Exploration and Evaluation of Machine Learning-Based Models for Predicting Enzymatic Reactions

Watanabe, Naoki; Morimoto, Masahiro; Ogawa, Teppei; Vavricka, Christopher J.; Kondo, Akihiko; Ogino, Chiaki; Araki, Michihiro

doi:10.1021/acs.jcim.9b00877

“…We acknowledge that different approaches, such as machine learning of structure-activity relationships of enzymes and substrates are a very promising alternative for large datasets, with a number of studies published recently. 59,60 However, for regimes of little data, as presented in this study, we believe that simple heuristic scoring schemes are a more robust and interpretable route toward success, and estimate the performance of EHreact to be satisfactory for use in computer-aided pathway design. As outlook, we plan to utilize EHreact to design multi-step synthesis pathways and enzymatic cascades.…”

Section: Discussionmentioning

confidence: 88%

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

Heid

¹

,

Goldman²,

Sankaranarayanan³

et al. 2021

Preprint

0

View full text Add to dashboard Cite

Data-driven computer-aided synthesis planning utilizing organic or biocatalyzed reactions from large databases has gained increasing interest in the last decade, sparking the development of numerous tools to extract, apply and score general reaction templates. The generation of reaction rules for enzymatic reactions is especially challenging, since substrate promiscuity varies between enzymes, causing the optimal levels of rule specificity and optimal number of included atoms to differ between enzymes. This complicates an automated extraction from databases and has promoted the creation of manually curated reaction rule sets. Here we present EHreact, a purely data-driven open-source software tool to extract and score reaction rules from sets of reactions known to be catalyzed by an enzyme at appropriate levels of specificity without expert knowledge. EHreact extracts and groups reaction rules into tree-like structures, Hasse diagrams, based on common substructures in the imaginary transition structures. Each diagram can be utilized to output a single or a set of reaction rules, as well as calculate the probability of a new substrate to be processed by the given enzyme by inferring information about the reactive site of the enzyme from the known reactions and their grouping in the template tree. EHreact heuristically predicts the activity of a given enzyme on a new substrate, outperforming current approaches in accuracy and functionality.

show abstract

“…We acknowledge that different approaches, such as machine learning of structure-activity relationships of enzymes and substrates are a very promising alternative for large datasets, with a number of studies published recently. 61,62 However, for regimes of little data, as presented in this study, we believe that simple heuristic scoring schemes are a more robust and interpretable route toward success, and estimate the performance of EHreact to be satisfactory for use in computer-aided pathway design. We plan to utilize EHreact to design multi-step synthesis pathways and enzymatic cascades.…”

Section: Discussionmentioning

confidence: 88%

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

Heid

¹

,

Goldman

²

,

Sankaranarayanan

³

et al. 2021

Preprint

0

View full text Add to dashboard Cite

Data-driven computer-aided synthesis planning utilizing organic or biocatalyzed reactions from large databases has gained increasing interest in the last decade, sparking the development of numerous tools to extract, apply and score general reaction templates. The generation of reaction rules for enzymatic reactions is especially challenging, since substrate promiscuity varies between enzymes, causing the optimal levels of rule specificity and optimal number of included atoms to differ between enzymes. This complicates an automated extraction from databases and has promoted the creation of manually curated reaction rule sets. Here we present EHreact, a purely data-driven open-source software tool to extract and score reaction rules from sets of reactions known to be catalyzed by an enzyme at appropriate levels of specificity without expert knowledge. EHreact extracts and groups reaction rules into tree-like structures, Hasse diagrams, based on common substructures in the imaginary transition structures. Each diagram can be utilized to output a single or a set of reaction rules, as well as calculate the probability of a new substrate to be processed by the given enzyme by inferring information about the reactive site of the enzyme from the known reactions and their grouping in the template tree. EHreact heuristically predicts the activity of a given enzyme on a new substrate, outperforming current approaches in accuracy and functionality.

show abstract

“…We acknowledge that different approaches, such as machine learning of structure− activity relationships of enzymes and substrates, are a very promising alternative for large data sets, with a number of studies published recently. 61,62 However, for regimes of little data, as presented in this study, we believe that simple heuristic scoring schemes are a more robust and interpretable route toward success and estimate the performance of EHreact to be satisfactory for use in computer-aided pathway design. We plan to utilize EHreact to design multistep synthesis pathways and enzymatic cascades.…”

Section: ■ Conclusionmentioning

confidence: 89%

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

Heid

¹

,

Goldman

²

,

Sankaranarayanan

³

et al. 2021

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Data-driven computer-aided synthesis planning utilizing organic or biocatalyzed reactions from large databases has gained increasing interest in the last decade, sparking the development of numerous tools to extract, apply, and score general reaction templates. The generation of reaction rules for enzymatic reactions is especially challenging since substrate promiscuity varies between enzymes, causing the optimal levels of rule specificity and optimal number of included atoms to differ between enzymes. This complicates an automated extraction from databases and has promoted the creation of manually curated reaction rule sets. Here, we present EHreact, a purely data-driven open-source software tool, to extract and score reaction rules from sets of reactions known to be catalyzed by an enzyme at appropriate levels of specificity without expert knowledge. EHreact extracts and groups reaction rules into tree-like structures, Hasse diagrams, based on common substructures in the imaginary transition structures. Each diagram can be utilized to output a single or a set of reaction rules, as well as calculate the probability of a new substrate to be processed by the given enzyme by inferring information about the reactive site of the enzyme from the known reactions and their grouping in the template tree. EHreact heuristically predicts the activity of a given enzyme on a new substrate, outperforming current approaches in accuracy and functionality.

show abstract

Exploration and Evaluation of Machine Learning-Based Models for Predicting Enzymatic Reactions

Cited by 22 publications

References 48 publications

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

EHreact: Extended Hasse Diagrams for the Extraction and Scoring of Enzymatic Reaction Templates

Contact Info

Product

Resources

About