Portfolio Search with Precedent-Based Selection for Multi-Step Retrosynthesis
Abstract
Computer-aided synthesis planners propose routes from a target molecule back to purchasable starting materials, and they are usually compared by how often they find such a route. A route that reaches purchasable stock, however, often contains steps that have no precedent in the chemical literature and that a chemist would not run. We present P*, a planner that searches with a portfolio of complementary strategies and then keeps, for each target, the route whose every step has literature precedent. Each strategy pairs a reaction score with an estimate of how hard an intermediate is to make. The reaction scores are built from chemical heuristics, such as how common a transformation is and how many of its precursors are purchasable. We find that the strategies that reach stock most often produce the fewest precedented routes, because rewarding purchasable precursors favors multicomponent disconnections that are rarely reported. Different strategies therefore find different precedented routes, and together they find precedented routes for about twice as many development targets as the best single strategy. P* produces fully precedented routes for 61% of approved drugs and clinical candidates, 41% of novel expert-curated targets and 49% of USPTO-190 targets, more than any published planner on all three sets. It exceeds the strongest of them, and it improves route validity with every single-step model we tested. Its best single strategy is also an effective route finder and solves 88.0% of ChEMBL-1000 targets.
Related
- RetroAgent: Harnessing LLMs to Search Over Structured Memory for Agentic Retrosynthesis Planning
- Chemist-aligned retrosynthesis by ensembling diverse inductive bias models
- RetroMol: Knowledge Distillation of Energy-Based Retrosynthesis Models
- URSA: Chemistry-Aware Benchmark for Utilitarian Retrosynthesis Assessment
- Improving computer-aided retrosynthesis of chemically intuitive disconnections
BibTeX
@misc{sun2026portfolio,
title = {{Portfolio Search with Precedent-Based Selection for Multi-Step Retrosynthesis}},
author = {Fang Sun and Yizhou Sun},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15011122/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15011122/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Portfolio Search with Precedent-Based Selection for Multi-Step Retrosynthesis
AU - Fang Sun
AU - Yizhou Sun
PY - 2026
DA - 2026/10/27
PB - ChemRxiv
DO - 10.26434/chemrxiv.15011122/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15011122/v1
AB - Computer-aided synthesis planners propose routes from a target molecule back to purchasable starting materials, and they are usually compared by how often they find such a route. A route that reaches purchasable stock, however, often contains steps that have no precedent in the chemical literature and that a chemist would not run. We present P*, a planner that searches with a portfolio of complementary strategies and then keeps, for each target, the route whose every step has literature precedent. Each strategy pairs a reaction score with an estimate of how hard an intermediate is to make. The reaction scores are built from chemical heuristics, such as how common a transformation is and how many of its precursors are purchasable. We find that the strategies that reach stock most often produce the fewest precedented routes, because rewarding purchasable precursors favors multicomponent disconnections that are rarely reported. Different strategies therefore find different precedented routes, and together they find precedented routes for about twice as many development targets as the best single strategy. P* produces fully precedented routes for 61% of approved drugs and clinical candidates, 41% of novel expert-curated targets and 49% of USPTO-190 targets, more than any published planner on all three sets. It exceeds the strongest of them, and it improves route validity with every single-step model we tested. Its best single strategy is also an effective route finder and solves 88.0% of ChEMBL-1000 targets.
ER -