Inferring experimental procedures from text-based representations of chemical reactions

Abstract The experimental execution of chemical reactions is a context-dependent and time-consuming process, often solved using the experience collected over multiple decades of laboratory work or searching similar, already executed, experimental protocols. Although data-driven schemes, such as retr...

Full description

Saved in:
Bibliographic Details
Main Authors: Alain C. Vaucher, Philippe Schwaller, Joppe Geluykens, Vishnu H. Nair, Anna Iuliano, Teodoro Laino
Format: Article
Language:English
Published: Nature Portfolio 2021-05-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-021-22951-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849221115814084608
author Alain C. Vaucher
Philippe Schwaller
Joppe Geluykens
Vishnu H. Nair
Anna Iuliano
Teodoro Laino
author_facet Alain C. Vaucher
Philippe Schwaller
Joppe Geluykens
Vishnu H. Nair
Anna Iuliano
Teodoro Laino
author_sort Alain C. Vaucher
collection DOAJ
description Abstract The experimental execution of chemical reactions is a context-dependent and time-consuming process, often solved using the experience collected over multiple decades of laboratory work or searching similar, already executed, experimental protocols. Although data-driven schemes, such as retrosynthetic models, are becoming established technologies in synthetic organic chemistry, the conversion of proposed synthetic routes to experimental procedures remains a burden on the shoulder of domain experts. In this work, we present data-driven models for predicting the entire sequence of synthesis steps starting from a textual representation of a chemical equation, for application in batch organic chemistry. We generated a data set of 693,517 chemical equations and associated action sequences by extracting and processing experimental procedure text from patents, using state-of-the-art natural language models. We used the attained data set to train three different models: a nearest-neighbor model based on recently-introduced reaction fingerprints, and two deep-learning sequence-to-sequence models based on the Transformer and BART architectures. An analysis by a trained chemist revealed that the predicted action sequences are adequate for execution without human intervention in more than 50% of the cases.
format Article
id doaj-art-dcccf88e9c3d4e90a2cd8b02fcfe56f3
institution Kabale University
issn 2041-1723
language English
publishDate 2021-05-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-dcccf88e9c3d4e90a2cd8b02fcfe56f32024-11-24T12:35:23ZengNature PortfolioNature Communications2041-17232021-05-0112111110.1038/s41467-021-22951-1Inferring experimental procedures from text-based representations of chemical reactionsAlain C. Vaucher0Philippe Schwaller1Joppe Geluykens2Vishnu H. Nair3Anna Iuliano4Teodoro Laino5IBM Research EuropeIBM Research EuropeIBM Research EuropeIBM Research EuropeDipartimento di Chimica e Chimica Industriale, Università di PisaIBM Research EuropeAbstract The experimental execution of chemical reactions is a context-dependent and time-consuming process, often solved using the experience collected over multiple decades of laboratory work or searching similar, already executed, experimental protocols. Although data-driven schemes, such as retrosynthetic models, are becoming established technologies in synthetic organic chemistry, the conversion of proposed synthetic routes to experimental procedures remains a burden on the shoulder of domain experts. In this work, we present data-driven models for predicting the entire sequence of synthesis steps starting from a textual representation of a chemical equation, for application in batch organic chemistry. We generated a data set of 693,517 chemical equations and associated action sequences by extracting and processing experimental procedure text from patents, using state-of-the-art natural language models. We used the attained data set to train three different models: a nearest-neighbor model based on recently-introduced reaction fingerprints, and two deep-learning sequence-to-sequence models based on the Transformer and BART architectures. An analysis by a trained chemist revealed that the predicted action sequences are adequate for execution without human intervention in more than 50% of the cases.https://doi.org/10.1038/s41467-021-22951-1
spellingShingle Alain C. Vaucher
Philippe Schwaller
Joppe Geluykens
Vishnu H. Nair
Anna Iuliano
Teodoro Laino
Inferring experimental procedures from text-based representations of chemical reactions
Nature Communications
title Inferring experimental procedures from text-based representations of chemical reactions
title_full Inferring experimental procedures from text-based representations of chemical reactions
title_fullStr Inferring experimental procedures from text-based representations of chemical reactions
title_full_unstemmed Inferring experimental procedures from text-based representations of chemical reactions
title_short Inferring experimental procedures from text-based representations of chemical reactions
title_sort inferring experimental procedures from text based representations of chemical reactions
url https://doi.org/10.1038/s41467-021-22951-1
work_keys_str_mv AT alaincvaucher inferringexperimentalproceduresfromtextbasedrepresentationsofchemicalreactions
AT philippeschwaller inferringexperimentalproceduresfromtextbasedrepresentationsofchemicalreactions
AT joppegeluykens inferringexperimentalproceduresfromtextbasedrepresentationsofchemicalreactions
AT vishnuhnair inferringexperimentalproceduresfromtextbasedrepresentationsofchemicalreactions
AT annaiuliano inferringexperimentalproceduresfromtextbasedrepresentationsofchemicalreactions
AT teodorolaino inferringexperimentalproceduresfromtextbasedrepresentationsofchemicalreactions