Aug 2026· Frontiers in Immunology· Vol 17· 0 citations· 18 references
Medicine
TL;DR
Existing immunological datasets can support feature-constrained transfer learning for data-efficient prioritization of antibody:antigen interactions across closely related Sarbecoviruses, particularly when conserved epitope regions can be aligned and limited target-specific measurements are available for calibration.
Abstract
Antiviral antibody discovery is constrained by the enormous combinatorial space of antibody paratopes and viral epitopes, while novel approaches to search for these vast possibilities such as machine learning models for antibody affinity prediction typically require large target-specific datasets that are rarely available for emerging pathogens. To address this problem, we investigated the ability of transfer learning to reduce the required training data and still create a robust predictive model of antibody affinities. We show that antibody-affinity prediction can be locally transferred from a data-rich viral target to a related data-limited viral target by combining transfer learning, epitope-centered antigen encoding, antibody CDR descriptors, and targeted experimental validation. Using SARS-CoV-2 receptor-binding-domain (RBD) interaction data as the source domain, we developed dual-encoder neural-network models that capture functional determinants of antibody recognition and supports prediction of SARS-CoV-1 variant:antibody affinities despite limited direct training data. High-throughput ELISA measurements generated compact validation datasets spanning heavy-chain-only antibodies (HCAbs) and conventional antibodies, enabling iterative model refinement through similarity-guided, model-informed data expansion. Across expanded withheld-antibody evaluations, the final models showed improved but antibody-dependent predictive performance, with mean Pearson correlations of 0.71 ± 0.12 for SARS-CoV-2 and 0.63 ± 0.09 for SARS-CoV-1, and best-performing holdouts reaching 0.85 and 0.78, respectively. These results demonstrate that existing immunological datasets can support feature-constrained transfer learning for data-efficient prioritization of antibody:antigen interactions across closely related Sarbecoviruses, particularly when conserved epitope regions can be aligned and limited target-specific measurements are available for calibration.
Drug repurposing offers a time-efficient strategy for identifying therapeutics against emerging pathogens such as SARS-CoV-2. In this study, we apply MPS2IT-DTI (Molecule and Protein Sequence to Image Transformer for Drug-Target Interaction), a deep learning framework that represents molecular (SMILES) and protein (FASTA) sequences as images using k-mer frequency encoding, enabling convolutional neural networks to capture spatial compositional patterns associated with biochemical interactions. A curated dataset (BindingDB-FDA) containing 83,165 binding interactions from 1640 FDA-approved ligands and 3270 targets was constructed from BindingDB, with binding scores derived from the KIBA scoring system. An enhanced variant, MPS2IT+MN, incorporating max-norm regularization, was introduced to improve generalization. The model was applied to predict binding affinities between 33 FDA-approved antiviral drugs and six key SARS-CoV-2 non-structural proteins. Results consistently identified five antivirals - MK-5172 (Grazoprevir), Simeprevir, Lopinavir, Etravirine, and Atazanavir - as top-ranked candidates across all targets. Comparative analysis with the MT-DTI model demonstrated competitive and, in several cases, superior ranking performance despite a simpler architecture. Importantly, these predictions are supported by independent experimental and clinical evidence, highlighting the potential of image-based representations as a computationally efficient and biologically meaningful approach for drug-target interaction prediction and drug repurposing.
Jackson G. de Souza, Marcelo A. C. Fernandes, Raquel de Melo Barbosa· Computational biology and ch...· 0 citations
This study integrates large language model-guided antibody design with experimental validation to assess the real-world performance of computationally optimized antibodies, and will clarify how effectively AI-generated sequences translate into functional high-affinity binders, informing the development of future therapeutics.
Sanaz Zebardast, F. Ashraf, Zihao Zhang et al.· Journal of Immunology· 0 citations
A combined experimental and computational characterization of SARS-CoV-2 variant-specific neutralisation across infection, vaccination and hybrid immunity-driven cohorts is provided, offering a hypothesis-generating framework to contextualize observed differences in antibody responses and epitope recognition across variants.
Jyoti Sawant, Ajit Patil, Madhuri Thakar et al.· Frontiers in Immunology· 0 citations
Two human-derived monoclonal antibodies are characterized that recognize conserved epitopes on the SARS-CoV-2 RBD and retain activity across antigenically distinct variants, and conserved, mutationally constrained epitopes may serve as targets for vaccines designed to elicit antibody responses resilient to ongoing SARS-CoV-2 evolution and future sarbecovirus emergence.
M. Abernathy, William B. Foreman, Jasmyn A. Lopez et al.· bioRxiv· 0 citations
ABSTRACT Antibodies are renowned for their ability to bind diverse targets with high affinity and specificity, yet identifying binders with predefined epitope specificity remains a major challenge. In this study, we investigate the concept of mimic antibodies–antibodies that recapitulate the binding mode of a target’s cognate ligand. Through a systematic analysis of the Protein Data Bank (PDB), we show that such mimicry is widespread and arises through diverse structural mechanisms, such as single-loop, multi-loop and scattered interaction mimicry. These findings indicate that protein interfaces impose strong constraints on binding, leading to convergent interaction solutions that can be independently discovered by antibodies. Building on these findings, we developed a ligand-guided strategy to mine immune repertoire data by selecting antibodies whose predicted binding interfaces mimic the interaction motif of a cognate ligand. Applied to the interaction between interleukin-18 (IL-18) and its receptor alpha (IL-18RA), mimicry-guided screening of a 20,000-sequence repertoire yielded 31 candidates, 11 of which (35% hit rate) bound the IL-18RA D3 domain, with eight reaching sub-nanomolar affinities that surpass the cognate ligand. Our findings establish mimic antibodies as a promising strategy for rational antibody selection, engineering, and design, with broad implications for therapeutic antibody development and drug discovery.
Brennan Abanades, J. López-Morales, Ivana Tanasijević et al.· mAbs· 0 citations
It is shown that increased sampling and AlphaFold3 generally improve performance relative to default sampling and AlphaFold2, however predictive accuracy and improvement levels varied considerably among interface classes, with antibody-peptide complexes representing a challenge despite their small antigen size.
Rui Yin, S. Saravanakumar, S. Shi et al.· bioRxiv· 1 citation
