Bridging the gap between computational chemistry and macromolecular crystallography
(2019)- Abstract
- Knowledge of the atomic structure of biomolecules, such as proteins, is paramount to understanding their function and interactions in the human body. For example, knowledge of the atomic structure of a target protein is crucial for developing drugs that bind strongly to it and thus help cure diverse diseases.
Macromolecular crystallography is the forefront method for determining the atomic structure of proteins, especially through X-ray diffraction experiments. However, the data obtained from these experiments are not the atomic structure but need to be processed and interpreted before arriving at the individual positions of atoms in a protein. This intepretation is done through computational techniques that share some of the... (More) - Knowledge of the atomic structure of biomolecules, such as proteins, is paramount to understanding their function and interactions in the human body. For example, knowledge of the atomic structure of a target protein is crucial for developing drugs that bind strongly to it and thus help cure diverse diseases.
Macromolecular crystallography is the forefront method for determining the atomic structure of proteins, especially through X-ray diffraction experiments. However, the data obtained from these experiments are not the atomic structure but need to be processed and interpreted before arriving at the individual positions of atoms in a protein. This intepretation is done through computational techniques that share some of the algorithms and problems with computational chemistry.
In this thesis, we use several methods that combine computational chemistry and macromolecular crystallography for the study of multiple important proteins. Crystallographic refinement combined with quantum mechanical calculations (quantum refinement) is used to improve the X-ray structures of three metalloenzymes. Furthermore, a quantum refinement procedure for neutron structures is developed and applied to two important enzymes. We also investigate how to use and improve the existing information on dynamics from crystallography experiments. To this end, we test whether conformational entropy can be calculated directly from B-factors. Additionally, ensemble refinement is used to explore ligand dynamics in the binding site of galectin-3 and reveals hidden conformations that were not apparent in traditional crystallographic refinement methods. Finally, we study the modeling of water molecules in protein X-ray and neutron crystal structures. We show that molecular dynamics simulations can reproduce crystal water molecules, if protein movements are correctly taken into account. Moreover, we have developed a method to automatically improve the orientation of water molecules in neutron structures.
(Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/dde56e58-5feb-4e34-8bce-843437488651
- author
- Caldararu, Octav LU
- supervisor
-
- Ulf Ryde LU
- Esko Oksanen LU
- opponent
-
- Dr. Murshudov, Garib, MRC Laboratory of Molecular Biology, Cambridge, UK
- organization
- publishing date
- 2019
- type
- Thesis
- publication status
- published
- subject
- keywords
- Protein structure, X-ray crystallography, Neutron crystallography, Quantum mechanics, QM/MM, Quantum refinement, Ensemble refinement, Water structure
- pages
- 238 pages
- publisher
- Lund University
- defense location
- Kemicentrum, Lecture Hall F, Naturvetarvägen 14, Lund
- defense date
- 2019-12-06 09:00:00
- ISBN
- 978-91-7422-702-4
- 978-91-7422-703-1
- language
- English
- LU publication?
- yes
- id
- dde56e58-5feb-4e34-8bce-843437488651
- date added to LUP
- 2019-11-06 11:14:48
- date last changed
- 2022-06-15 09:33:15
@phdthesis{dde56e58-5feb-4e34-8bce-843437488651,
abstract = {{Knowledge of the atomic structure of biomolecules, such as proteins, is paramount to understanding their function and interactions in the human body. For example, knowledge of the atomic structure of a target protein is crucial for developing drugs that bind strongly to it and thus help cure diverse diseases.<br/>Macromolecular crystallography is the forefront method for determining the atomic structure of proteins, especially through X-ray diffraction experiments. However, the data obtained from these experiments are not the atomic structure but need to be processed and interpreted before arriving at the individual positions of atoms in a protein. This intepretation is done through computational techniques that share some of the algorithms and problems with computational chemistry.<br/>In this thesis, we use several methods that combine computational chemistry and macromolecular crystallography for the study of multiple important proteins. Crystallographic refinement combined with quantum mechanical calculations (quantum refinement) is used to improve the X-ray structures of three metalloenzymes. Furthermore, a quantum refinement procedure for neutron structures is developed and applied to two important enzymes. We also investigate how to use and improve the existing information on dynamics from crystallography experiments. To this end, we test whether conformational entropy can be calculated directly from B-factors. Additionally, ensemble refinement is used to explore ligand dynamics in the binding site of galectin-3 and reveals hidden conformations that were not apparent in traditional crystallographic refinement methods. Finally, we study the modeling of water molecules in protein X-ray and neutron crystal structures. We show that molecular dynamics simulations can reproduce crystal water molecules, if protein movements are correctly taken into account. Moreover, we have developed a method to automatically improve the orientation of water molecules in neutron structures.<br/>}},
author = {{Caldararu, Octav}},
isbn = {{978-91-7422-702-4}},
keywords = {{Protein structure, X-ray crystallography, Neutron crystallography, Quantum mechanics, QM/MM, Quantum refinement, Ensemble refinement, Water structure}},
language = {{eng}},
publisher = {{Lund University}},
school = {{Lund University}},
title = {{Bridging the gap between computational chemistry and macromolecular crystallography}},
url = {{https://lup.lub.lu.se/search/files/71557310/Only_kappa_full_resolution.pdf}},
year = {{2019}},
}