2024
DOI: 10.1101/2024.03.14.585103
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Atomically accurate de novo design of single-domain antibodies

Nathaniel R. Bennett,
Joseph L. Watson,
Robert J. Ragotte
et al.

Abstract: Despite the central role that antibodies play in modern medicine, there is currently no way to rationally design novel antibodies to bind a specific epitope on a target. Instead, antibody discovery currently involves time-consuming immunization of an animal or library screening approaches. Here we demonstrate that a fine-tuned RFdiffusion network is capable of designing de novo antibody variable heavy chains (VHH's) that bind user-specified epitopes. We experimentally confirm binders to four disease-relevant e… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
5
0

Year Published

2024
2024
2025
2025

Publication Types

Select...
4
3
1

Relationship

0
8

Authors

Journals

citations
Cited by 36 publications
(5 citation statements)
references
References 37 publications
0
5
0
Order By: Relevance
“…Also, the number of patients eligible for existing TCR or TMA-based therapies could be expanded by rapidly testing for interactions with allotypes belonging to the same T-CREG as the original HLA which restricts a known receptor 22 . In the future, de novo receptor design [49][50][51][52] could allow for binding modes that are directed towards conserved regions of the MHC-I surface, enabling cross-HLA targeting of public epitopes across different T-CREGs. Finally, our T-CREG analysis requires a minimal amount of information, simply the set of HLAs anticipated to bind the peptide, and is publicly available online through a code repository.…”
Section: Discussionmentioning
confidence: 99%
“…Also, the number of patients eligible for existing TCR or TMA-based therapies could be expanded by rapidly testing for interactions with allotypes belonging to the same T-CREG as the original HLA which restricts a known receptor 22 . In the future, de novo receptor design [49][50][51][52] could allow for binding modes that are directed towards conserved regions of the MHC-I surface, enabling cross-HLA targeting of public epitopes across different T-CREGs. Finally, our T-CREG analysis requires a minimal amount of information, simply the set of HLAs anticipated to bind the peptide, and is publicly available online through a code repository.…”
Section: Discussionmentioning
confidence: 99%
“…Despite these caveats, our study provides the first comprehensive measurements of how HA mutations affect key molecular phenotypes that contribute to the pandemic risk posed by H5 influenza. Combining these measurements with rapid sequencing of ongoing H5 influenza outbreaks can improve monitoring of ongoing viral evolution, as well as inform the choice of candidate vaccine strains, the targeting of designed therapeutics 20,21 , and the prioritization of viral strains for additional experimental study 64,65 .…”
Section: Discussionmentioning
confidence: 99%
“…From a basic standpoint, they illuminate sequence-function relationships-for instance by comprehensively defining sites in HA's stalk and head that contribute to its stability, and mutations that affect HA's preference for α2-6-versus α2-3-linked sialic acids. From an applied standpoint, the measurements define which sites are constrained by the virus's need to maintain HA's essential cell entry function, thereby identifying targets for new approaches that design antibodies (17) and other therapeutics (16) to bind specific epitopes.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…These limitations necessitate a paradigm shift towards innovative approaches to address the evolving demands of antibody development. Indeed, recent efforts have led to deep learning networks and diffusion-based generative approaches to design novel antibody sequences, however the de novo design of antibodies without a base remains largely unexplored ( 710 ).…”
Section: Introductionmentioning
confidence: 99%