The general objective of this work is to generate information for the development of environmental policies by means of studying the structure-activity relationships of available toxicological data. This theoretical approach offers an alternative to current processes followed by the Environmental Protection Agency and other regulatory agencies, which rely on information generated from a battery of standardized toxicity assays employing animal models and other short term in vitro assays. The extensive and indiscriminate toxicological testing of chemicals is undesirable due to a combination of economical and practical reasons, including the rising social pressure to decrease animal research.

Our approach involves the study of the quantitative structure-activity relationships as a way to summarize and organize information, provide insights into the molecular mechanisms of biological activity, and offer statistical models to predict the toxicity of new or untested compounds. The basic assumption is that the biological activity elicited by a chemical depends on its structure. The specific aims are:

Specific Aim 1. To develop an easily accessible databank of information on mutagenic nitroaromatic compounds. This databank includes Ames mutagenicities, optimized structure, MOLCAD surfaces in VRML format, LUMO and HOMO energies, and octanol/water partition coefficients. This information is hosted by the Biomedical Visualization Center at UAMS (BVC) and will be Internet accessible. Presently the databank is about 500 Mb, and increasing rapidly in size with the addition of the VRML surfaces.

Specific Aim 2. Expand existing QSAR models by compiling and analyzing toxicological information in other toxicological tests and chemical types. Comparison of QSARs should help to establish the overlapping information provided by each test. Analysis of outliers could indicate a different mechanism of toxicity and provide grounds for new experiments that will be performed at the National Center for Toxicological Research. While computational studies will be conducted at the Biomedical Visualization Center at UAMS (BVC).