A new 3D molecular structure representation using quantum topology with application to structure-property relationships

Citation
Bk. Alsberg et al., A new 3D molecular structure representation using quantum topology with application to structure-property relationships, CHEM INTELL, 54(2), 2000, pp. 75-91
Citations number
61
Categorie Soggetti
Spectroscopy /Instrumentation/Analytical Sciences
Journal title
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
ISSN journal
01697439 → ACNP
Volume
54
Issue
2
Year of publication
2000
Pages
75 - 91
Database
ISI
SICI code
0169-7439(200012)54:2<75:AN3MSR>2.0.ZU;2-2
Abstract
We present a new 3D molecular structure representation based on Richard F.W . Bader's quantum topological atoms in molecules (AIM) theory for use in qu antitative structure-property/activity relationship (QSPR/QSAR) modeling. C entral to this structure representation using quantum topology (StruQT) are critical points located on the electron density distribution of the molecu les. Other gradient fields such as the Laplacian of the electron density di stribution can also be used. The type of critical point of particular inter est is the bond critical point (BCP) which is here characterized by using t he following three parameters: electron density rho, the Laplacian del rho and the ellipticity epsilon. This representation has the advantage that the re is no need to probe a large number of lattice points in 3D space to capt ure the important parts of the 3D electronic structure as is necessary in, e.g. comparative field analysis (CoMFA). We tested the new structure representation by predicting the wavelength of the lowest UV transition for a system of 18 anthocyanidins. Different quant itative structure-property relationship (QSPR) models are constructed using several chemometric/machine learning methods such as standard partial leas t squares regression (PLS), truncated PLS variable selection, genetic algor ithm-based variable selection and genetic programming (GP). These models id entified bonds that either take part in decreasing or increasing the domina nt excitation wavelength. The models also correctly emphasized on the invol vement of the conjugated pi system for predicting the wavelength through fl agging the BCP ellipticity parameters as important for this particular data set. (C) 2000 Elsevier Science B.V. All rights reserved.