OPRD-Imp: a dataset of experimentally observed impurities and side products in multi-step pharmaceutical syntheses
Abstract
Predicting reaction impurities is a critical yet persistently data-limited challenge in pharmaceutical process chemistry. While machine learning models for reaction outcome prediction have advanced rapidly, benchmarking their ability to predict impurities remains impractical due to the absence of curated, publicly accessible datasets. We present OPRD-Imp, a dataset of 229 organic reactions drawn from 62 publications in Organic Process Research & Development, each annotated with experimentally observed impurities and their mechanistic origins. A total of 784 impurity entries is provided, classified into four origin categories. Reactions are represented as reaction SMILES and impurities as individual SMILES strings, with multi-step synthetic route connectivity preserved. All data were manually extracted from the source publications by trained chemists following a standardized extraction protocol, with expert chemical validation and manual structure drawing. OPRD-Imp is intended to serve as a benchmark resource for evaluating models that predict reaction impurities, providing experimentally reported impurity annotations and mechanistic origin labels that are not systematically represented in conventional reaction datasets.
Related
- Recent Analytical Methods for N-Nitrosamine Analysis as Impurities in Drug Substances and Drug Products.
- Comparative Feasibility Assessment of a Simplified Mossel EE–MacConkey Workflow for Detection of Enterobacteriaceae in Selected Non-Sterile Pharmaceutical Products
- Enantioselective Total Syntheses of Bousigonine A and Bisleuconothine B
- Optimized Biomimetic Syntheses of Discorhabdin B and Aleutianamine Drives a Deeper Exploration of Their Anticancer Potential
- Divergent Bioinspired Total Syntheses of Caryophyllene Sesquiterpenoids: Nigrosporinol Sulfoxide B, Punctaporonins C and U, and Desuccinoylpunctaporonin C.
BibTeX
@misc{kim2026oprdimp,
title = {{OPRD-Imp: a dataset of experimentally observed impurities and side products in multi-step pharmaceutical syntheses}},
author = {Heejeong Kim and Seonbin Kim and Jihwan Kim and Tim Pinkhassik and Magdalena Lederbauer and Suong Tran and Connor W. Coley and Juyoung Yoon and Joonyoung F. Joung},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.07488},
eprint = {2611.07488},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.07488},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - OPRD-Imp: a dataset of experimentally observed impurities and side products in multi-step pharmaceutical syntheses
AU - Heejeong Kim
AU - Seonbin Kim
AU - Jihwan Kim
AU - Tim Pinkhassik
AU - Magdalena Lederbauer
AU - Suong Tran
AU - Connor W. Coley
AU - Juyoung Yoon
AU - Joonyoung F. Joung
PY - 2026
DA - 2026/11/05
PB - arXiv
DO - 10.48550/arxiv.2611.07488
UR - https://arxiv.org/abs/2611.07488
AB - Predicting reaction impurities is a critical yet persistently data-limited challenge in pharmaceutical process chemistry. While machine learning models for reaction outcome prediction have advanced rapidly, benchmarking their ability to predict impurities remains impractical due to the absence of curated, publicly accessible datasets. We present OPRD-Imp, a dataset of 229 organic reactions drawn from 62 publications in Organic Process Research & Development, each annotated with experimentally observed impurities and their mechanistic origins. A total of 784 impurity entries is provided, classified into four origin categories. Reactions are represented as reaction SMILES and impurities as individual SMILES strings, with multi-step synthetic route connectivity preserved. All data were manually extracted from the source publications by trained chemists following a standardized extraction protocol, with expert chemical validation and manual structure drawing. OPRD-Imp is intended to serve as a benchmark resource for evaluating models that predict reaction impurities, providing experimentally reported impurity annotations and mechanistic origin labels that are not systematically represented in conventional reaction datasets.
ER -