Thesis supervisor: Beáta Albert
co-supervisor: Szilárd Fejér
Location of studies (in Hungarian): Csíkszereda Abbreviation of location of studies: Csík
Description of the research topic:
Currently the most accepted methods for the study of protein structures is the X ray crystallog-raphy, which in turn can be difficult and problematic, since the crystallisation does not occur under physiological conditions and the conformations of the proteins can not be relevant.
The structure of proteins can be examined under native conditions only frozen with liquid nitrogen, on TEM, a special version of the cryo-electron microscope. The result in this case is a low resolu-tion electron density map from which a computer can determine a structure.
For structure determination we will use the ADMA (Adjustable Density Matrix Assembler) method. The method is designed for ab initio quantum chemical calculations for macromolecules, especially proteins structure, which are assembled from its smaller parts.
Looking at the local interactions of small, fuzzy fragments, on the base of Mulliken-Mezey scheme can be formed density matrixes, from which can be calculated properties as the electron density, electrostatic potential, and total energy.
Based on the electron density can be formed mathematical maps with 21X41 elements, which can be improved trough mathematical optimization.
At present all points of the map numerical codes, Betti numbers are made up, so we want to reduce that to order just one point to every Betti number. This reduces the required computing capacity and computing time. The success of mathematical optimization will be verified with a protein with unknown crystal structure.
Login
