Dokument: Approaches to Automatic Structural Interpretation of Cryo Electron Microscopy Data

Titel:	Approaches to Automatic Structural Interpretation of Cryo Electron Microscopy Data
URL für Lesezeichen:	https://docserv.uni-duesseldorf.de/servlets/DocumentServlet?id=61942
URN (NBN):	urn:nbn:de:hbz:061-20230215-081244-6
Kollektion:	Dissertationen
Sprache:	Englisch
Dokumententyp:	Wissenschaftliche Abschlussarbeiten » Dissertation
Medientyp:	Text
Autor:	Schäfer, Luisa [Autor]
Dateien:	[Dateien anzeigen] Adobe PDF [Details] 180,87 MB in einer Datei [ZIP-Datei erzeugen] Dateien vom 06.02.2023 / geändert 06.02.2023
Beitragende:	Prof. Dr. Schröder, Gunnar [Gutachter] Jun.-Prof. Strodel, Birgit [Gutachter]
Dewey Dezimal-Klassifikation:	500 Naturwissenschaften und Mathematik » 530 Physik
Beschreibung:	Due to their decisive involvement in physiological processes of living organisms, elucidation of the three-dimensional structure of proteins is essential for biomedical research and drug development. Cryogenic Electron Microscopy (cryo-EM) is a state-of-the-art method to experimentally determine a protein structure. It enables the reconstruction of a three-dimensional density map, revealing the structure of a protein with a resolution of up to a few angstrom. However, depending on map resolution, interpreting the density map in terms of an atomic model of the protein is often challenging. Computational tools are needed to facilitate this process. This thesis presents four different methods for automatic structural interpretation of cryo-EM density maps. (1) As contribution to the EMDB Model Metrics Challenge 2019, we have developed a procedure to optimise a given protein structure into a cryo-EM density map, regarding the conformational heterogeneity embodied in the cryo-EM data. (2) In some situations a cryo-EM map allows for straight-forward modelling of the main-chain of the protein, but the assignment of side-chains is often ambiguous. We introduce a method to automatically sample and rank many different side-chain assignments and apply it to a density map representing the structure of IAPP fibrils. (3) We present a routine to flexibly fit fragments from a library of backbone conformations to a given protein trace. The fitted fragments offer more detailed insights into the underlying protein structure than the plain trace. Furthermore, the fragment fitting can easily be integrated in larger frameworks for automatic protein structure modelling. (4) If only the density map and no other structural information is available, the first step is to determine the topology of the protein. We have developed a novel integrative approach to do so. It combines the information provided by the density map with information derived from predicted inter-residue distances. We show, that incorporating distance predictions can correct errors in topology and improve traces that were built based on density information alone.
Lizenz:	Dieses Werk ist lizenziert unter einer Creative Commons Namensnennung 4.0 International Lizenz
Fachbereich / Einrichtung:	Mathematisch- Naturwissenschaftliche Fakultät » WE Physik
Dokument erstellt am:	15.02.2023
Dateien geändert am:	15.02.2023
Promotionsantrag am:	18.10.2022
Datum der Promotion:	10.01.2023