| Accounting for solvation in thermodynamic studies is one issue requiring further work so that computational models may be more advantageously applied to studies of systems in solution, especially studies of large systems such as aqueous biosystems. In the area of applied molecular biology, this capability is very significant, since computational thermodynamic studies can supply valuable information to assist in the design of molecules with specific desired binding or conformational properties, such as inhibitors. This dissertation addresses the issue by suggesting a new approach for the calculation of solvation free energies using a modified reference interaction site model (RISM) integral equation approach in which molecular simulations are used to provide the solvent structure around a solute at infinite dilution in the form of radial distribution functions. The intent is to compensate for insufficiencies arising in the standard RISM approach, and in this way to establish an alternative to the very time consuming free energy simulations, which are usually depended upon for high accuracy and have seen some recent applications, although falling somewhat short as far as practicality. Chapter 2 of this dissertation investigates how it is possible to implement such a scheme, with consideration given to the inherent errors associated with such an approach. In Chapter 3, the resulting coupled RISM simulation methodology is first tested by its application to determine absolute solvation free energies of some small molecules.; Applications of relative solvation free energy determination by the coupled RISM/simulation methodology to the conformational analysis of the alanine dipeptide, and to the tautomeric equilibria of the DNA base cytosine and one of its analogues, are then described in Chapters 4 and 5, respectively. Chapter 6 is concerned with the determination of a potential of mean force profile for a simple atom transfer reaction in aqueous solution. By means of these applications, the accuracy of the developed methodology is tested by comparison with standard free energy simulations and experimental data. |
