This project intends to apply methods of theoretical chemistry and statistical methodologies to the study of biomolecular structures and interactions. Systems of interest include DNA and DNA metabolizing enzymes, mammalian p450 enzymes, the Zn finger domain of BRCA1, the Gla domain of the coagulation proteins, and the p21 product of the ras oncogene. Methods of approach include quantum calculations as well as classical force field approaches, database searching techniques, and molecular graphics. Due to its high charge concentration and flexibility, DNA has been very difficult to simulate realistically. Recent developments in the treatment of long range electrostatics have allowed us to produce long stable simulations of DNA. We plan to look into structural questions involving damaged DNA. We are modeling mammalian p450's using the homologous bacterial proteins. These enzymes as a class interact with a wide variety of substrates, and yet their specificity can be finely tuned by specific amino acid substitutions. In the absence of a known structure of a mammalian p450 we are focussing on a geometric model of the binding pocket. The model is based on the known structure of a bacterial p450. The p21 protein is a molecular """"""""switch"""""""" central to a variety of growth and developmental signaling pathways in human tissues. The """"""""switch"""""""" operates by means of a hydrolysis of the GTP cofactor. Our interest is in the intrinsic rate of hydrolysis in normal and mutant p21.
Showing the most recent 10 out of 52 publications
