A graph-convolutional neural network model for the prediction of chemical reactivity

By Connor W. Coley, Wengong Jin, Luke Rogers, T. Jamison, Tommi S. Jaakkola, William H. Green Jr., Regina Barzilay, Klavs F. Jensen. Chemical Science, 2018

515 citations recorded by the source corpus.

We present a supervised learning approach to predict the products of organic reactions given their reactants, reagents, and solvent(s).

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Materials Chemistry

View Paper: https://www.semanticscholar.org/paper/c3d8fb489e4dfeef912efab64cc7a158f3d816fd

How researchers see this paper

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

In Artificial Intelligence

In Computational Theory and Mathematics

In Materials Chemistry

Citations in the corpus

AWB-GCN: A Graph Convolutional Network Accelerator with Runtime Workload Rebalancing (2019, background): “…in a large number of real-world applications [8], including electric grid cascading failure analysis [32], prediction of chemical reactivity [33], prediction of synthesized material properties [34], polypharmacy side-effect modeling [35], accurate advertisement in E-commerce [36], and…”
Unsupervised Attention-Guided Atom-Mapping (2020, methodology): “, 2020), reaction graph neural network algorithms (Jin et al., 2017; Coley et al., 2019), reactant-reagent role assignments (Schneider et al.”
A review on machine learning approaches and trends in drug discovery (2021, background): “On the other hand, in [55] implemented another graph-convolutional neural network for the prediction of chemical reactivity.”
Graph Representation Learning in Biomedicine and Healthcare (2021, unknown): “Producing atom-centred representations based on a weighted combination of their neighbours’ features (via an attention mechanism) can be used to model interactions among reactants and to predict the outcomes of the reactions 154 .”
Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction (2018, unknown): “Figure 2: The 6 reactions in the human test set [29] not predicted within top-2 using our model trained on the augmented mixed USPTO_MIT set.”
Geometric deep learning on molecular representations (2021, methodology): “For modelling tasks that are usually characterized by the need for engineered rules (for example, molecule construction for de novo design and reactive site specification for CASP), GDL extends the existing methodological repertoire.”
Tackling Climate Change with Machine Learning (2019, background): “ To develop cleaner ammonia, chemists may be able to invent electrochemical strategies for lower-temperature ammonia production [386, 387]. Given the potential of ML for predicting chemical reactions [385], ML may also be able to help with the discovery of new materials for electrocatalysts and/or p”
Modern Hopfield Networks for Few- and Zero-Shot Reaction Prediction (2021, methodology): “USPTO-lg is based on USPTO-410k by Coley et al. (2019b).”
A robotic platform for flow synthesis of organic compounds informed by AI planning (2019, methodology): “…of reaction steps considered plausible by the binary classifier can be evaluated by the more rigorous forward prediction model, which explicitly generates product molecules based on the provided reactants and reaction conditions ( 37 , 38 ); the exact model used is described in detail in ( 37 ).”