|
PARENT SESSION
1F - QSAR Poster Hall 8:30 AM - Tuesday, 29 April 2003 Chair: Schüürmann, G.1, 1 Co-chair: Verhaar, H.J.M.2, Cronin, M.3, 2 3
(TUP/41) Intramolecular Hydrogen Bonding - Electrotopological Indices vs. 3D Geometries for 661 Antibacterials and Non-Antibacterials.
Aptula, Aynur1, Kühne, Ralph1, Ebert, Ralf-Uwe1, Netzeva, Tatiana2, Cronin, Mark2, Schüürmann, Gerrit1, 1 Department of Chemical Ecotoxicology, UFZ Centre for Environmental Research, Leipzig, Germany2 School of Pharmacy and Chemistry, Liverpool John Moores University, Liverpool, England
ABSTRACT- In previous studies, a data set of 661 compounds with many drug-like structures was subjected to modeling classification in terms of antibacterials and non-antibacterials using discriminant analysis as well as binary logistic regression. To this end, both 2D and 3D molecular descriptors were generated, both of which lead to a statistically sound discrimination between both activity classes, with overall classification rates of around 90%. Interestingly, the 2D study that involved electrotopological indices pointed to intramolecular hydrogen bonding as a major discriminating feature besides hydrophobicity, whilst AM1-level 3D geometric and electronic parameters lead to discriminant functions that involved both positively and negatively charged partial surface areas as well as global donor delocalizability. In the present investigation, a detailed analysis of the occurrence of intramolecular hydrogen bonding in both the 249 antibacterials and 412 non-antibacterials is performed, using force-field (Chem-X, SYBYL) as well as AM1 geometries. The results are compared with electrotopological hydrogen bonding descriptors generated for this compound set, and the discussion includes a comparison of further geometrical features such as dihedral angles across distinct chemical functionalities. This work was supported in part by the European Union IMAGETOX Research Training Network (HPRN-CT- 1999-00015).
Key words: 3D geometry, Hydrogen bonding, Force field, AM1
|