The long-term objective of this proposal is to produce and apply methods and tools for chemically accurate and computationally efficient molecular simulations. The targets of the simulations are organic, physical, and biological systems (including proteins and peptides). The results of this work will be available to the general scientific community. The emphasis of the methodological part of this proposal will be on accurate and computationally inexpensive inclusion of electrostatic polarization. The following specific goals will be set with the above long-term objective in mind: A software suit will be developed for molecular modeling with a fast polarization technique employed. The technique has been previously proposed and tested by the author. Its application resulted in obtaining accurate results with a ca. an order of magnitude increase in computational speed. Force field parameters to be used with the model will be produced for a variety of biologically significant organic and protein molecules. The author has an extensive amount of experience with such tasks, including development of polarizable and non-polarizable protein force fields. In the proposed research, the main source of the target fitting data will be high-level quantum mechanical calculations. Two major directions of applications are planned. First, accurate predictions of protein pKa values will be conducted. Accurate assessment of pKa values of protein residues is important in predicting protein stability, solubility, and biological activity. The first two major targets in this project will be the turkey ovomuvoid third domain and ribonuclease Sa. Second, the fast polarization technique, combined with the Monte Carlo method, will be applied to studies of effective farnesyl transferase inhibitors. While farnesyl transferase has been acknowledged as a very valid target for anti-cancer inhibition, synthetic reproduction of some of the most potent natural inhibitors (such as the CP-type) is often very challenging due to the structure of the inhibitors. The author hopes that the accuracy of the polarizable force field, combined with the speed of the employed implementation and the true thermodynamic sampling in the Monte Carlo form will allow to assess binding energy tendencies adequately enough to suggest easier-to synthesize potent inhibitors - analogues of the natural products and peptidomimetics.
Showing the most recent 10 out of 13 publications
