Dissertation
A generalized framework for computational antibody design
Doctor of Philosophy (Ph.D.), Drexel University
Jun 2015
DOI:
https://doi.org/10.17918/00008354
Abstract
A novel knowledge-based computational antibody design methodology an application have been developed for the design of antibodies and antibody therapeutics to targets of interest. It has been rigorously benchmarked to reproduce native antibody structures and sequences, and has been used to successfully design an antibody to a target antigen as a proof of concept of de novo design. Our lab's recent clustering of the structures of antibody Complementary Determining Regions (CDRs) in the Protein Data Bank (PDB) enabled the development of the antibody analysis software and database, PyIgClassify, which forms the core of this knowledge-based approach where we use the structural clusters and associated sequence profiles to guide design. The Program uses the Rosetta energy function and new methods for reliable CDR grafting, cluster-based constraints, profile-based sequence design, and flexible-backbone design to computationally test hundreds of thousands of prospective antibody designs in the span of days. Methods were developed for quickly analyzing structural properties of potential designs using the Rosetta FeatureReporter framework, while a Graphical User Interface (GUI), PyIgDesign, was created to aid in antibody design selection. The Rosetta Antibody Design (RAbD) program is meant to sample the diverse sequence, structure, and binding space of an antibody to an antigen. It was developed to enable near-complete customization for a variety of design strategies and projects and can design antibodies for a broad range of applications, from affinity of existing antibody-antigen complex structures to de novo design in which whole new sequences and antibody-antigen contacts are created.
Metrics
31 File views/ downloads
23 Record Views
Details
- Title
- A generalized framework for computational antibody design
- Creators
- Jared Adolf-Bryfogle
- Contributors
- Roland L. Dunbrack Jr. (Advisor) - Drexel University, Drexel University (1970-)
- Awarding Institution
- Drexel University
- Degree Awarded
- Doctor of Philosophy (Ph.D.)
- Publisher
- Drexel University; Philadelphia, Pennsylvania
- Number of pages
- xii, 280 pages
- Resource Type
- Dissertation
- Language
- English
- Academic Unit
- Biochemistry and Molecular Biology; College of Medicine; Drexel University
- Other Identifier
- 991021888841004721