Finding meta stable states in proteins Identification of meta stable states in VPAL-peptide's dynamics
Examensarbete för masterexamen
Applied physics (MPAPP), MSc
Proteins are very large biological molecules, responsible for many functions in living cells and organisms. Ever since they were recognized as a distinct class in 1789 by Antoine Fourcroy and others, proteins were, and still are due to their vital importance and high complexity, subject of studies of scientists from various fields of science all over the world. It was discovered, that each protein consists of one or more chains of amino acids compactly folded in space. This fully folded structure protein attain from the initial unfolded structure through intermediate ones. Moreover, scientists believe that protein properties and functions strongly depend on these structures. Fully folded structure, called native state, is the most preferable, due to the lowest overall free energy, so once it is reached, it holds for a long time, which increases chances for success in the experimental research of it (using, for example, X-ray crystallography or Nuclear Magnetic Resonance techniques). Unfortunately, this is not the case for intermediate structures, due to the either too fast (hard to record and notice) or too slow (hard to separate) folding process. That is when united physical, mathematical, and computer research dominates. This master thesis work presents a mathematical approach for studying intermediate protein structures. These intermediate structures, usually called meta stable states, are usable in multiple fields of research. For example, biochemistry and pharmaceutics are using them to optimize the shape of the drug molecules for achieving the best possible binding properties with respect to the target molecules; In biophysics it helps to find the folding pathways in the potential energy landscape; and in computational biology to reduce the amount of data needed to be stored obtained from the molecular dynamics simulations and also so that the problem can be partitioned into smaller pieces and be run in parallel on multiple computers. The goal of this work was to establish, if it is possible to apply spectral method for finding meta stable states of proteins. This was achieved through 3 major steps. First step was to obtain data for analysis, namely to perform molecular dynamics simulations resulting in the protein dynamics trajectory. Second step was to prepare this data for the analysis, namely to divide conformational trajectory into microstates and construct transition probability matrix. And, finally, the third step was to analyze data, by applying spectral method, which resulted in meta stable states. Tests were performed on the small peptide (peptide is the name for protein with the length of the chain less than 100 amino acids) consisting of only 4 amino acids named Valine-Proline-Alanine-Liucine (VPAL) . This peptide, despite of its small size, attains properties of a protein, i.e. (folds through / has) meta stable states, however, because of its size, it makes it possible to get results comparatively quickly. Results obtained for VPAL-peptide were confirmed by those known in literature, which let us to conclude that it is possible to apply spectral method for finding meta stable states.
Fysik , Physical Sciences