QSPR Study of Fluorescence Wavelengths (λex/λem) Based on the Heuristic Method and Radial Basis Function Neural Networks

University of Paris-Est, La Haye-Descartes, Centre, France
QSAR & Combinatorial Science (Impact Factor: 1.55). 02/2006; 25(2):147 - 155. DOI: 10.1002/qsar.200510142


The Quantitative Structure–Property Relationship (QSPR) method was performed to study the fluorescence excitation wavelengths (λex) and emission wavelengths (λem) of 64 fluorescent probes. The probes included the derivatives of dansyl, bimane, pyrene, benzofurazan, nonaphthalene, coumarin, anthracene and fluorescein, with the wavelength ranging from 300 nm to 600 nm. The Heuristic Method (HM) and Radial Basis Function Neural Networks (RBFNNs) were employed to construct linear and nonlinear prediction models, respectively. The proposed linear models for λex and λem contain five descriptors with the squared correlation coefficients R2 of 0.888 and 0.897, respectively. Better prediction results were obtained from RBFNN model, with the squared correlation coefficients R2 of 0.948 and 0.939 for λex and λem, respectively. The descriptors used in the models were discussed in detail too.

113 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: Quantitative structure-retention relationship (QSRR) models have been successfully developed for the prediction of the retention factor (log k) in the biopartitioning micellar chromatography (BMC) of 66 organic pollutants. Heuristic method (HM) and radial basis function neural networks (RBFNN) were utilized to construct the linear and non-linear QSRR models, respectively. The optimal QSRR model was developed based on a 6-17-1 radial basis function neural network architecture using molecular descriptors calculated from molecular structure alone. The RBFNN model gave a correlation coefficient (R2) of 0.8464 and root-mean-square error (RMSE) of 0.1925 for the test set. This paper provided a useful model for the predicting the log k of other organic compounds when experiment data are unknown.
    Analytica chimica acta 09/2007; 598(1):12-8. DOI:10.1016/j.aca.2007.07.016 · 4.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The quantitative structure-property relationship (QSPR) method was used to model the fluorescence excitation wavelengths (lambda(ex)) of 42 boronic acid-based fluorescent biosensors (30 in the training set and 12 in the test set). In this QSPR study, unsupervised forward selection (UFS), stepwise multiple linear regression (SMLR), partial least squares regression (PLS) and associative neural networks (ASNN) were employed to simulate linear and nonlinear models. All models were validated by a test set and Tropsha's validation model. The resulting ASNN nonlinear model demonstrates significant improvement on the predictive ability of the neural network compared to the SMLR and PLS linear models. The descriptors used in the models are discussed in detail. These QSPR models are useful tools for the prediction of fluorescence excitation wavelengths of arylboronic acids.
    Journal of Molecular Modeling 07/2008; 14(6):441-9. DOI:10.1007/s00894-008-0293-0 · 1.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Many chemicals cause latent harm, such as erratic diseases and change of climate, and therefore it is necessary to evaluate environmentally safe levels of dangerous chemicals. Quantitative Structure-Toxicity Relationship (QSTR) analysis has become an indispensable tool in ecotoxicological risk assessments. Our paper used QSTR to deal with the modeling of the acute toxicity of 92 substituted benzenes. The molecular descriptors representing the structural features of the compounds were calculated by CODESSA program. Heuristic Method (HM) and Radial Basis Function Neural Networks (RBFNNs) were utilized to construct the linear and the nonlinear QSTR models, respectively. The predictive results were in agreement with the experimental values. The optimal QSTR model which was established based on RBFNNs gave a correlation coefficient (R-2) of 0.893, 0.876, 0.889 and Root-Mean-Square Error (RMSE) of 0.220, 0.205, 0.218 for the training set, the test set, and the whole set, respectively. RBFNNs proved to be a v
    QSAR & Combinatorial Science 08/2008; 27(8). DOI:10.1002/qsar.200710096 · 1.55 Impact Factor
Show more