Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images
Proteins and other biomolecules form dynamic macromolecular machines that carry out essential biological processes responsible for life. However, studying the mechanisms of these biomolecular complexes at relevant atomic-scale resolutions is an extraordinarily challenging task in structural biology....
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2022
|
| Online Access: | https://hdl.handle.net/1721.1/144512 |
| _version_ | 1826191846005538816 |
|---|---|
| author | Zhong, Ellen D. |
| author2 | Berger, Bonnie |
| author_facet | Berger, Bonnie Zhong, Ellen D. |
| author_sort | Zhong, Ellen D. |
| collection | MIT |
| description | Proteins and other biomolecules form dynamic macromolecular machines that carry out essential biological processes responsible for life. However, studying the mechanisms of these biomolecular complexes at relevant atomic-scale resolutions is an extraordinarily challenging task in structural biology. This thesis presents new algorithms that address the computational bottlenecks at the frontier of structure determination of dynamic biomolecular complexes via cryo-electron microscopy (cryo-EM).
In single particle cryo-EM, the central problem is to reconstruct the 3D structure of a target biomolecular complex from a set of noisy and randomly oriented 2D projection images, a challenging inverse problem especially when instances of the imaged biomolecular complex exhibit structural heterogeneity.
The main contribution of this thesis is a machine learning system, cryoDRGN, for reconstructing continuous distributions of biomolecular structures from cryo-EM images. Underpinning the cryoDRGN method is a deep generative model parameterized by a new neural representation of cryo-EM volumes and a learning algorithm to optimize this representation from unlabeled 2D cryo-EM images. Released as an open source software tool, cryoDRGN has been applied on real datasets to uncover heterogeneity in high resolution datasets, discover new conformations of large macromolecular machines and visualize continuous trajectories of their motion. This thesis also describes an extension, cryoDRGN2, for learning this model from unposed images, i.e. ab initio reconstruction. Finally, this thesis presents emerging directions in analyzing the learned manifold of cryo-EM structures and in incorporating atomic model priors into cryo-EM reconstruction. |
| first_indexed | 2024-09-23T09:02:11Z |
| format | Thesis |
| id | mit-1721.1/144512 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T09:02:11Z |
| publishDate | 2022 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1445122022-08-30T03:59:23Z Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images Zhong, Ellen D. Berger, Bonnie Davis, Joseph H. Massachusetts Institute of Technology. Computational and Systems Biology Program Proteins and other biomolecules form dynamic macromolecular machines that carry out essential biological processes responsible for life. However, studying the mechanisms of these biomolecular complexes at relevant atomic-scale resolutions is an extraordinarily challenging task in structural biology. This thesis presents new algorithms that address the computational bottlenecks at the frontier of structure determination of dynamic biomolecular complexes via cryo-electron microscopy (cryo-EM). In single particle cryo-EM, the central problem is to reconstruct the 3D structure of a target biomolecular complex from a set of noisy and randomly oriented 2D projection images, a challenging inverse problem especially when instances of the imaged biomolecular complex exhibit structural heterogeneity. The main contribution of this thesis is a machine learning system, cryoDRGN, for reconstructing continuous distributions of biomolecular structures from cryo-EM images. Underpinning the cryoDRGN method is a deep generative model parameterized by a new neural representation of cryo-EM volumes and a learning algorithm to optimize this representation from unlabeled 2D cryo-EM images. Released as an open source software tool, cryoDRGN has been applied on real datasets to uncover heterogeneity in high resolution datasets, discover new conformations of large macromolecular machines and visualize continuous trajectories of their motion. This thesis also describes an extension, cryoDRGN2, for learning this model from unposed images, i.e. ab initio reconstruction. Finally, this thesis presents emerging directions in analyzing the learned manifold of cryo-EM structures and in incorporating atomic model priors into cryo-EM reconstruction. Ph.D. 2022-08-29T15:52:31Z 2022-08-29T15:52:31Z 2022-05 2022-06-14T23:25:03.996Z Thesis https://hdl.handle.net/1721.1/144512 In Copyright - Educational Use Permitted Copyright MIT http://rightsstatements.org/page/InC-EDU/1.0/ application/pdf Massachusetts Institute of Technology |
| spellingShingle | Zhong, Ellen D. Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images |
| title | Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images |
| title_full | Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images |
| title_fullStr | Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images |
| title_full_unstemmed | Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images |
| title_short | Machine Learning for Reconstructing Dynamic Protein Structures from Cryo-EM Images |
| title_sort | machine learning for reconstructing dynamic protein structures from cryo em images |
| url | https://hdl.handle.net/1721.1/144512 |
| work_keys_str_mv | AT zhongellend machinelearningforreconstructingdynamicproteinstructuresfromcryoemimages |