Back to search

Cornell University

Model-Based Statistical Estimation Algorithms For Functional Structural Virology

Abstract

dc:description.abstract

A technique that is used widely in structural biology utilizing transmission electron microscopy (TEM) is single-particle cryo-EM, in which one projection image of a plunge-frozen specimen of multiple identical copies of a macromolecule, oriented randomly on a thin layer of vitreous ice, is acquired [88, 33]. The key assumption of each particle being identical allows one to combine projections of particles with the same orientation to increase SNR and to combine averaged projections of the structure at different views to create a high resolution 3D reconstruction of a given structure [33, 34]. This technique is especially attractive in studying viruses, which are relatively rigid and large macromolecular complexes that are made up of identical subunits arranged in a regular pattern [104, 43]. However, several properties of this technique limit its applicability in structural virology. These limitations are summarized below along with new experimental and computational methods that we help develop to address these issues. 1) Single-particle cryo-EM is an in vitro technique. To gain in vivo structural information, we utilize whole-cell cryo-electron tomography (CET), in which a tilt series of projection images of a single cell infected with virus particles is acquired and a 3-D tomogram is reconstructed from these projections. Together with pattern recognition techniques, we are able to study how a virus particle interacts with cellular components of its host cell during its lifecycle in a reliable way. 2) The requirement of identical copies of a structure prevents the use of cryo-EM from studying viruses that are pleomorphic. Because 3D information of each particle is obtained in CET data, CET of purified particles is a powerful technique to analyze components of virus particles that are similar to each other in a collective way. One excellent example is the glycoprotein spikes of enveloped viruses. We develop a computational method that can be used to analyze subregions of CET data cubes in a reliable and efficient way. 3) Single-particle cryo-EM is a static imaging method that only provide snap-shots of structural states of a virus. By developing a physical model based on image statistics of cryo-EM data, we show how one can accurately predict the conformational dynamics of a virus structure based on its cryo-EM reconstruction and obtain information concerning the mechanism of how a virus functions. Hopefully, this set of tools will enable biologists to study viruses in a more comprehensive way.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Biomedical Engineering
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Biomedical Engineering
Grantor
Cornell University
Year dc:date.issued
2012

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Wang, Kang
Chair dc:contributor.chair
  • Doerschuk, Peter
Committee members dc:contributor.committeemember
  • Reeves, Anthony P
  • Parker, John Stuart Leslie
  • Johnson, John E

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Dc Identifier Bibid
7745257

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Wang, Kang. Model-Based Statistical Estimation Algorithms For Functional Structural Virology. Doctor of Philosophy thesis, Cornell University, 2012. https://hdl.handle.net/1813/29383