The role of hydration and flexibility in computational antibody design
The role of hydration and flexibility in computational antibody design
This thesis studies how molecular simulation can identify aspects of antibody-antigen interfaces that are important for binding and affinity prediction. Using the enhanced sampling method of Replica Exchange with Solute Scaling (REST2), limitations of current computational antibody design methods are explored and then improved upon in an in silico affinity maturation attempt. The first results chapter justifies the creation of a new, long timescale dataset of extensive antibody simulations, and the effect of REST2 on sampling of Complementarity Determining Region (CDR) loops which dictate the majority of antibody binding. The results show that REST2 significantly increases CDR sampling, and that CDRs have different dynamics in the presence and absence of their antigen. The second results chapter inspects antibody-antigen interface interactions from the REST2 dataset. The analysis confirms that antibodies regularly interact with waters at the interface, a feature that is only captured by high resolution crystal structures. More interactions with the antigen and lower antigen-bound CDR flexibility correlated with better antibody binding, suggesting that solvent mediation and dynamics should be considered when predicting antibody affinity. The last results chapter integrates REST2 simulation into UCB’s in silico affinity maturation technology. An adapted REST2 approach for mutating an antibody against a human target into a cynomolgus monkey surrogate antibody is presented, with examples of how simulation can guide structure-based rational design. The computational workflow reproduced the best binder from the display-based affinity maturation campaign and designed a novel variant with affinity on par with the best experimentally discovered binder. This shows that REST2 simulation, when combined with existing in silico affinity maturation methods, can add value to antibody design.
University of Southampton
Wong, Mabel, Tung Yuet
97923566-eaa6-41af-b1af-83bb57d61753
Wong, Mabel, Tung Yuet
97923566-eaa6-41af-b1af-83bb57d61753
Essex, Jonathan
1f409cfe-6ba4-42e2-a0ab-a931826314b5
Wong, Mabel, Tung Yuet
(2022)
The role of hydration and flexibility in computational antibody design.
University of Southampton, Doctoral Thesis, 162pp.
Record type:
Thesis
(Doctoral)
Abstract
This thesis studies how molecular simulation can identify aspects of antibody-antigen interfaces that are important for binding and affinity prediction. Using the enhanced sampling method of Replica Exchange with Solute Scaling (REST2), limitations of current computational antibody design methods are explored and then improved upon in an in silico affinity maturation attempt. The first results chapter justifies the creation of a new, long timescale dataset of extensive antibody simulations, and the effect of REST2 on sampling of Complementarity Determining Region (CDR) loops which dictate the majority of antibody binding. The results show that REST2 significantly increases CDR sampling, and that CDRs have different dynamics in the presence and absence of their antigen. The second results chapter inspects antibody-antigen interface interactions from the REST2 dataset. The analysis confirms that antibodies regularly interact with waters at the interface, a feature that is only captured by high resolution crystal structures. More interactions with the antigen and lower antigen-bound CDR flexibility correlated with better antibody binding, suggesting that solvent mediation and dynamics should be considered when predicting antibody affinity. The last results chapter integrates REST2 simulation into UCB’s in silico affinity maturation technology. An adapted REST2 approach for mutating an antibody against a human target into a cynomolgus monkey surrogate antibody is presented, with examples of how simulation can guide structure-based rational design. The computational workflow reproduced the best binder from the display-based affinity maturation campaign and designed a novel variant with affinity on par with the best experimentally discovered binder. This shows that REST2 simulation, when combined with existing in silico affinity maturation methods, can add value to antibody design.
Text
Mabel Wong_PhD ChemistryThesis_EssexGroup_2022.04.04
- Version of Record
Text
Permission to deposit_thesis_form
Restricted to Repository staff only
More information
Submitted date: 4 April 2022
Identifiers
Local EPrints ID: 467507
URI: http://eprints.soton.ac.uk/id/eprint/467507
PURE UUID: d14550f1-8c5d-4575-9c1a-5bf051e1baa1
Catalogue record
Date deposited: 12 Jul 2022 16:32
Last modified: 17 Mar 2024 07:22
Export record
Contributors
Author:
Mabel, Tung Yuet Wong
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics