Deep generative modeling for structure-guided protein design
Abstract/Contents
- Abstract
- In recent years, deep generative modeling has garnered attention as a powerful way to approximate the probability distributions of high-dimensional data, and synthesize an infinite amount of novel data from finite datasets. In the field of protein design, where structural data is low in abundance and expensive to obtain, generative modeling has the potential to "fill" many gaps in known conformational space, increasing the scope of designable proteins, and allowing us to automate design in a data-guided way. This thesis describes my work in developing and training generative models for computational protein design, and describes the reformulation of a general protein design task as a conditional generation problem, solved by latent variable optimization within a deep generative model. I begin with a description of early explorations of protein representation and computer vision methods, leading to the development of a new 3D generative model for proteins, and culminating with the invention of an algorithm that uses generative design to create binding proteins against arbitrary user-specified epitopes. We name this algorithm "Sculptor".
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource. |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2022; ©2022 |
Publication date | 2022; 2022 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Eguchi, Raphael Ryuichi |
---|---|
Degree supervisor | Huang, Possu |
Thesis advisor | Huang, Possu |
Thesis advisor | Das, Rhiju |
Thesis advisor | Harbury, Pehr |
Thesis advisor | Kim, Peter, 1958- |
Degree committee member | Das, Rhiju |
Degree committee member | Harbury, Pehr |
Degree committee member | Kim, Peter, 1958- |
Associated with | Stanford University, Department of Biochemistry |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Raphael R. Eguchi. |
---|---|
Note | Submitted to the Department of Biochemistry. |
Thesis | Thesis Ph.D. Stanford University 2022. |
Location | https://purl.stanford.edu/cm738cq4005 |
Access conditions
- Copyright
- © 2022 by Raphael Ryuichi Eguchi
Also listed in
Loading usage metrics...