CGRtools: Python Library for Molecule, Reaction, and Condensed Graph of Reaction Processing

Nugmanov, Ramil; Mukhametgaleev, Ravil; Akhmetshin, Tagir; Gimadiev, Timur; Afonina, Valentina A.; Madzhidov, Timur; Varnek, Alexandre

doi:10.1021/acs.jcim.9b00102

Cited by 69 publications

(100 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, a similar workflow was successfully used for the generation of novel molecular structures possessing desirable biological activity 26 . The SMILES/CGR strings were prepared using the in-house CGRtools library 45 . Unlike conventional reaction SMILES strings, SMILES/CGR depicts a pseudo-molecule with extended features corresponding to dynamic bonds and dynamic atoms, see Figure 1.…”

Section: Modeling and Postprocessing Workflowsmentioning

confidence: 99%

“…The initial dataset was preprocessed with in-house scripts based on the CGRtools library. 45 The curation includes the standardization (aromatization and functional group standardization), removal of empty reactions (those where the products and reactants are exactly the same, or no reactants or products are recorded) and reactions with valence errors. For curated reactions, atom-toatom mapping (AAM) was performed using the ChemAxon Automapper tool which is a part of the JChem toolkit.…”

Section: Datasets and Data Curationmentioning

confidence: 99%

“…CGRtools library (version 3) 45 was used for the reactions cleaning, their transformation to CGRs, conversion of CGRs into SMILES/CGR and processing of generated SMILES/CGR back into reactions.…”

Section: Reaction Data Treatmentmentioning

confidence: 99%

“…Unlike the canonical reaction equation, each atom in CGR is present only once, which significantly reduces the length of the string. Moreover, the CGR (and, hence, SMILES/CGR) contains information about the reaction center and its close neighborhood 45 . The conventional reaction equation can be easily derived from CGR.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Discovery of Novel Chemical Reactions by Deep Generative Recurrent Neural Network

Bort¹,

Baskin²,

Сидоров³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Here, we report an application of Artificial Intelligence techniques to generate novel chemical reactions of the given type. A sequence-to-sequence autoencoder was trained on the USPTO reaction database. Each reaction was converted into a single Condensed Graph of Reaction (CGR), followed by their translation into on-purpose developed SMILES/GGR text strings. The autoencoder latent space was visualized on the two-dimensional generative topographic map, from which some zones populated by Suzuki coupling reactions were targeted. These served for the generation of novel reactions by sampling the latent space points and decoding them to SMILES/CGR.<br>

show abstract

Section: Modeling and Postprocessing Workflowsmentioning

confidence: 99%

Section: Datasets and Data Curationmentioning

confidence: 99%

Section: Reaction Data Treatmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Discovery of Novel Chemical Reactions by Deep Generative Recurrent Neural Network

Bort¹,

Baskin²,

Сидоров³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Stored in an SD File, it is mainly employed for machine learning applications, similarity search, and classification. [14,6] Recently, a SMIRKS-like format for CGR was implemented concomitant with the development of Python-based tools to operate on them (CGRTools) [19]. However, this format cannot be used directly for, e.g., string-based comparisons of reactions.…”

Section: Introductionmentioning

confidence: 99%

ReactionCode: a new versatile format for searching, analysis, classification, transform, and encoding/decoding of reactions

Delannée¹,

Nicklaus²

2020

Preprint

View full text Add to dashboard Cite

In the past two decades a lot of different formats for molecules and reactions have been created. These formats were mostly developed for the purposes of identifiers, representation, classification, analysis and data exchange. A lot of efforts have been made on molecule formats but only few for reactions where the endeavors have been made mostly by companies leading to proprietary formats. Here, we developed a new open-source format which allows to encode and decode a reaction into multi-layers machine readable code, which aggregates reactants and products into a condensed graph of reaction (CGR). This format is flexible and can be used in a context of reaction similarity searching and classification. It is also designed for database organization, machine learning applications and as a new transform reaction language.

show abstract

Predicting Highly Enantioselective Catalysts Using Tunable Fragment Descriptors**

et al. 2023

Self Cite

View full text Add to dashboard Cite

Catalyst optimization processes typically rely on inductive and qualitative assumptions of chemists based on screening data. While machine learning models using molecular properties or calculated 3D structures enable quantitative data evaluation, costly quantum chemical calculations are often required. In contrast, readily available binary fingerprint descriptors are timeand cost-efficient, but their predictive performance remains insufficient. Here, we describe a machine learning model based on fragment descriptors, which are fine-tuned for asymmetric catalysis and represent cyclic or polyaromatic hydrocarbons, enabling robust and efficient virtual screening. Using training data with only moderate selectivities, we designed theoretically and validated experimentally new catalysts showing higher selectivities in a challenging asymmetric tetrahydropyran synthesis.

show abstract

CGRtools: Python Library for Molecule, Reaction, and Condensed Graph of Reaction Processing

Cited by 69 publications

References 22 publications

Discovery of Novel Chemical Reactions by Deep Generative Recurrent Neural Network

Discovery of Novel Chemical Reactions by Deep Generative Recurrent Neural Network

ReactionCode: a new versatile format for searching, analysis, classification, transform, and encoding/decoding of reactions

Predicting Highly Enantioselective Catalysts Using Tunable Fragment Descriptors**

Contact Info

Product

Resources

About