Published:Journal of Chromatographic Science,
ISSN 0021-9665 Volume
45, Number 7, August 2007, pp. 400-404
Quantitative Structure–Property Relationship
Study of Retention Time of Some Pesticides in Gas Chromatography
?M.R. Hadjmohammadi, M.H.Fatemi, and K. Kamel
Department of Chemistry, University of Mazandaran, P.O. Box 453,
Babolsar, Iran
?A quantitative structure–property relationship
(QSPR) study based on multiple linear regression (MLR) and artificial
neural network (ANN) techniques is carried out to investigate
the retention time behavior of some pesticides on the DB-5ms fused-silica
column in gas chromatography. Five descriptors selected in the
MLR model are: first component WHIM index (E1v), highest eigenvalue
n.7 of burden matrix / weighted by atomic van der waals volume
(BEHv7); average connectivity index Chi-2 (X2a), 3D-MoRSE signal
23 weighted by atomic Sanderson electronegativity (MoR23m); and
principal moments of inertia B (PMIB). A 5-5-1 ANN is also generated
to investigate the retention behavior of described pesticides
using the same descriptors MLR model as inputs. The statistical
parameters derived from MLR and ANN for all molecules are: correlation
coefficient (R)MLR = 0.929, standard errors (SE)MLR = 3.452, RANN
= 0.943, and SEANN = 3.112. The mean of relative errors between
the MLR and ANN calculated and the experimental values of the
retention times for the prediction set are 13.8% and 9.04%, respectively.
The correlation coefficient and standard error of ANN model compared
with MLR models showed the superiority of ANNs over regression
models. This is partly due to the fact that ANN considers the
interaction between different parameters as well as nonlinear
relation.
Reproduction
of editorial content of this journal is prohibited without publisher’s
permission.
This
article is available in its entirety by fax for $40.00.
Visa, MasterCard & Amex accepted.
To
order electronicallyclick here or call: 847-647-2900 ext. 1323 or fax request to: 847-647-1155.
Please
indicate JCS volume and issue along with
page numbers.
For permission to reuse material from Journal of Chromatographic Science, please access www.copyright.com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of uses.
