Active synthesizability prediction for a virtual space of 395 million catalyst candidates with small data
- 1Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 3, Groningen, Groningen, Netherlands, 9747 AG
Abstract
N-heterocyclic carbenes (NHCs) are well established as catalysts and ligands in many chemical transformations. Recently, interest has grown in pyridine[1,5]imidazolium salt pre-catalysts, stable precursors to NHCs, because they are relatively easy to synthesize and highly adaptable. This work introduces a data-driven workflow to identify which pyridine[1,5]imidazolium salts are readily synthetically accessible, starting from exploratory synthesis attempts for 127 compounds. Our approach relies on a feedback loop with a supervised machine learning model to estimate and dynamically update the synthesizability for a combinatorial space of about 394 million pre-catalysts derived from commercially available precursors. Our final model estimates that 4.78% of the entire space are feasible, mapping the space of readily accessible candidates and providing a valuable set for future inverse design studies. We validated the prediction performance of our approach based on both literature data and a holdout set occupying extremes of the combinatorial space.
Related
- EVALUATING MOLECULAR DYNAMICS FRAMES FOR BINDING AFFINITY PREDICTION AND VIRTUAL SCREENING
- Synthesizability: The Open Question in Digital MOF Discovery
- Runtime-steered ultrashort molecular dynamics enables million-pose protein–ligand screening
- Systematic mapping of bimetallic catalyst combinations for floating catalyst chemical vapor deposition growth of single-walled carbon nanotubes
- Acylals Derived from Acrylic Anhydride as Latent Anhydride Crosslinkers for Catalyst-Free Vitrimers
BibTeX
@misc{hovorka2026active,
title = {{Active synthesizability prediction for a virtual space of 395 million catalyst candidates with small data}},
author = {Hannes Hovorka and Noa Huismann and Yanchao Wang and Gian-Michele Cherchi and Michael Lesa and Shiquan Zhou and Robert Pollice},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15010127/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010127/v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Active synthesizability prediction for a virtual space of 395 million catalyst candidates with small data
AU - Hannes Hovorka
AU - Noa Huismann
AU - Yanchao Wang
AU - Gian-Michele Cherchi
AU - Michael Lesa
AU - Shiquan Zhou
AU - Robert Pollice
PY - 2026
DA - 2026/10/08
PB - ChemRxiv
DO - 10.26434/chemrxiv.15010127/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010127/v1
AB - N-heterocyclic carbenes (NHCs) are well established as catalysts and ligands in many chemical transformations. Recently, interest has grown in pyridine[1,5]imidazolium salt pre-catalysts, stable precursors to NHCs, because they are relatively easy to synthesize and highly adaptable. This work introduces a data-driven workflow to identify which pyridine[1,5]imidazolium salts are readily synthetically accessible, starting from exploratory synthesis attempts for 127 compounds. Our approach relies on a feedback loop with a supervised machine learning model to estimate and dynamically update the synthesizability for a combinatorial space of about 394 million pre-catalysts derived from commercially available precursors. Our final model estimates that 4.78% of the entire space are feasible, mapping the space of readily accessible candidates and providing a valuable set for future inverse design studies. We validated the prediction performance of our approach based on both literature data and a holdout set occupying extremes of the combinatorial space.
ER -