Quantum Mechanical Methods for Prediction of Thermodynamic Properties and QSAR Modeling
Nov 11, Fri, 2005
3:30pm - 4:30pm, Alcove F, Level 3 of the KU Union
The role of computational methods in assisting experimentalists has expanded dramatically in recent years due to more powerful computing resources. In this seminar, I will discuss the use of semiempirical and ab initio calculations in a collaborative effort with experimental research to study the polymerization reaction of a potential dental restorative. Modeling of these reactions is now possible on desktop machines in a reasonable amount of time which was not possible just a few years ago. Comparing thermodynamic trends for different reactions, a new mechanism was predicted to participate which had not previously been reported in the literature. This theoretical work enabled assignment of a large number of unexplained MALDI-TOF peaks and showed unambiguously that copolymerization was involved. In another example, I will describe a Quantitative Structure Activity Relationship (QSAR) model based on quantum mechanical descriptors which correlates molecular structure with the property of skin sensitization. While the model does not replace experimental work, it is now being used for preliminary prediction of untested compounds and has reduced the number of animal tests necessary by eliminating range finding experiments.