Active Learning framework for Antibody Binding Affinity Prediction
- 1Shiv Nadar University, Delhi-NCR
Abstract
Antibody engineering campaigns must prioritise variants from large combinatorial libraries under limited assay budgets. We present a Gaussian process (GP) active-learning benchmark for antibody-antigen binding affinity prediction across eight public AbBiBench assays spanning influenza hemagglutinin antibodies CR6261 and CR9114 and AAYL-series scFv libraries against the SARS-CoV-2 Spike HR2 peptide. The framework embeds variable heavy and light (VH/VL) sequences with ESM-2 650M, trains GP surrogates over compressed protein-language-model representations, and evaluates kernel, acquisition, representation, classical featurisation, and labelling-budget choices under a unified budget of 10\% of each full pool with a label-blind seed and a fixed held-out evaluation split. Across the benchmark, Tanimoto+UCB has the highest mean Recall@2\% point estimate among tested kernel-acquisition configurations, achieving \textbf{57.8\%} mean Recall@2\% across five random seeds (68.3\% when scored against the acquisition-eligible pool), compared with 9.6\% for random selection -- a gap that is significant by a paired Wilcoxon test across datasets (p=0.0078); its point estimate is not significantly different from RQ+UCB or Tanimoto+EI at five seeds. We adopt the continuous Tanimoto kernel of \citet{gessner2024}. The same configuration recovers the largest fractions of top binders in the influenza libraries (74.2\% on \texttt{3gbn\_h1}, 82.7\% on \texttt{3gbn\_h9}, and 84.9\% on \texttt{4fqi\_h3}) while exposing harder AAYL combinatorial landscapes where recall remains lower and more variable. An additional three-seed representation ablation, run under the original protocol and not repeated after the evaluation corrections, ranked ESM-2 650M highest (mean Recall@2\% 0.673 ± 0.28) among ProtBert, AntiBERTy, AbLang2, ProGen2, and classical encodings. On BindingGYM, the same configuration matches or exceeds ALLM-Ab on two of three datasets using the 100-sample held-out test metric. Structure-mapped Integrated Gradients and KernelSHAP analyses recover CDR-region attribution patterns consistent with contact-derived paratope geometry, though threshold-free (AUROC) analysis shows this qualitative agreement is not matched by strong residue-level discrimination. These results support Tanimoto-kernel GP active learning as a simple, reproducible strategy for label-efficient antibody affinity optimisation with qualitative structural validation.
Related
- Leakage-controlled benchmarking reveals generalization limits of deep learning for protein-ligand binding affinity prediction
- Benchmarking Large Language Models for Assay-Aware Drug-Target Binding Affinity Prediction
- Antibody-Antigen Affinity Prediction with Chain-Aware Protein Language Modeling
- Boltz2-Notebook: An Interactive Google Colab Platform for Diffusion-Based Biomolecular Structure and Binding Affinity Prediction using the Boltz2 model.
- RA-PLA: Retrieval-augmented graph convolutional networks for protein-ligand binding affinity prediction
BibTeX
@misc{krishna2026active,
title = {{Active Learning framework for Antibody Binding Affinity Prediction}},
author = {C. S. Krishna and H. Singh and S. P. Srivastava and R. K. SINGH},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.17.759787},
url = {https://www.biorxiv.org/content/10.64898/2026.10.17.759787v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Active Learning framework for Antibody Binding Affinity Prediction
AU - C. S. Krishna
AU - H. Singh
AU - S. P. Srivastava
AU - R. K. SINGH
PY - 2026
DA - 2026/10/17
PB - bioRxiv
DO - 10.64898/2026.10.17.759787
UR - https://www.biorxiv.org/content/10.64898/2026.10.17.759787v1
AB - Antibody engineering campaigns must prioritise variants from large combinatorial libraries under limited assay budgets. We present a Gaussian process (GP) active-learning benchmark for antibody-antigen binding affinity prediction across eight public AbBiBench assays spanning influenza hemagglutinin antibodies CR6261 and CR9114 and AAYL-series scFv libraries against the SARS-CoV-2 Spike HR2 peptide. The framework embeds variable heavy and light (VH/VL) sequences with ESM-2 650M, trains GP surrogates over compressed protein-language-model representations, and evaluates kernel, acquisition, representation, classical featurisation, and labelling-budget choices under a unified budget of 10\% of each full pool with a label-blind seed and a fixed held-out evaluation split. Across the benchmark, Tanimoto+UCB has the highest mean Recall@2\% point estimate among tested kernel-acquisition configurations, achieving \textbf{57.8\%} mean Recall@2\% across five random seeds (68.3\% when scored against the acquisition-eligible pool), compared with 9.6\% for random selection -- a gap that is significant by a paired Wilcoxon test across datasets (p=0.0078); its point estimate is not significantly different from RQ+UCB or Tanimoto+EI at five seeds. We adopt the continuous Tanimoto kernel of \citet{gessner2024}. The same configuration recovers the largest fractions of top binders in the influenza libraries (74.2\% on \texttt{3gbn\_h1}, 82.7\% on \texttt{3gbn\_h9}, and 84.9\% on \texttt{4fqi\_h3}) while exposing harder AAYL combinatorial landscapes where recall remains lower and more variable. An additional three-seed representation ablation, run under the original protocol and not repeated after the evaluation corrections, ranked ESM-2 650M highest (mean Recall@2\% 0.673 ± 0.28) among ProtBert, AntiBERTy, AbLang2, ProGen2, and classical encodings. On BindingGYM, the same configuration matches or exceeds ALLM-Ab on two of three datasets using the 100-sample held-out test metric. Structure-mapped Integrated Gradients and KernelSHAP analyses recover CDR-region attribution patterns consistent with contact-derived paratope geometry, though threshold-free (AUROC) analysis shows this qualitative agreement is not matched by strong residue-level discrimination. These results support Tanimoto-kernel GP active learning as a simple, reproducible strategy for label-efficient antibody affinity optimisation with qualitative structural validation.
ER -