Development of a Novel Fingerprint for Chemical Reactions and Its Application to Large-Scale Reaction Classification and Similarity

By Nadine Schneider, Daniel M. Lowe, Roger A. Sayle, Gregory A. Landrum. 2015

182 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Philippe Schwaller
2. Marinka Zitnik
3. Tianfan Fu
4. Jimeng Sun
5. Jure Leskovec
6. Sepp Hochreiter
7. Kexin Huang
8. Teodoro Laino

In Materials Chemistry

1. Philippe Schwaller
2. Sepp Hochreiter
3. Tianfan Fu
4. Jimeng Sun
5. Günter Klambauer
6. Philipp Renz
7. Jure Leskovec
8. Teodoro Laino

In Molecular Biology

1. Marinka Zitnik
2. Kexin Huang
3. Jure Leskovec
4. Cao Xiao
5. Sepp Hochreiter
6. Jimeng Sun
7. Tianfan Fu
8. Günter Klambauer

In Computational Chemistry & Drug Discovery

1. Sepp Hochreiter
2. Günter Klambauer
3. Jure Leskovec
4. Philippe Schwaller
5. Teodoro Laino
6. Alpha Albert Lee
7. C. Bekas
8. Connor W. Coley

Citations in the corpus

• Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development (2021, methodology): “USPTO-50K: USPTO (United States Patent and Trademark Office) 50K consists of 50K extracted atommapped reactions with 10 reaction types (Schneider et al. 2015).”
• Modern Hopfield Networks for Few- and Zero-Shot Reaction Prediction (2021, background): “Each point is coloured according to its class as defined in (Schneider et al., 2015).”
• Molecular representations in AI-driven drug discovery: a review and practical guide (2020, methodology): “[81] have used the difference fingerprint with the atom-pair variant to build a machine learning system for a 50-class reaction classification model.”
• Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction (2018, background): “NameRXN [26, 27]) are themselves based on libraries of expert rules and templates.”