offline from a partially observed hypergraph of valid chemical reactions. We demonstrate the utility of DESP in improving solve rates and reducing the number of search expansions by biasing synthesis planning towards expert goals on multiple new benchmarks. DESP can make use of existing one-step retrosynthesis models, and we anticipate its performance to scale as these one-step model capabilities improve. |
| title | Double-Ended Synthesis Planning with Goal-Constrained Bidirectional Search |
| topic | Artificial Intelligence Quantitative Methods |
| url | https://arxiv.org/abs/2407.06334 |