A hierarchical approach to quantitative structure - activity relationship (QSAR) modeling has been used to the estimate the complement-inhibitory potency of 105 benzamidines. This hierarchical approach uses topostructural, topochemical, and geometric parameters in a stepwise fashion to build increasingly more complex models. The results show that topostructural indices alone, specifically ID, predict inhibitory potency reasonably well. The addition of topochemical and geometrical parameters to the set of descriptors provides only marginal improvement in predictive power. However, when taken alone, the geometric parameter 3DW provides a more stable model than the topostructural one.
|Original language||English (US)|
|Number of pages||6|
|Journal||Journal of chemical information and computer sciences|
|State||Published - Jan 1 1999|
|Event||Proceedings of the 1998 First Indo-U.S. Workshop on Mathematical Chemistry - Santiniketan, India|
Duration: Jan 9 1998 → Jan 13 1998