Structure-Aware Antibody Design with Affinity-Optimized Inverse Folding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhao, Xinyan, Tang, Yi-Ching, Monsia, Rivaaj, Cantu, Victor J., Ramesh, Ashwin Kumar, Liu, Xiaozhong, An, Zhiqiang, Jiang, Xiaoqian, Kim, Yejin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908723311542272
author Zhao, Xinyan
Tang, Yi-Ching
Monsia, Rivaaj
Cantu, Victor J.
Ramesh, Ashwin Kumar
Liu, Xiaozhong
An, Zhiqiang
Jiang, Xiaoqian
Kim, Yejin
author_facet Zhao, Xinyan
Tang, Yi-Ching
Monsia, Rivaaj
Cantu, Victor J.
Ramesh, Ashwin Kumar
Liu, Xiaozhong
An, Zhiqiang
Jiang, Xiaoqian
Kim, Yejin
contents Motivation: The clinical efficacy of antibody therapeutics critically depends on high-affinity target engagement, yet laboratory affinity-maturation campaigns are slow and costly. In computational settings, most protein language models (PLMs) are not trained to favor high-affinity antibodies, and existing preference optimization approaches introduce substantial computational overhead without clear affinity gains. Therefore, this work proposes SimBinder-IF, which converts the inverse folding model ESM-IF into an antibody sequence generator by freezing its structure encoder and training only its decoder to prefer experimentally stronger binders through preference optimization. Results: On the 11-assay AbBiBench benchmark, SimBinder-IF achieves a 55 percent relative improvement in mean Spearman correlation between log-likelihood scores and experimentally measured binding affinity compared to vanilla ESM-IF (from 0.264 to 0.410). In zero-shot generalization across four unseen antigen-antibody complexes, the correlation improves by 156 percent (from 0.115 to 0.294). SimBinder-IF also outperforms baselines in top-10 precision for ten-fold or greater affinity improvements. A case study redesigning antibody F045-092 for A/California/04/2009 (pdmH1N1) shows that SimBinder-IF proposes variants with substantially lower predicted binding free energy changes than ESM-IF (mean Delta Delta G -75.16 vs -46.57). Notably, SimBinder-IF trains only about 18 percent of the parameters of the full ESM-IF model, highlighting its parameter efficiency for high-affinity antibody generation.
format Preprint
id arxiv_https___arxiv_org_abs_2512_17815
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Structure-Aware Antibody Design with Affinity-Optimized Inverse Folding
Zhao, Xinyan
Tang, Yi-Ching
Monsia, Rivaaj
Cantu, Victor J.
Ramesh, Ashwin Kumar
Liu, Xiaozhong
An, Zhiqiang
Jiang, Xiaoqian
Kim, Yejin
Computational Engineering, Finance, and Science
Biomolecules
Motivation: The clinical efficacy of antibody therapeutics critically depends on high-affinity target engagement, yet laboratory affinity-maturation campaigns are slow and costly. In computational settings, most protein language models (PLMs) are not trained to favor high-affinity antibodies, and existing preference optimization approaches introduce substantial computational overhead without clear affinity gains. Therefore, this work proposes SimBinder-IF, which converts the inverse folding model ESM-IF into an antibody sequence generator by freezing its structure encoder and training only its decoder to prefer experimentally stronger binders through preference optimization. Results: On the 11-assay AbBiBench benchmark, SimBinder-IF achieves a 55 percent relative improvement in mean Spearman correlation between log-likelihood scores and experimentally measured binding affinity compared to vanilla ESM-IF (from 0.264 to 0.410). In zero-shot generalization across four unseen antigen-antibody complexes, the correlation improves by 156 percent (from 0.115 to 0.294). SimBinder-IF also outperforms baselines in top-10 precision for ten-fold or greater affinity improvements. A case study redesigning antibody F045-092 for A/California/04/2009 (pdmH1N1) shows that SimBinder-IF proposes variants with substantially lower predicted binding free energy changes than ESM-IF (mean Delta Delta G -75.16 vs -46.57). Notably, SimBinder-IF trains only about 18 percent of the parameters of the full ESM-IF model, highlighting its parameter efficiency for high-affinity antibody generation.
title Structure-Aware Antibody Design with Affinity-Optimized Inverse Folding
topic Computational Engineering, Finance, and Science
Biomolecules
url https://arxiv.org/abs/2512.17815