scholarly journals Predicting Enzymatic Reactions with a Molecular Transformer

2021 ◽  
Author(s):  
David Kreutter ◽  
Philippe Schwaller ◽  
Jean-Louis Reymond
Keyword(s):  
Organic Synthesis ◽  
Enzymatic Reactions ◽  
Patent Office ◽  
Reaction Products ◽  
Specific Enzyme ◽  
Machine Learning Model ◽  
The Us ◽  
Task Transfer ◽  
Synthetic Routes ◽  
Transformer Model

<p>The use of enzymes for organic synthesis allows for simplified, more economical and selective synthetic routes not accessible to conventional reagents. However, predicting whether a particular molecule might undergo a specific enzyme transformation is very difficult. <a>Here we used multi-task transfer learning to train the Molecular Transformer, a sequence-to-sequence machine learning model, with one million reactions from the US Patent Office (USPTO) database combined with 32,181 enzymatic transformations annotated with a text description of the enzyme. The resulting Enzymatic Transformer model predicts the structure and stereochemistry of enzyme-catalyzed reaction products with remarkable accuracy. One of the key novelties is that we combined the reaction SMILES language of only 405 atomic tokens with thousands of human language tokens describing the enzymes, such that our Enzymatic Transformer not only learned to interpret SMILES, but also the natural language as used by human experts to describe enzymes and their mutations.</a></p>

scholarly journals Predicting Enzymatic Reactions with a Molecular Transformer

2020 ◽  
Author(s):  
David Kreutter ◽  
Philippe Schwaller ◽  
Jean-Louis Reymond
Keyword(s):  
Kinetic Resolution ◽  
Computer Assisted ◽  
Enzymatic Reactions ◽  
Organic Transformations ◽  
Machine Learning Model ◽  
The Us ◽  
Task Transfer ◽  
Synthetic Routes ◽  
Transformer Model ◽  
Resolution Processes

The use of enzymes for organic synthesis allows for simplified, more economical and selective synthetic routes not accessible to conventional reagents. However, predicting whether a particular molecule might undergo a specific enzyme transformation is very difficult. Here we exploited recent advances in computer assisted synthetic planning (CASP) by considering the Molecular Transformer, which is a sequence-to-sequence machine learning model that can be trained to predict the products of organic transformations, including their stereochemistry, from the structure of reactants and reagents. We used multi-task transfer learning to train the Molecular Transformer with one million reactions from the US Patent Office (USPTO) database as a source of general chemistry knowledge combined with 32,000 enzymatic transformations, each one annotated with a text description of the enzyme. We show that the resulting Enzymatic Transformer model predicts the products formed from a given substrate and enzyme with remarkable accuracy, including typical kinetic resolution processes.

scholarly journals Predicting Enzymatic Reactions with a Molecular Transformer

2020 ◽  
Author(s):  
David Kreutter ◽  
Philippe Schwaller ◽  
Jean-Louis Reymond
Keyword(s):  
Kinetic Resolution ◽  
Computer Assisted ◽  
Enzymatic Reactions ◽  
Organic Transformations ◽  
Machine Learning Model ◽  
The Us ◽  
Task Transfer ◽  
Synthetic Routes ◽  
Transformer Model ◽  
Resolution Processes

The use of enzymes for organic synthesis allows for simplified, more economical and selective synthetic routes not accessible to conventional reagents. However, predicting whether a particular molecule might undergo a specific enzyme transformation is very difficult. Here we exploited recent advances in computer assisted synthetic planning (CASP) by considering the Molecular Transformer, which is a sequence-to-sequence machine learning model that can be trained to predict the products of organic transformations, including their stereochemistry, from the structure of reactants and reagents. We used multi-task transfer learning to train the Molecular Transformer with one million reactions from the US Patent Office (USPTO) database as a source of general chemistry knowledge combined with 32,000 enzymatic transformations, each one annotated with a text description of the enzyme. We show that the resulting Enzymatic Transformer model predicts the products formed from a given substrate and enzyme with remarkable accuracy, including typical kinetic resolution processes.

scholarly journals Predicting Enzymatic Reactions with a Molecular Transformer

2021 ◽  
Author(s):  
David Kreutter ◽  
Philippe Schwaller ◽  
Jean-Louis Reymond
Keyword(s):  
Organic Synthesis ◽  
Enzymatic Reactions ◽  
Synthetic Routes

The use of enzymes for organic synthesis allows for simplified, more economical and selective synthetic routes not accessible to conventional reagents. However, predicting whether a particular molecule might undergo a...

scholarly journals Enzymatic assay of d-mannose in serum

1997 ◽  
Vol 43 (3) ◽  
pp. 533-538 ◽  
Author(s):  
James R Etchison ◽  
Hudson H Freeze
Keyword(s):  
Bacillus Stearothermophilus ◽  
Anion Exchange Chromatography ◽  
Exchange Chromatography ◽  
Enzymatic Assay ◽  
Enzymatic Reactions ◽  
Reaction Products ◽  
Serum Samples ◽  
Healthy Human ◽  
Human Volunteers ◽  
Chromatography Step

Abstract We describe a new and improved enzymatic assay for determining the concentration of d-mannose in sera. Serum d-glucose is selectively converted to glucose-6 phosphate with the highly specific thermostable glucokinase (EC 2.7.1.2) from Bacillus stearothermophilus. The anionic reaction products and excess substrates are removed by a rapid and simple anion-exchange chromatography step in microcentrifuge spin columns. d-Mannose in the glucose-depleted sample is then assayed spectrophotometrically by using coupled enzymatic reactions. The quantitative elimination of glucose from the serum samples allowed the accurate and reproducible assay of serum mannose in the 0–200 μmol/L range. Recovery of mannose added to serum (5–200 μmol/L) was 94% ± 4.4%. The intraassay CV was 6.7% at 40 μmol/L mannose (n = 5; 39.6 ± 1.6 μmol/L) and 4.4% at 80 μmol/L (n = 11; 75.0 ± 1.8 μmol/L); the interassay CV at these concentrations was 12.2% (n = 7; 36.9 ± 2.1 μmol/L) and 9.8% (n = 7; 74.2 ± 2.7 μmol/L), respectively. Sera from 11 healthy human volunteers contained an average of 54.1 ± 11.9 μmol/L mannose (range 36–81 μmol/L).

scholarly journals Datasets and Their Influence on the Development of Computer Assisted Synthesis Planning Tools in the Pharmaceutical Domain

2019 ◽  
Author(s):  
Amol Thakkar ◽  
Thierry Kogej ◽  
Jean-Louis Reymond ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum
Keyword(s):  
Model Performance ◽  
Computer Assisted ◽  
Patent Office ◽  
United States Patent ◽  
Planning Tool ◽  
Virtual Libraries ◽  
Synthesis Planning ◽  
Synthetic Routes ◽  
The Cost ◽  
States Patent

<p>Computer Assisted Synthesis Planning (CASP) has gained considerable interest as of late. Herein we investigate a template-based retrosynthetic planning tool, trained on a variety of datasets consisting of up to 17.5 million reactions. We demonstrate that models trained on datasets such as internal Electronic Laboratory Notebooks (ELN), and the publicly available United States Patent Office (USPTO) extracts, are sufficient for the prediction of full synthetic routes to compounds of interest in medicinal chemistry. As such we have assessed the models on 1,731 compounds from 41 virtual libraries for which experimental results were known. Furthermore, we show that accuracy is a misleading metric for assessment of the ‘filter network’, and propose that the number of successfully applied templates, in conjunction with the overall ability to generate full synthetic routes be examined instead. To this end we found that the specificity of the templates comes at the cost of generalizability, and overall model performance. This is supplemented by a comparison of the underlying datasets and their corresponding models.</p>

Transition Metal Complexes in Organic Synthesis, Part 73. Synthetic Routes to Naturally Occurring Furocarbazoles

Heterocycles ◽  
2004 ◽  
Vol 63 (10) ◽  
pp. 2393 ◽  
Author(s):  
Hans-Joachim Knölker ◽  
Wolfgang Fröhner ◽  
Micha P. Krahl ◽  
Kethiri R. Reddy
Keyword(s):  
Transition Metal ◽  
Metal Complexes ◽  
Organic Synthesis ◽  
Transition Metal Complexes ◽  
Naturally Occurring ◽  
Synthetic Routes

Nitrosocarbonyl Carbohydrate Derivatives: Hetero Diels–Alder and Ene Reaction Products for Useful Organic Synthesis

Synthesis ◽  
2020 ◽  
Author(s):  
Paolo Quadrelli ◽  
Marco Corti ◽  
Marco Leusciatti ◽  
Mattia Moiola ◽  
Mariella Mella
Keyword(s):  
Organic Synthesis ◽  
Tertiary Amine ◽  
Diels Alder ◽  
Mild Oxidation ◽  
Reaction Products ◽  
Ene Reaction ◽  
Nitrile Oxides

AbstractThe generation and trapping of two new nitrosocarbonyl intermediates bearing carbohydrate-based chiral substituents is achieved by the mild oxidation of the corresponding nitrile oxides with tertiary amine N-oxides. Their capture with suitable dienes and alkenes afforded the corresponding hetero Diels–Alder cycloadducts and ene adducts from fair to excellent yields. The entire methodology looks highly promising by the easy conversion of aldoximes into hydroxymoyl halides, widening the access to nitrosocarbonyls, versatile tools in organic synthesis.

scholarly journals Modelling the Effects of Ageing Time of Starch on the Enzymatic Activity of Three Amylolytic Enzymes

2012 ◽  
Vol 2012 ◽  
pp. 1-13 ◽  
Author(s):  
Nelson P. Guerra ◽  
Lorenzo Pastrana Castro
Keyword(s):  
Enzymatic Activity ◽  
Conformational Changes ◽  
Active Site ◽  
Ageing Time ◽  
Enzymatic Reactions ◽  
Reaction Products ◽  
Inhibitory Effects ◽  
Amylolytic Enzymes ◽  
Modified Forms

The effect of increasing ageing time (t) of starch on the activity of three amylolytic enzymes (Termamyl, San Super, and BAN) was investigated. Although all the enzymatic reactions follow michaelian kinetics,vmaxdecreased significantly (P<0.05) andKMincreased (although not always significantly) with the increase int. The conformational changes produced in the starch chains as a consequence of the ageing seemed to affect negatively the diffusivity of the starch to the active site of the enzymes and the release of the reaction products to the medium. A similar effect was observed when the enzymatic reactions were carried out with unaged starches supplemented with different concentrations of gelatine [G]. The inhibition in the amylolytic activities was best mathematically described by using three modified forms of the Michaelis-Menten model, which included a term to consider, respectively, the linear, exponential, and hyperbolic inhibitory effects oftand [G].

scholarly journals Of Smart Phone Wars and Software Patents

2013 ◽  
Vol 27 (1) ◽  
pp. 67-86 ◽  
Author(s):  
Stuart Graham ◽  
Saurabh Vishnubhakat
Keyword(s):  
Smart Phone ◽  
The United States ◽  
Patent System ◽  
Patent Office ◽  
Competing Values ◽  
Administrative Reforms ◽  
The Us ◽  
Software Patents ◽  
Over Time ◽  
The Web

Among the main criticisms currently confronting the US Patent and Trademark Office are concerns about software patents and what role they play in the web of litigation now proceeding in the smart phone industry. We will examine the evidence on the litigation and the treatment by the Patent Office of patents that include software elements. We present specific empirical evidence regarding the examination by the Patent Office of software patents, their validity, and their role in the smart phone wars. More broadly, this article discusses the competing values at work in the patent system and how the system has dealt with disputes that, like the smart phone wars, routinely erupt over time, in fact dating back to the very founding of the United States. The article concludes with an outlook for systematic policymaking within the patent system in the wake of major recent legislative and administrative reforms. Principally, the article highlights how the US Patent Office acts responsibly when it engages constructively with principled criticisms and calls for reform, as it has during the passage and now implementation of the landmark Leahy–Smith America Invents Act of 2011.

Sign in / Sign up

Export Citation Format

Share Document