ChemInform Abstract: Computer-Assisted Topological Analysis and Completion of Chemical Reactions.

ChemInform ◽  
1990 ◽  
Vol 21 (49) ◽  
Author(s):  
C. BEHNKE ◽  
J. BARGON
Keyword(s):  
Chemical Reactions ◽  
Topological Analysis ◽  
Computer Assisted

Valence states of atoms and their conversions

1984 ◽  
Vol 49 (5) ◽  
pp. 1247-1261 ◽  
Author(s):  
Jaroslav Koča ◽  
Milan Kratochvíl ◽  
Milan Kunz ◽  
Vladimír Kvasnička
Keyword(s):  
Chemical Reactions ◽  
Computer Assisted ◽  
Efficient Tool ◽  
Common Features ◽  
Systematic Application ◽  
Organic Syntheses ◽  
Valence States ◽  
Algebraic Formalism

The algebraic formalism for the description of valence states of atoms and their interconversions is elaborated. It offers a possibility to construct and trace mechanistic paths of chemical reactions, the problem of which is of great importance in computer-assisted organic syntheses. Its systematic application gives exhaustive lists of possible mechanistic paths, and furthermore, very efficient tool to classify chemical reactions and look for their common features.

Reaction of Tropone with a Homopyrrole. The Result of a Computer-Assisted Search for Unique Chemical Reactions

1988 ◽  
Vol 27 (11) ◽  
pp. 1558-1559 ◽  
Author(s):  
Dietmar Forstmeyer ◽  
Johannes Bauer ◽  
Eric Fontain ◽  
Rainer Herges ◽  
Rudolf Herrmann ◽  
...  
Keyword(s):  
Chemical Reactions ◽  
Computer Assisted ◽  
Unique Chemical

scholarly journals DRACON: Disconnected Graph Neural Network for Atom Mapping in Chemical Reactions

2020 ◽  
Author(s):  
Filipp Nikitin ◽  
Olexandr Isayev ◽  
Vadim Strijov
Keyword(s):  
Neural Network ◽  
Chemical Reactions ◽  
Computer Assisted ◽  
Excellent Performance ◽  
Reaction Prediction ◽  
Latent Vector ◽  
Atom Mapping ◽  
Proposed Model ◽  
Disconnected Graphs ◽  
Disconnected Graph

<p>Machine learning solved many challenging problems in computer-assisted synthesis prediction (CASP). We formulate a reaction prediction problem in terms of node-classification in a disconnected graph of source molecules and generalize a graph convolution neural network for disconnected graphs. Here we demonstrate that our approach can successfully predict reaction outcome and atom-mapping during a chemical transformation. A set of experiments using the USPTO dataset demonstrates excellent performance and interpretability of the proposed model. Implicitly learned latent vector representation of chemical reactions strongly correlates with the class of the chemical reaction. Reactions with similar templates group together in the latent vector space.</p>

Topological analysis of energy hypersurface

1980 ◽  
Vol 45 (9) ◽  
pp. 2463-2473 ◽  
Author(s):  
Jiří Pancíř
Keyword(s):  
Total Energy ◽  
Chemical Reactions ◽  
Reaction Mechanisms ◽  
Driving Force ◽  
Topological Analysis ◽  
Matrix Elements ◽  
Qualitative Information ◽  
Linear Change ◽  
Hartree Fock ◽  
Energy Hypersurface

Topological approximations discussed in the preceeding paper are used for construction of topological energy hypersurfaces of chemical reactions on two levels of a sophistication. A method of a linear change of off diagonal matrix elements reproduces adequately Woodward-Hoffmann rules but it provides only qualitative information about energetical relations. A physically more rigorous hypothesis where the reaction driving force is a function of hybridization changes at reaction sites provides energy hypersurfaces which are correct from a quantitative viewpoint. The method suggested treats the total energy as a function of abstract reaction angles which reflect topological changes accompanying the reaction course. A reaction mechanisms analysis leads then to the analysis of topological energy hypersurfaces which can be treated similarly to that of conventional geometrical hypersurfaces. Formulas for atomic integrals necessary in the construction of Hartree-Fock and CI matrices are derived in the basis of reaction angles mentioned and approximations leading to the semiempirical type of a calculation are discussed.

Topological aspects of chemical reactivity. Valency changes in the course of chemical reactions

1987 ◽  
Vol 52 (11) ◽  
pp. 2603-2612 ◽  
Author(s):  
Robert Ponec
Keyword(s):  
Electron Density ◽  
Chemical Reactions ◽  
Chemical Treatment ◽  
Quantum Chemical ◽  
Chemical Reactivity ◽  
Topological Analysis ◽  
Topological Description

The Jug and Gopinathan approach to the use of valency changes for the characterization of chemical reactions is generalized by incorporating it into the framework of the recently proposed topological description of chemical reactivity in terms of the overlap determinant method. The conclusions of the simple topological analysis agree with the results of the direct quantum chemical treatment, indicating thus that the overlap determinant method depicts correctly the principal features of the electron density reorganization occuring during the chemical reactions.

scholarly journals DRACON: Disconnected Graph Neural Network for Atom Mapping in Chemical Reactions

2020 ◽  
Author(s):  
Filipp Nikitin ◽  
Olexandr Isayev ◽  
Vadim Strijov
Keyword(s):  
Neural Network ◽  
Chemical Reactions ◽  
Computer Assisted ◽  
Reaction Prediction ◽  
Latent Vector ◽  
Atom Mapping ◽  
Proposed Model ◽  
Unsupervised Approach ◽  
Disconnected Graphs ◽  
Disconnected Graph

Machine learning solved many challenging problems in computer-assisted synthesis prediction (CASP). We formulate a reaction prediction problem in terms of node-classification in a disconnected graph of source molecules and generalize a graph convolution neural network for disconnected graphs. Here we demonstrate that our approach can successfully predict reaction outcome and atom-mapping during a chemical transformation. A set of experiments using the USPTO dataset demonstrates excellent performance and interpretability of the proposed model. Our model uses an unsupervised approach to atom-mapping and bridges the gap between data-driven and traditional rule-based methods. Implicitly learned latent vector representation of chemical reactions strongly correlates with the class of the chemical reaction. Reactions with similar templates group together in the latent vector space.

scholarly journals DRACON: Disconnected Graph Neural Network for Atom Mapping in Chemical Reactions

2020 ◽  
Author(s):  
Filipp Nikitin ◽  
Olexandr Isayev ◽  
Vadim Strijov
Keyword(s):  
Neural Network ◽  
Chemical Reactions ◽  
Computer Assisted ◽  
Excellent Performance ◽  
Reaction Prediction ◽  
Latent Vector ◽  
Atom Mapping ◽  
Proposed Model ◽  
Disconnected Graphs ◽  
Disconnected Graph

<p>Machine learning solved many challenging problems in computer-assisted synthesis prediction (CASP). We formulate a reaction prediction problem in terms of node-classification in a disconnected graph of source molecules and generalize a graph convolution neural network for disconnected graphs. Here we demonstrate that our approach can successfully predict reaction outcome and atom-mapping during a chemical transformation. A set of experiments using the USPTO dataset demonstrates excellent performance and interpretability of the proposed model. Implicitly learned latent vector representation of chemical reactions strongly correlates with the class of the chemical reaction. Reactions with similar templates group together in the latent vector space.</p>

scholarly journals Computer-assisted topological analysis of renal allograft inflammation adds to risk evaluation at diagnosis of humoral rejection

2017 ◽  
Vol 92 (1) ◽  
pp. 214-226 ◽  
Author(s):  
Antoine Sicard ◽  
Vannary Meas-Yedid ◽  
Maud Rabeyrin ◽  
Alice Koenig ◽  
Stephanie Ducreux ◽  
...  
Keyword(s):  
Risk Evaluation ◽  
Renal Allograft ◽  
Topological Analysis ◽  
Computer Assisted ◽  
Humoral Rejection

3-D organization of fibrinogen receptors using computer reconstruction and whole-mount microscopy

1988 ◽  
Vol 46 ◽  
pp. 30-31
Author(s):  
E. T. O'Toole ◽  
R. R. Hantgan ◽  
J. C. Lewis
Keyword(s):  
Whole Blood ◽  
Colloidal Gold ◽  
Blood Platelets ◽  
Cell Contact ◽  
Computer Assisted ◽  
Adherent Cells ◽  
Differential Centrifugation ◽  
Computer Reconstruction ◽  
Sds Page ◽  
Whole Mount

Thrombocytes (TC), the avian equivalent of blood platelets, support hemostasis by aggregating at sites of injury. Studies in our lab suggested that fibrinogen (fib) is a requisite cofactor for TC aggregation but operates by an undefined mechanism. To study the interaction of fib with TC and to identify fib receptors on cells, fib was purified from pigeon plasma, conjugated to colloidal gold and used both to facilitate aggregation and as a receptor probe. Described is the application of computer assisted reconstruction and stereo whole mount microscopy to visualize the 3-D organization of fib receptors at sites of cell contact in TC aggregates and on adherent cells.Pigeon TC were obtained from citrated whole blood by differential centrifugation, washed with Ca++ free Hank's balanced salts containing 0.3% EDTA (pH 6.5) and resuspended in Ca++ free Hank's. Pigeon fib was isolated by precipitation with PEG-1000 and the purity assessed by SDS-PAGE. Fib was conjugated to 25nm colloidal gold by vortexing and the conjugates used as the ligand to identify fib receptors.

Sign in / Sign up

Export Citation Format

Share Document