Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction

@article{Schwaller2019MolecularTA,
  title={Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction},
  author={Philippe Schwaller and Teodoro Laino and Th{\'e}ophile Gaudin and Peter Bolgar and Christopher A. Hunter and Costas Bekas and Alpha Albert Lee},
  journal={ACS Central Science},
  year={2019},
  volume={5},
  pages={1572 - 1583}
}
Organic synthesis is one of the key stumbling blocks in medicinal chemistry. [] Key Method Molecular Transformer makes predictions by inferring the correlations between the presence and absence of chemical motifs in the reactant, reagent, and product present in the data set. Our model requires no handcrafted rules and accurately predicts subtle chemical transformations. Crucially, our model can accurately estimate its own uncertainty, with an uncertainty score that is 89% accurate in terms of classifying…

Figures and Tables from this paper

Evaluating the Performance of a Transformer-based Organic Reaction Prediction Model

Forward and retrosynthetic organic reaction prediction are challenging applications of artificial intelligence (AI) research in chemistry. IBM’s freely available RXN for Chemistry

Molecular Transformer unifies reaction prediction and retrosynthesis across pharma chemical space.

TLDR
This work shows that an attention-based machine translation model - Molecular Transformer - tackles both reaction prediction and retrosynthesis by learning from the same dataset, and expects this versatile framework to be broadly applicable to problems such as reaction condition prediction, reagent prediction and yield prediction.

Root-aligned SMILES: a tight representation for chemical reaction prediction

TLDR
This article proposes the root-aligned SMILES (R-SMILES), which is a tightly aligned one-to-one mapping between the product and the reactant SMilES for more e-cient synthesis prediction and shows that it significantly outperforms them all, demonstrating the superiority of the proposed method.

Retrosynthetic reaction pathway prediction through neural machine translation of atomic environments

Designing efficient synthetic routes for a target molecule remains a major challenge in organic synthesis. Atom environments are ideal, stand-alone, chemically meaningful building blocks providing a

RetroTRAE : retrosynthetic prediction with Transformer Retrosynthetic reaction pathway prediction through neural 1 machine translation of atomic environments

7 Designing efficient synthetic routes for a target molecule remains a major challenge in organic syn8 thesis. Atom environments are ideal, stand-alone, chemically meaningful building blocks

A generalized-template-based graph neural network for accurate organic reactivity prediction

TLDR
A chemistry-motivated graph neural network called LocalTransform is proposed, which learns organic reactivity based on generalized reaction templates to describe the net changes in electron configuration between the reactants and products and exhibits state-of-the-art product prediction accuracy.

Sequence-based modeling recasts

Herein we present a new retrosynthesis prediction method, viz. RetroTRAE, which uses fragment-based tokenization combined with the Transformer architecture. RetroTRAE mimics chemical reasoning, and

Predicting chemical reaction outcomes: A grammar ontology‐based transformer framework

Discovering and designing novel materials is a challenging problem as it often requires searching a combinatorially large space of potential candidates. Evaluation of all candidates experimentally is

Quantitative interpretation explains machine learning models for chemical reaction prediction and uncovers bias

TLDR
This paper quantitatively interpret the Molecular Transformer, the state-of-the-art model for reaction prediction, and develops a framework to attribute predicted reaction outcomes both to specific parts of reactants, and to reactions in the training set.

RetroTRAE: retrosynthetic translation of atomic environments with Transformer

We present a new single-step retrosynthesis prediction method, viz. RetroTRAE, using fragment-based tokenization and the Transformer architecture. RetroTRAE mimics chemical reasoning, and predicts
...

References

SHOWING 1-10 OF 56 REFERENCES

Graph Transformation Policy Network for Chemical Reaction Prediction

TLDR
This work proposes Graph Transformation Policy Network (GTPN), a novel generic method that combines the strengths of graph neural networks and reinforcement learning to learn reactions directly from data with minimal chemical knowledge.

Prediction of Organic Reaction Outcomes Using Machine Learning

TLDR
A model framework for anticipating reaction outcomes that combines the traditional use of reaction templates with the flexibility in pattern recognition afforded by neural networks is reported.

Linking the Neural Machine Translation and the Prediction of Organic Chemistry Reactions

TLDR
A gated recurrent unit based sequence-to-sequence model and a parser to generate input tokens for model from reaction SMILES strings were built to translate 'reactants and reagents' to 'products'.

Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction.

TLDR
It is reported that deep neural networks can learn to resolve reactivity conflicts and to prioritize the most suitable transformation rules.

Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network

TLDR
This paper proposes a template-free approach to efficiently explore the space of product molecules by first pinpointing the reaction center -- the set of nodes and edges where graph edits occur, and demonstrates that the model accuracy rivals the performance of domain experts.

Predicting Electron Paths

TLDR
This work designs a method to extract approximate reaction paths from any dataset of atom-mapped reaction SMILES strings and shows that the model achieves state-of-the-art results on a subset of the UPSTO reaction dataset.

Neural Networks for the Prediction of Organic Chemistry Reactions

TLDR
This work explores the use of neural networks for predicting reaction types, using a new reaction fingerprinting method and combines this predictor with SMARTS transformations to build a system which, given a set of reagents and reactants, predicts the likely products.

A Generative Model For Electron Paths

TLDR
A method to extract approximate reaction paths from any dataset of atom-mapped reaction SMILES strings and shows that the model achieves excellent performance on an important subset of the USPTO reaction dataset, comparing favorably to the strongest baselines.

Machine Learning in Computer-Aided Synthesis Planning.

TLDR
Two critical aspects of CASP and recent machine learning approaches to both challenges are focused on, including the problem of retrosynthetic planning and anticipating the products of chemical reactions, which can be used to validate proposed reactions in a computer-generated synthesis plan.

What's What: The (Nearly) Definitive Guide to Reaction Role Assignment

TLDR
A new fingerprint-based and data-driven approach to assign reaction roles which is also applicable to rather unbalanced and noisy reaction schemes and is available as part of the RDKit toolkit.
...