Determination of thermodynamic and kinetic properties of proteins from atomistic computer simulations
Authors
Jiménez Cruz, Camilo Andrés
ORCID
Loading...
Other Contributors
García, Angel E.
Giedt, Joel
Meunier, Vincent
Garde, Shekhar
Makhatadze, George I.
Giedt, Joel
Meunier, Vincent
Garde, Shekhar
Makhatadze, George I.
Issue Date
2013-08
Keywords
Physics
Degree
PhD
Terms of Use
This electronic version is a licensed copy owned by Rensselaer Polytechnic Institute, Troy, NY. Copyright of original work retained by author.
Full Citation
Abstract
Proteins undergo fluctuations spanning a wide range of timescales. Atom oscillations and bond vibration happen in femtoseconds, while the timescales of non-local contacts and domain rearrangements range from miliseconds to seconds to occur. As a result of this hierarchy of processes, protein folding is a rare event, and thus, the dwelling times in local metastable states are much longer than the transition times between such states. Since the function of proteins is determined by their atomic motions, a detailed understanding of protein folding can result in significant advances in biology and medicine. Molecular Dynamics (MD) simulations of biomolecular systems, based on physical models of the interactions, provide atomic-level description of these processes. While MD directly samples temporal and spatial resolutions not available to experiments, it presents new challenges in the generation, validation and analysis of datasets. In this thesis, we focus on the characterization of folding/unfolding equilibrium and kinetic properties of a beta hairpin and globular mini protein. We use extensive all-atom, explicit solvent, Replica Exchange Molecular Dynamics (REMD) simulations, to generate the equilibrium ensemble of configurations of the systems studied and selected mutants. Our results high- light the importance of non-native interactions in the relative stability of the folded state. Kinetic networks for the systems were generated by constructing Markov State Models (MSMs), using thousands of independent MD simulations. We provide new approaches to solve some of the challenges of building MSMs. Specifically, we develop novel strategies for the discretization of the configurational space, and apply them to the model systems.
Description
August 2013
School of Science
School of Science
Department
Dept. of Physics, Applied Physics, and Astronomy
Publisher
Rensselaer Polytechnic Institute, Troy, NY
Relationships
Rensselaer Theses and Dissertations Online Collection
Access
Restricted to current Rensselaer faculty, staff and students. Access inquiries may be directed to the Rensselaer Libraries.