What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?

Antibody therapeutic development is a major focus in healthcare. To accelerate drug development, significant efforts have been directed toward the in silico design and screening of antibodies for which high modeling accuracy is necessary. To probe AlphaFold3’s (AF3) capabilities and limitations, we...

Full description

Saved in:
Bibliographic Details
Main Authors: Fatima N. Hitawala, Jeffrey J. Gray
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:mAbs
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/19420862.2025.2545601
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849231356540747776
author Fatima N. Hitawala
Jeffrey J. Gray
author_facet Fatima N. Hitawala
Jeffrey J. Gray
author_sort Fatima N. Hitawala
collection DOAJ
description Antibody therapeutic development is a major focus in healthcare. To accelerate drug development, significant efforts have been directed toward the in silico design and screening of antibodies for which high modeling accuracy is necessary. To probe AlphaFold3’s (AF3) capabilities and limitations, we tested AF3’s ability to capture the fine details and interplay between antibody structure prediction and antigen docking accuracy. With one seed, AF3 achieves a 10.2% and 13.3% high-accuracy docking success rate for antibodies and nanobodies, respectively. AF3-like models Boltz-1 and Chai-1 achieve 4.08% and 0% high-accuracy rates for antibodies, and 5% and 3.33% for nanobodies, respectively. With twenty seeds, AF3 achieves a median unbound CDR H3 RMSD accuracy of 2.9 Å … and 2.2 Å … for antibodies and nanobodies, respectively. Both AF3-like models Boltz-1 and Chai-1 improve further on antibodies (2.08 Å … and 2.71 Å …, respectively), but do poorly on nanobodies (3.78 Å … , 3.63 Å …). CDR H3 accuracy boosts AF3 complex prediction accuracy, with antigen context improving CDR H3 accuracy, particularly for loops longer than 15 residues. Combining ipTM-HA and I-pLDDT with [Formula: see text] improves discriminative power for correctly docked antibody and nanobody complexes. However, AF3’s 65% failure rate for antibody and nanobody docking (with single seed sampling) demonstrates a need to further improve antibody modeling tools.
format Article
id doaj-art-6a0974b523d34cad8c7b82e4d97f9fe6
institution Kabale University
issn 1942-0862
1942-0870
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series mAbs
spelling doaj-art-6a0974b523d34cad8c7b82e4d97f9fe62025-08-21T05:34:19ZengTaylor & Francis GroupmAbs1942-08621942-08702025-12-0117110.1080/19420862.2025.2545601What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?Fatima N. Hitawala0Jeffrey J. Gray1Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD, USADepartment of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD, USAAntibody therapeutic development is a major focus in healthcare. To accelerate drug development, significant efforts have been directed toward the in silico design and screening of antibodies for which high modeling accuracy is necessary. To probe AlphaFold3’s (AF3) capabilities and limitations, we tested AF3’s ability to capture the fine details and interplay between antibody structure prediction and antigen docking accuracy. With one seed, AF3 achieves a 10.2% and 13.3% high-accuracy docking success rate for antibodies and nanobodies, respectively. AF3-like models Boltz-1 and Chai-1 achieve 4.08% and 0% high-accuracy rates for antibodies, and 5% and 3.33% for nanobodies, respectively. With twenty seeds, AF3 achieves a median unbound CDR H3 RMSD accuracy of 2.9 Å … and 2.2 Å … for antibodies and nanobodies, respectively. Both AF3-like models Boltz-1 and Chai-1 improve further on antibodies (2.08 Å … and 2.71 Å …, respectively), but do poorly on nanobodies (3.78 Å … , 3.63 Å …). CDR H3 accuracy boosts AF3 complex prediction accuracy, with antigen context improving CDR H3 accuracy, particularly for loops longer than 15 residues. Combining ipTM-HA and I-pLDDT with [Formula: see text] improves discriminative power for correctly docked antibody and nanobody complexes. However, AF3’s 65% failure rate for antibody and nanobody docking (with single seed sampling) demonstrates a need to further improve antibody modeling tools.https://www.tandfonline.com/doi/10.1080/19420862.2025.2545601Antibody dockingnanobody dockingAlphaFold3benchmarkranking protocol
spellingShingle Fatima N. Hitawala
Jeffrey J. Gray
What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?
mAbs
Antibody docking
nanobody docking
AlphaFold3
benchmark
ranking protocol
title What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?
title_full What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?
title_fullStr What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?
title_full_unstemmed What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?
title_short What does AlphaFold3 learn about antibody and nanobody docking, and what remains unsolved?
title_sort what does alphafold3 learn about antibody and nanobody docking and what remains unsolved
topic Antibody docking
nanobody docking
AlphaFold3
benchmark
ranking protocol
url https://www.tandfonline.com/doi/10.1080/19420862.2025.2545601
work_keys_str_mv AT fatimanhitawala whatdoesalphafold3learnaboutantibodyandnanobodydockingandwhatremainsunsolved
AT jeffreyjgray whatdoesalphafold3learnaboutantibodyandnanobodydockingandwhatremainsunsolved