| The META program is an expert system that employs a model consisting of a set of rules to predict the transformation of chemicals in different environments. META searches the structure of a given molecule for specific structural sequences and modifies them, thus creating a transformed chemical. A model to predict the degradation pathways of organic compounds under UV light was constructed. This photodegradation model was validated and optimized by a META_TREE genetic algorithm using data reported in the literature. After completion, the photodegradation dictionary was applied to the study of a group of industrial compounds for which degradation pathways were published. A very good fit between experimental and predicted data (hit/miss ratio around 92%) was achieved. Thus, the UV degradation model expands the applicability of META beyond the bounds of metabolism for which it was originally devised.; Parkinson's disease is caused by a dopamine deficiency state in the forebrain area. Dopamine receptor agonists, MAO-B inhibitors, and N-methyl-D-aspartate receptor antagonists are all known to have antiparkinson effect. We have created computer models for these three groups of compounds in an effort to identify the relationship between chemical structure and observed beneficial activities using the MCASE methodology. In an effort to create a more potent drug, the structural features found to be responsible for each of the three types of active compounds have been combined in the hope that the resultant drug will act by multiple activity mechanisms. The proposed antiparkinson compounds bear pharmacophores of MAO-B inhibitors, DA agonists and NMDA antagonists, which could synergistically enhance their antiparkinson effect. Two such compounds have been synthesized and are now being tested for activity. |
