Efficient, Polarizable Charge Model for Molecular Simulation

Computers serve chemists from various disciplines as an indispensable tool to understand, but also to perform experiments. The Nobel-Committee recently underlined the importance of computational research by awarding the 2013 Nobel Prize in Chemistry to M. Karplus, M. Levitt and A. Warshel for their contributions to the field of computer-assisted simulations on the interaction of chemical systems. A vast variety of different tools are available to theoretical chemists today. In an ideal world, the most accurate method would always be used. Unfortunately, for all relevant questions this is computationally far too demanding. Therefore, theoreticians have to develop and use more approximated methods, balancing accuracy and computational effort. The method of choice in the pharmaceutical industry are so-called force fields. They are used in the early stages of nearly every drug design project to select promising candidates for the use as future medication. The force field method allows to simulate entire proteins, which very often are the targets of potential drugs, with adequate accuracy. Although this technique has already been developed for over 35 years, some errors introduced by the used approximations still persist. For this reason, efforts have been made in recent years to develop force fields based on a physically sounder description of the investigated systems. One of the main sources of error in state-of-the-art force fields is the simplification that molecules do not respond to their environment and vice versa, whereas in reality they do. The project focuses on the improvement of a next generation force field (a so-called polarizable force field), where molecules are able to respond to their environment, causing a significant error reduction. The goal of the presented project is the development of a very efficient and easy-to-handle polarizable force field. To achieve this, the project will follow an innovative, easy methodology in comparisons to previous development projects: Instead of designing a completely independent force field, we will build on a very popular state-of-the-art force field and refurbish it with additional functions. This will result in a robust, fast, and also efficient polarizable force field. Thereby, a method with increased accuracy but only minor additional computational costs will be developed. As force fields are heavily used in the field of pharmaceutical chemistry, we expect that, amongst others, especially the field of modern drug development will benefit from this project.

Computers serve chemists from various disciplines as an indispensable tool to understand, but also to perform experiments. The Nobel-Committee recently underlined the importance of computational research by awarding the 2013 Nobel Prize in Chemistry to M. Karplus, M. Levitt and A. Warshel for their contributions to the field of computer-assisted simulations on the interaction of chemical systems. A vast variety of different tools are available to theoretical chemists today. In an ideal world, the most accurate method would always be used. Unfortunately, for all relevant questions this is computationally far too demanding. Therefore, theoreticians have to develop and use more approximated methods, balancing accuracy and computational effort. The method of choice in the pharmaceutical industry are so-called force fields. They are used in the early stages of nearly every drug design project to select promising candidates for the use as future medication. The force field method allows to simulate entire proteins, which very often are the targets of potential drugs, with adequate accuracy. Although this technique has already been developed for over 35 years, some errors introduced by the used approximations still persist. For this reason, efforts have been made in recent years to develop force fields based on a physically sounder description of the investigated systems. One of the main sources of error in state-of-the-art force fields is the erroneous description of the molecules electrostatics. In this project we developed a new method to describe the electrostatic of a molecule in a given environment more accurately. The method has its foundation in more accurate quantum chemical calculation and therefore does not rely, in contrast to earlier methods, on fortuitous error compensation. In a second step we developed parameters for a next generation force field (a so-called polarizable force field), where molecules are able to respond to their environment, causing a significant error reduction. In the context of this project we developed a parameterization schema for a very efficient and easy-to-handle polarizable force field. This was achieved by following an innovative, easy methodology in comparisons to previous development projects: Instead of designing a completely independent force field, we build on a very popular state-of-the-art force field and refurbished it with additional functions.