Atom-to-Atom Mapping: A Benchmarking Study of Popular Mapping Algorithms and Consensus Strategies

Madzhidov, Timur; Lin, Arkadii; Nugmanov, Ramil; Dyubankova, Natalia; Gimadiev, Timur; Wegner, Jörg K.; Rakhimbekova, Assima; Akhmetshin, Tagir; Ibragimova, Zarina; Varnek, Alexandre; Suleymanov, Rail; Ceulemans, Hugo; Verhoeven, Jonas

doi:10.26434/chemrxiv.13012679

Cited by 2 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For 86% (4706) of these reactions, the mappings are fully consistent (Figure 3b), with all atom pairs being consistent between MetaCyc and the results of RDT. This is similar to the reported findings from other comparative analyses (Preciat Gonzalez et al., 2017; Madzhidov et al., 2020, Figure 3a). The remaining 14% (760) of reactions showed inconsistencies between MetaCyc and the results of RDT, which are due to the different algorithms for their generation.…”

Section: Resultssupporting

confidence: 93%

“…RDT, in its original implementation, is not designed to allow multiple mappings; recently Madzhidov et al. (2020) created an extended version of RDT, featuring this functionality. Usage of this new RDT version in the proposed workflow is possible and will be attempted in future extensions.…”

Section: Resultsmentioning

confidence: 99%

“…RDT is open source and can be executed locally, which facilitates automated comparative analyses. At least two evaluations of the existing AAM approaches with different databases of reactions have been conducted (Madzhidov et al., 2020; Preciat Gonzalez, & et al., 2017). In these evaluations, the atom mappings created by several AAM tools, such as DREAM (First et al., 2012), MWED (Latendresse et al., 2012), and ChemAxon AutoMapper (ChemAxon, 2021), were compared against manually curated atom mappings.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An automated workflow that generates atom mappings for large‐scale metabolic models and its application to Arabidopsis thaliana

et al. 2022

View full text Add to dashboard Cite

SUMMARY Quantification of reaction fluxes of metabolic networks can help us understand how the integration of different metabolic pathways determines cellular functions. Yet, intracellular fluxes cannot be measured directly but are estimated with metabolic flux analysis (MFA), which relies on the patterns of isotope labeling of metabolites in the network. The application of MFA also requires a stoichiometric model with atom mappings that are currently not available for the majority of large‐scale metabolic network models, particularly of plants. While automated approaches such as the Reaction Decoder Toolkit (RDT) can produce atom mappings for individual reactions, tracing the flow of individual atoms of the entire reactions across a metabolic model remains challenging. Here we establish an automated workflow to obtain reliable atom mappings for large‐scale metabolic models by refining the outcome of RDT, and apply the workflow to metabolic models of Arabidopsis thaliana. We demonstrate the accuracy of RDT through a comparative analysis with atom mappings from a large database of biochemical reactions, MetaCyc. We further show the utility of our automated workflow by simulating 15N isotope enrichment and identifying nitrogen (N)‐containing metabolites which show enrichment patterns that are informative for flux estimation in future 15N‐MFA studies of A. thaliana. The automated workflow established in this study can be readily expanded to other species for which metabolic models have been established and the resulting atom mappings will facilitate MFA and graph‐theoretic structural analyses with large‐scale metabolic networks.

show abstract

Section: Resultssupporting

confidence: 93%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An automated workflow that generates atom mappings for large‐scale metabolic models and its application to Arabidopsis thaliana

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In the last years, the use of machine learning and data-driven approaches proved to be a very effective way to capture patterns from complex chemistry knowledge collections. 13 The extraction of chemical reaction rules from large data sets of traditional organic chemistry reactions 14 is one of the most successful examples 15 of providing transparency and explainability with AI applications in chemistry. While traditional synthetic organic chemistry went through its renaissance period thanks to recent development in machine learning and the availability of public chemical reaction datasets, the impact in biochemistry remained mostly bounded to the context of metabolic pathways prediction.…”

Section: Introductionmentioning

confidence: 99%

Molecular Transformer-aided Biocatalysed Synthesis Planning

Probst¹,

Manica²,

Teukam³

et al. 2021

Preprint

View full text Add to dashboard Cite

Enzyme catalysts are an integral part of green chemistry strategies towards a more sustainable and resource-efficient chemical synthesis. However, the use of enzymes on unreported substrates and their specific stereo- and regioselectivity are domain-specific knowledge factors that require decades of field experience to master. This makes the retrosynthesis of given targets with biocatalysed reactions a significant challenge. Here, we use the molecular transformer architecture to capture the latent knowledge about enzymatic activity from a large data set of publicly available biochemical reactions, extending forward reaction and retrosynthetic pathway prediction to the domain of biocatalysis. We introduce the use of a class token based on the EC classification scheme that allows to capture catalysis patterns among different enzymes belonging to the same hierarchical families. The forward prediction model achieves an accuracy of 49.6% and 62.7%, top-1 and top-5 respectively, while the single-step retrosynthetic model shows a round-trip accuracy of 39.6% and 42.6%, top-1 and top-10 respectively. Trained models and curated data are made publicly available with the hope of promoting enzymatic catalysis and making green chemistry more accessible through the use of digital technologies.

show abstract

Atom-to-Atom Mapping: A Benchmarking Study of Popular Mapping Algorithms and Consensus Strategies

Cited by 2 publications

References 9 publications

An automated workflow that generates atom mappings for large‐scale metabolic models and its application to Arabidopsis thaliana

An automated workflow that generates atom mappings for large‐scale metabolic models and its application to Arabidopsis thaliana

Molecular Transformer-aided Biocatalysed Synthesis Planning

Contact Info

Product

Resources

About