“Found in Translation”: predicting outcomes of complex organic chemistry reactions using neural sequence-to-sequence models

By Philippe Schwaller, Théophile Gaudin, David Lanyi, C. Bekas, Teodoro Laino. Chemical Science, 2017

325 citations recorded by the source corpus.

Using a text-based representation of molecules, chemical reactions are predicted with a neural machine translation model borrowed from language processing.

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

View Paper: https://www.semanticscholar.org/paper/f8c306a5e8a19f5806bac9f26de14d09f7686aac

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

Geometric deep learning on molecular representations (2021, unknown): “SMILES strings – in which letters are used to represent atoms, and symbols and numbers are used to encode bond types, connectivity, branching, and stereochemistry (Figure 3a) – have become the most frequently employed data representation method for sequence-based deep learning [19, 46].”
A graph-convolutional neural network model for the prediction of chemical reactivity (2018, unknown): “[31] Philippe Schwaller, Theophile Gaudin, David Lanyi, Costas Bekas, and Teodoro Laino.”
Combining generative artificial intelligence and on-chip synthesis for de novo drug design (2020, background): “…explore the permissive structural diversity of the designs, several strategies can be adopted; for example, choosing different optimization criteria for the generative deep learning model, or including other artificial intelligence models 12,13 in the definition of compatible organic reactions.”
Modern Hopfield Networks for Few- and Zero-Shot Reaction Prediction (2021, background): “In this case, the reaction prediction can be considered as a translation task, in which the reactants represent a sentence in the input language and the products the sentence in the output language (Nam & Kim, 2016; Schwaller et al., 2018a; Liu et al., 2017; Schwaller et al., 2018b).”
LSTMs Exploit Linguistic Attributes of Data (2018, unknown): “Since LSTMs better memorize (and thus better count) on language data than on non-language data, and counting plays a role in encoder-decoder models, our work could also lead to improved training for sequence-to-sequence models in non-language applications (e.g., Schwaller et al., 2017).”
Artificial intelligence in drug discovery: recent advances and future perspectives (2021, unknown): “With regard to template-free forward synthesis prediction, natural language processing approaches based on transformer [184] or recurrent neural network architectures [185] are also becoming popular.”