AbBiBench: A Benchmark for Antibody Binding Affinity Maturation and Design

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhao, Xinyan, Tang, Yi-Ching, Singh, Akshita, Cantu, Victor J, An, KwanHo, Lee, Junseok, Stogsdill, Adam E, Hamdi, Ibraheem M, Ramesh, Ashwin Kumar, An, Zhiqiang, Jiang, Xiaoqian, Kim, Yejin
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909836191465472
author Zhao, Xinyan
Tang, Yi-Ching
Singh, Akshita
Cantu, Victor J
An, KwanHo
Lee, Junseok
Stogsdill, Adam E
Hamdi, Ibraheem M
Ramesh, Ashwin Kumar
An, Zhiqiang
Jiang, Xiaoqian
Kim, Yejin
author_facet Zhao, Xinyan
Tang, Yi-Ching
Singh, Akshita
Cantu, Victor J
An, KwanHo
Lee, Junseok
Stogsdill, Adam E
Hamdi, Ibraheem M
Ramesh, Ashwin Kumar
An, Zhiqiang
Jiang, Xiaoqian
Kim, Yejin
contents We introduce AbBiBench (Antibody Binding Benchmarking), a benchmarking framework for antibody binding affinity maturation and design. Unlike previous strategies that evaluate antibodies in isolation, typically by comparing them to natural sequences with metrics such as amino acid recovery rate or structural RMSD, AbBiBench instead treats the antibody-antigen (Ab-Ag) complex as the fundamental unit. It evaluates an antibody design's binding potential by measuring how well a protein model scores the full Ab-Ag complex. We first curate, standardize, and share more than 184,500 experimental measurements of antibody mutants across 14 antibodies and 9 antigens-including influenza, lysozyme, HER2, VEGF, integrin, Ang2, and SARS-CoV-2-covering both heavy-chain and light-chain mutations. Using these datasets, we systematically compare 15 protein models including masked language models, autoregressive language models, inverse folding models, diffusion-based generative models, and geometric graph models by comparing the correlation between model likelihood and experimental affinity values. Additionally, to demonstrate AbBiBench's generative utility, we apply it to antibody F045-092 in order to introduce binding to influenza H1N1. We sample new antibody variants with the top-performing models, rank them by the structural integrity and biophysical properties of the Ab-Ag complex, and assess them with in vitro ELISA binding assays. Our findings show that structure-conditioned inverse folding models outperform others in both affinity correlation and generation tasks. Overall, AbBiBench provides a unified, biologically grounded evaluation framework to facilitate the development of more effective, function-aware antibody design models.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04235
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AbBiBench: A Benchmark for Antibody Binding Affinity Maturation and Design
Zhao, Xinyan
Tang, Yi-Ching
Singh, Akshita
Cantu, Victor J
An, KwanHo
Lee, Junseok
Stogsdill, Adam E
Hamdi, Ibraheem M
Ramesh, Ashwin Kumar
An, Zhiqiang
Jiang, Xiaoqian
Kim, Yejin
Quantitative Methods
Artificial Intelligence
Computational Engineering, Finance, and Science
Machine Learning
Biomolecules
We introduce AbBiBench (Antibody Binding Benchmarking), a benchmarking framework for antibody binding affinity maturation and design. Unlike previous strategies that evaluate antibodies in isolation, typically by comparing them to natural sequences with metrics such as amino acid recovery rate or structural RMSD, AbBiBench instead treats the antibody-antigen (Ab-Ag) complex as the fundamental unit. It evaluates an antibody design's binding potential by measuring how well a protein model scores the full Ab-Ag complex. We first curate, standardize, and share more than 184,500 experimental measurements of antibody mutants across 14 antibodies and 9 antigens-including influenza, lysozyme, HER2, VEGF, integrin, Ang2, and SARS-CoV-2-covering both heavy-chain and light-chain mutations. Using these datasets, we systematically compare 15 protein models including masked language models, autoregressive language models, inverse folding models, diffusion-based generative models, and geometric graph models by comparing the correlation between model likelihood and experimental affinity values. Additionally, to demonstrate AbBiBench's generative utility, we apply it to antibody F045-092 in order to introduce binding to influenza H1N1. We sample new antibody variants with the top-performing models, rank them by the structural integrity and biophysical properties of the Ab-Ag complex, and assess them with in vitro ELISA binding assays. Our findings show that structure-conditioned inverse folding models outperform others in both affinity correlation and generation tasks. Overall, AbBiBench provides a unified, biologically grounded evaluation framework to facilitate the development of more effective, function-aware antibody design models.
title AbBiBench: A Benchmark for Antibody Binding Affinity Maturation and Design
topic Quantitative Methods
Artificial Intelligence
Computational Engineering, Finance, and Science
Machine Learning
Biomolecules
url https://arxiv.org/abs/2506.04235