EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

AVIDbase : A biologically accurate structural dataset of nanobody‐antigen complexes

Tadej Medved iD, Jurij Lah iD, Goran Miličić, San Hadži iD

Protein Science · 2026

DOI: 10.1002/pro.70773 ↗

At a glance

Journal article in Protein Science, 2026. Written by Tadej Medved and 3 co-authors. Institution: Biologics Drug Product Development, Novartis LLC, Mengeš, Slovenia. Major MEDLINE topics: Single-Domain Antibodies, Databases, Protein, Antigen-Antibody Complex, Antigens. 39 references. The full text is freely available (open access, CC BY 4.0). Funded by: Novartis LLC, Javna Agencija za Raziskovalno Dejavnost RS, Slovensko nacionalno superračunalniško omrežje (SLING).

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Accurate structural data in a standardized format is one of the key factors behind the success of machine learning (ML)‐based methods for protein design and structure prediction. However, their application to nanobody‐antigen complexes has lower success rates compared to globular protein complexes, partly due to the limited amount of high‐quality structural data. While several dedicated databases already exist, automated assembly pipelines frequently overlook various artifacts, which act as additional noise and limit the effectiveness of ML applications. Common issues include incorrectly defined antigen assemblies, redundancy bias, inclusion of crystal contacts, strained geometry due to crystal packing, as well as missing density or post‐translational modifications near the interface. To address these issues, we present A ntigen‐ V HH I nterface D atabase ( AVIDbase ), a highly curated dataset of nanobody‐antigen structures. In addition to correcting structural artifacts, the dataset provides a nonredundant set of structures with standardized chain identifiers, harmonized metadata, and cleaned atomic coordinates in a ready‐to‐use format for ML applications. AVIDbase is available on GitHub ( github.com/Novartis/AVIDbase ) and Zenodo ( doi.org/10.5281/zenodo.20488703 ).

Subjects

MeSH terms (MEDLINE)

Single-Domain AntibodiesDatabases, ProteinAntigen-Antibody ComplexAntigensMachine LearningProtein ConformationModels, Molecular

Author keywords

PTMassemblydatabaseinterfacenanobody

Substances

Single-Domain AntibodiesAntigen-Antibody ComplexAntigens

Authors and institutions

  • Biologics Drug Product Development, Novartis LLC, Mengeš, Slovenia.
  • Faculty of Chemistry and Chemical Technology, University of Ljubljana, Ljubljana, Slovenia.

ORCID profiles: Tadej Medved, Jurij Lah, San Hadži

Funding

  • Novartis LLC
  • Javna Agencija za Raziskovalno Dejavnost RS
  • Slovensko nacionalno superračunalniško omrežje (SLING)

Cite

APA

Medved, T., Lah, J., Miličić, G., & Hadži, S. (2026). AVIDbase : A biologically accurate structural dataset of nanobody‐antigen complexes. Protein Science. https://doi.org/10.1002/pro.70773

Vancouver

Medved T, Lah J, Miličić G, Hadži S. AVIDbase : A biologically accurate structural dataset of nanobody‐antigen complexes. Protein Science. 2026. doi:10.1002/pro.70773

BibTeX
@article{Medved2026,
  author = {Medved, Tadej and Lah, Jurij and Miličić, Goran and Hadži, San},
  title = {AVIDbase : A biologically accurate structural dataset of nanobody‐antigen complexes},
  journal = {Protein Science},
  year = {2026},
  doi = {10.1002/pro.70773}
}
Start for free →