Structure-musk odour relationship studies of tetralin and indan compounds using neural networks

Cherqaoui, Driss; Esseffar, M.; Villemin, Didier; Cense, J. M.; Chastrette, M.; Zakarya, Driss

doi:10.1039/a709269e

Cited by 33 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…That is why we choose to estimate their relative contribution. The contribution of each descriptor was estimated from the trained SVM using a technique proposed by Cherqaoui et al [41]. Figure 5 indicates that the relative importance of the descriptors varied in the following order:…”

Section: Descriptors Contributionmentioning

confidence: 99%

QSAR Studies of HEPT Derivatives Using Support Vector Machines

et al. 2009

Self Cite

View full text Add to dashboard Cite

Human Immunodeficiency Virus type 1 reverse transcriptase is an important target for chemotherapeutic agents against the AIDS disease. 1-[2-Hydroxyethoxy-methyl]-6-(phenylthio) thymine] derivatives are potent nonnucleoside reverse transcriptase inhibitors. In the present work, quantitative structure-activity relationship analysis for a set of 79 HEPT derivatives has been investigated by means of support vector machines. The relationships between structure and activity were examined quantitatively using descriptors encoding the steric, hydrophobic, electronic and structural features of 1-[2-hydroxyethoxy-methyl]-6-(phenylthio) thymine] derivatives. The performance and predictive capability of support vector machines method are investigated and compared with other methods such as artificial neural network and multiple linear regression methods. The obtained results indicate that the support vector machines model with the kernel radial basis function can be employed as a forceful tool for quantitative structureactivity relationship studies. The contribution of each descriptor to the structure-activity relationships was evaluated.

show abstract

Section: Descriptors Contributionmentioning

confidence: 99%

QSAR Studies of HEPT Derivatives Using Support Vector Machines

et al. 2009

Self Cite

View full text Add to dashboard Cite

show abstract

“…Aiming to figure out the impact of each input in the model, we performed a sensibility analysis in two ways (Cherqaoui et al, 1998). In the first, the descriptor under study is removed from the model and the statistical coefficient R 2 train for the training set is analyzed.…”

Section: Variables' Interpretation Of the Best Modelmentioning

confidence: 99%

“…In the other method, the mean of the absolute deviation values 4 mi between the observed and the estimated value for all compounds is calculated. Finally, the contribution factor C i (Cherqaoui et al, 1998) of descriptor i (i = 1-11) is given by…”

Section: Variables' Interpretation Of the Best Modelmentioning

confidence: 99%

QSAR studies of imidazopyridine derivatives as Et-PKG inhibitors using the PSO-SVM approach

Cheng

Zhang

2009

Med Chem Res

View full text Add to dashboard Cite

In this work, the multiple linear regression (MLR) and support vector machine (SVM) methods were applied for modeling and predicting the inhibitory activity of imidazopyridine derivatives with two-dimensional (2D) autocorrelation descriptors calculated from the molecular structure alone for the first time. We define the new objective function as fitness, and it can guide the particle swarm optimization (PSO) method to select important descriptors which are responsible for the inhibitory activity of these compounds. The square of the correlation coefficient (R 2 = 0.897), the square of correlation coefficients of the test set R 2 ext ¼ 0:660 À Á , and the obtained statistical parameter of the calibrating set in the PSO-SVM model was 0.743, which demonstrated the reliability of the model. The PSO-SVM model is superior over the PSO-MLR method in the dataset with imidazopyridine derivatives as Et-PKG inhibitors. Our best quantitative structure-activity relationship model illustrates the importance of an adequate distribution of atomic properties represented in topological frames and reveals atomic masses, van der Waals volumes, Sanderson electronegativities, and polarizabilities to be the most influential atomic properties in the structures of the imidazopyridine derivatives.

show abstract

“…Chastrette et al (1995) have investigated 105 molecules comprising nitrobenzene, carbonyl tetralins and carbonyl indans with musk fragrance, using a threelayer back-propagation neural network. In another study Cherqaoui et al (1998) have used the same method, with multilayer neural network using the back-propagation algorithm, to investigate a series of tetralins and indans with musk fragrance. Furthermore, Zakarya et al (1999) tested a classification of camphor odour compounds by means of ANNs with a back-propagation algorithm and K nearest neighbor.…”

Section: Former Workmentioning

confidence: 99%

“…Those charts, also found in wine tasting industry enable to set a standard description of odours but subtle differences may arise from one author to another, especially when considering odour secondary notes. Evaluation of the odour note of a novel molecule is then a key issue that has been attempted for several primary notes through the use of Structure -Odour Relationships (SOR): for camphor (Chastrette et al, 1996), for sandalwood (Zakarya et al, 1997), for musk (Cherqaoui et al, 1998) and for ambergris (Kovatcheva et al, 2004).…”

Section: Former Workmentioning

confidence: 99%

Computer-aided aroma design. II. Quantitative structure–odour relationship

Korichi¹,

Gerbaud²,

Talou³

et al. 2008

Chemical Engineering and Processing: Process Intensification

View full text Add to dashboard Cite

Computer Aided Aroma Design (CAAD) is likely to become a hot issue as the REACH EC document targets many aroma compounds to require substitution. The two crucial steps in CAMD are the generation of candidate molecules and the estimation of properties, which can be difficult when complex molecular structures like odours are sought and their odour quality are definitely subjective or their odour intensity are partly subjective as stated in Rossitier's review (1996). The CAAD methodology and a novel molecular framework were presented in part I. Part II focuses on a classification methodology to characterize the odour quality of molecules based on Structure -Odour Relation (SOR). Using 2D and 3D molecular descriptors, Linear Discriminant Analysis (LDA) and Artificial Neural Network are compared in favour of LDA. The classification into balsamic / non balsamic quality was satisfactorily solved. The classification among five sub notes of the balsamic quality was less successful, partly due to the selection of the Aldrich's Catalog as the reference classification. For the second case, it is shown that the sweet sub note considered in Aldrich's Catalog is not a relevant sub note, confirming the alternative and popular classification of Jaubert et al., (1995), the field of odours.

show abstract

Structure-musk odour relationship studies of tetralin and indan compounds using neural networks

Cited by 33 publications

References 15 publications

QSAR Studies of HEPT Derivatives Using Support Vector Machines

QSAR Studies of HEPT Derivatives Using Support Vector Machines

QSAR studies of imidazopyridine derivatives as Et-PKG inhibitors using the PSO-SVM approach

Computer-aided aroma design. II. Quantitative structure–odour relationship

Contact Info

Product

Resources

About