Search PubMed⌕ Search

Biomedical subjects

M Chastrette

Publications and source records attributed to M Chastrette.

10 recordsLinked to original sources
PubMed1998

Data management in olfaction studies.

Problems arising in the collection of data in olfaction studies are discussed in relation with the specific nature of perception and description of odors. Olfactory data depend strongly on individual physiological differences, on measurement methods and on psychological factors. Classifications of odors are necessary to put some order in odor descriptions which are used in structure-odor relationships. Published classifications have been based on empirical, semi-empirical, or statistical approaches. In the last category data may be obtained using semantic descriptions, or profiles, or similarity estimations. Examples of each kind of classification are presented and discussed. The intensity data are most often threshold data. They also present a huge variability which makes it difficult to relate them to physicochemical properties. Data on thresholds and attempts to standardize them are presented and discussed.

Data Interpretation, Statistical↗
PubMed1998

Modelling the human olfactory stimulus-response function.

Two models, derived from the equations of Michaelis-Menten and Hill, were adapted to olfaction. Their ability to model human olfactory stimulus-response relationships was compared with that of the classical laws of Fechner and Stevens. First, these four models were systematically compared on data available in the literature concerning 20 pure odorous compounds. At the lower concentrations of the odorous compound, the model of Stevens was found to be as good as the model of Hill. However, when the concentration range was extended further and included the concentration at half the maximum intensity, the model of Hill was found to be better. Second, the four models were tested on different parts of a true stimulus-response sigmoid curve with 5% noise added. The comparison confirmed the results obtained when experimental data were used. Third, the hypothesis that the psychophysical response is the sum of sigmoidal responses generated at the more peripheral parts of the olfactory system was examined, assuming a binomial distribution of receptor affinities. Within a very large range of variation in their characteristics, the sums of several sigmoids are indeed correctly modelled by Hill equations with exponents reflecting the distribution of receptor affinities.

Humans↗
PubMed1997

Trends in structure-odor relationships.

A number of structure-odor relationships concerning odor intensity and odor quality are presented. The sets which were used, the statistical methods which were employed and the principal results of the studies are analyzed and discussed. An attempt was made to recognize principal trends in the field of structure-odor relationships.

Humans↗
PubMed1994

A nonlinear map of substituent constants for selecting test series and deriving structure-activity relationships. 1. Aromatic series.

A nonlinear mapping (NLM) analysis was performed on a set of 166 aromatic substituents described by six variables encoding hydrophobic (pi), steric (MR), and electronic effects (HBA, HBD, F, and R). NLM allowed to easily summarize the main information contained in the original data table. By means of collections of graphs, it was possible to relate the structure of the substituents to their pi, MR, HBA, HBD, F, and R values. The proposed approach provides a useful and easy tool for the selection of test series and for deriving structure-activity relationships.

Alanine↗
PubMed1994

A nonlinear map of substituent constants for selecting test series and deriving structure-activity relationships. 2. Aliphatic series.

A nonlinear mapping (NLM) analysis was performed on a set of 103 aliphatic substituents described by five variables encoding hydrophobic (Fr), steric (MR), and electronic effects (HBA, HBD, and F). NLM allowed to easily summarize the main information contained in the original data table. By means of collections of graphs, it was possible to relate the structure of the aliphatic substituents to their Fr, MR, HBA, HBD, and F values. The proposed approach provides a useful and easy tool for the selection of test series and for deriving structure-activity relationships.

Antifungal Agents↗
PubMed1993

Estimating pesticide field half-lives from a backpropagation neural network.

The field half-lives of 110 pesticides were modelled using a backpropagation neural network (NN). The molecules were described by means of the frequency of 17 structural fragments. Before training the NN, different scaling transformations were assayed. Best results were obtained with correspondence factor analysis which also allowed a reduction of dimensionality. The training and testing sets of the NN analysis gave 95.5% and 84.6% of good classifications, respectively. Comparison with discriminant factor analysis showed that a backpropagation NN was more appropriate to model the field half-lives of pesticides.

Discriminant Analysis↗
PubMed1993

Adapting the structure of a neural network to extract chemical information. Application to structure-odour relationships.

Two types of neural networks were used to establish relationships between chemical structure and musk odour of 79 nitrobenzenic compounds. Substituents on the five free sites of the benzene ring (one position was always occupied by a t-butyl group) were described using three volume descriptors and three electronegativity descriptors. Musk odour was coded by a binary variable. First a classical network with two hidden layers containing six and three neurons was used. This network gave a better classification (94%) than that obtained by linear discriminant analysis (81%). The odour was then predicted using a leave-ten-out procedure, with 77% of correct prediction for the whole sample. Then a dual two-way network was built to mimic the symmetry of the problem (two sides on a molecule, two muskophore patterns). This network recognized both patterns already known to chemists and gave 99% of correct classifications by taking into account substitution in all positions. As a side benefit of the modified network structure it was possible to evaluate the influence of each of 19 substituents in each of the five possible positions.

Discriminant Analysis↗
PubMed1989

The stochastic regression analysis as a tool in ecotoxicological QSAR studies.

Correspondence factor analysis (CFA) was used in conjunction with linear regression analysis to examine the structure-activity relationships of 50 benzene derivatives tested on Pimephales promelas. From nine molecular descriptions (numbers of C, H, O, N, Br, Cl, NO2, OH, and NH2 included in the molecules), CFA made it possible to define five new independent variables which were introduced in a stepwise regression analysis procedure to describe the acute toxicity (96-h LC50) of the aromatic compounds. The model log 1/C = -0.727F1 + 1.248F3 + 4.052 (r = 0.918; s = 0.270) is more relevant to describe the ecotoxicological behavior of the studied compounds on the fathead minnow than that obtained with principal components (log 1/C = 0.151 PC1 -0.271 PC2 + 4.124; r = 0.737; s = 0.460). The heuristic potency of this particular statistical analysis, which is called stochastic regression analysis, is discussed in detail.

Animals↗
PubMedDate not supplied

Structure-odor relationships: using neural networks in the estimation of camphoraceous or fruity odors and olfactory thresholds of aliphatic alcohols.

Structure-odor relationships were established for a sample of 99 aliphatic alcohols using a three-layer backpropagation neural network. The molecular structure was described using a common skeleton with six possible substitutions. Substituents were described using only their van der Waals volumes. The discrimination between fruity and camphoraceous odors of 67 compounds gave good results in classification (100%) and prediction (85%) phases. With the global set, the network correctly classified and predicted the camphoraceous character of compounds (100% and 95% respectively) but gave poorer results for the fruity character (87% and 74% respectively). Calculations of pOLs (pOL = -log (olfactory threshold expressed in mol/L)) of 45 camphoraceous compounds were also made. When all camphoraceous compounds were used to establish the model, 91% of the pOLs were correctly estimated. When attempts were made to predict the pOL values of 10% of the compounds from a model designed using 90% of the sample, only 74% of the pOLs were correctly estimated.

Alcohols↗