On the use of real-world datasets for reaction yield prediction
Open Access
- 13 March 2023
- journal article
- research article
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 14 (19) , 4997-5005
Abstract
An attributed graph neural network predicts the yield of Suzuki–Miyaura and Buchwald–Hartwig reactions for datasets from high-throughput experimentation (HTE) but not for a more diverse real-world dataset from electronic lab notebooks (ELNs).Keywords
All Related Versions
Funding Information
- National Science Foundation (CHE-1925607)
- National Science Foundation (CHE-2202693)
This publication has 56 references indexed in Scilit:
- De Novo Design of Bioactive Small Molecules by Artificial IntelligenceMolecular Informatics, 2018
- Optimizing Chemical Reactions with Deep Reinforcement LearningACS Central Science, 2017
- Computer-Assisted Retrosynthesis Based on Molecular SimilarityACS Central Science, 2017
- Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?Scientific Reports, 2017
- A Holy Grail in Chemistry: Computational Catalyst Design: Feasible or Fiction?Accounts of Chemical Research, 2017
- Ror2 signaling regulates Golgi structure and transport through IFT20 for tumor invasivenessScientific Reports, 2017
- Suzuki–Miyaura cross-coupling optimization enabled by automated feedbackReaction Chemistry & Engineering, 2016
- ZINC 15 – Ligand Discovery for EveryoneJournal of Chemical Information and Modeling, 2015
- Route Design in the 21st Century: The ICSYNTH Software Tool as an Idea Generator for Synthesis PredictionOrganic Process Research & Development, 2015
- Mining Electronic Laboratory Notebooks: Analysis, Retrosynthesis, and Reaction Based EnumerationJournal of Chemical Information and Modeling, 2012