James Devillers
Research Subjects
James Devillers has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
Cécile Guillon
1 shared publicationDaniel Domine
1 shared publicationWalter Karcher
1 shared publication
Latest Publications
- 1 May 2000
- Vol. 3 (3) , 275-9
New trends in (Q)SAR modeling with topological indices.
Topological indices are mathematical entities allowing encoding of structures of molecules. The general principle of their calculation is briefly presented and some recent studies dealing with their use in SAR and QSAR are reviewed. No attempt is made to present the topological indices as a panacea for describing molecules; the goal is only to stress that their proper use yields valuable modeling results.
- 1 May 2000
- Vol. 3 (3) , 275-9
New trends in (Q)SAR modeling with topological indices.
Topological indices are mathematical entities allowing encoding of structures of molecules. The general principle of their calculation is briefly presented and some recent studies dealing with their use in SAR and QSAR are reviewed. No attempt is made to present the topological indices as a panacea for describing molecules; the goal is only to stress that their proper use yields valuable modeling results.
- 1 January 2000
- Published by Taylor & Francis in Toxicology Mechanisms and Methods
- Vol. 10 (3) , 181-193
A NEURAL NETWORK SAR MODEL FOR ALLERGIC CONTACT DERMATITIS
Allergic contact dermatitis results for 259 compounds were used for deriving a qualitative structure-activity relationship (SAR) model. Chemicals were described by means of one physicochemical descriptor, one topological index, and twelve structural alerts (i.e., a value of 1 indicated the presence of the structural feature in a molecule, 0 the absence). A three-layer feed forward neural network trained by the back-propagation algorithm was used as statistical engine. The comparison of the simulation performances of the obtained model to those produced by a classical linear discriminant analysis clearly revealed the usefulness of the nonlinear methods of modeling skin sensitization.
- 1 January 2000
- Published by Taylor & Francis in Toxicology Mechanisms and Methods
- Vol. 10 (3) , 181-193
A NEURAL NETWORK SAR MODEL FOR ALLERGIC CONTACT DERMATITIS
Allergic contact dermatitis results for 259 compounds were used for deriving a qualitative structure-activity relationship (SAR) model. Chemicals were described by means of one physicochemical descriptor, one topological index, and twelve structural alerts (i.e., a value of 1 indicated the presence of the structural feature in a molecule, 0 the absence). A three-layer feed forward neural network trained by the back-propagation algorithm was used as statistical engine. The comparison of the simulation performances of the obtained model to those produced by a classical linear discriminant analysis clearly revealed the usefulness of the nonlinear methods of modeling skin sensitization.
- 1 January 2000
- Published by Taylor & Francis in Toxicology Mechanisms and Methods
- Vol. 10 (1) , 69-79
PREDICTION OF TOXICITY OF ORGANOPHOSPHORUS INSECTICIDES AGAINST THE MIDGE, CHIRONOMUS RIPARIUS, VIA A QSAR NEURAL NETWORK MODEL INTEGRATING ENVIRONMENTAL VARIABLES
A quantitative structure-activity relationship (QSAR) investigation was done for estimating the toxicity of organophosphorus compounds measured against midge larvae (Chironomus riparius) under varying temperature (11, 18, and 25°C) and pH (6, 7, and 8) conditions and with or without sediment. Chemicals were described by means of an autocorrelation vector encoding lipophilicity. A three-layer feedforward neural network trained by the back-propagation algorithm was used as the statistical engine for deriving a powerful QSAR model accounting for environmental constraints.
- 1 January 2000
- Published by Taylor & Francis in Toxicology Mechanisms and Methods
- Vol. 10 (1) , 69-79
PREDICTION OF TOXICITY OF ORGANOPHOSPHORUS INSECTICIDES AGAINST THE MIDGE, CHIRONOMUS RIPARIUS, VIA A QSAR NEURAL NETWORK MODEL INTEGRATING ENVIRONMENTAL VARIABLES
A quantitative structure-activity relationship (QSAR) investigation was done for estimating the toxicity of organophosphorus compounds measured against midge larvae (Chironomus riparius) under varying temperature (11, 18, and 25°C) and pH (6, 7, and 8) conditions and with or without sediment. Chemicals were described by means of an autocorrelation vector encoding lipophilicity. A three-layer feedforward neural network trained by the back-propagation algorithm was used as the statistical engine for deriving a powerful QSAR model accounting for environmental constraints.
- 1 September 1998
- Published by American Geophysical Union (AGU) in Journal of Pharmaceutical Sciences
- Vol. 87 (9) , 1086-1090
Simulating Lipophilicity of Organic Molecules with a Back-Propagation Neural Network
From a training set of 7200 chemicals, a back-propagation neural network (BNN) model was developed for calculating the 1-octanol/water partition coefficient (log P) of molecules containing nitrogen, oxygen, halogen, phosphorus, and/or sulfur atoms. Chemicals were described by means of autocorrelation vectors encoding hydrophobicity, molar refractivity, H-bonding acceptor ability, and H-bonding donor ability. A 35/32/1 composite network composed of four configurations was selected as the final model (root-mean-square error (RMS) = 0.37, r = 0.97) because it provided the best simulation results (RMS = 0.39, r = 0.98) on an external testing set of 519 molecules. This final model compared favorably with a recently published BNN model using variables (atoms and bonds) derived from connection matrices.
- 1 September 1998
- Published by American Geophysical Union (AGU) in Journal of Pharmaceutical Sciences
- Vol. 87 (9) , 1086-1090
Simulating Lipophilicity of Organic Molecules with a Back-Propagation Neural Network
From a training set of 7200 chemicals, a back-propagation neural network (BNN) model was developed for calculating the 1-octanol/water partition coefficient (log P) of molecules containing nitrogen, oxygen, halogen, phosphorus, and/or sulfur atoms. Chemicals were described by means of autocorrelation vectors encoding hydrophobicity, molar refractivity, H-bonding acceptor ability, and H-bonding donor ability. A 35/32/1 composite network composed of four configurations was selected as the final model (root-mean-square error (RMS) = 0.37, r = 0.97) because it provided the best simulation results (RMS = 0.39, r = 0.98) on an external testing set of 519 molecules. This final model compared favorably with a recently published BNN model using variables (atoms and bonds) derived from connection matrices.