An interpretable open platform for sequence-based antibody developability prediction
- 1Institute for Protein Innovation
Abstract
Antibody developability is increasingly predictable from sequence, yet software and trained models are rarely made available. We present DELPHI, open software for training developability predictors from labelled antibody assay data, together with ready-to-run, retrainable models. DELPHI compares 25 language-model and classifier combinations under CDR H3-cluster cross-validation that reduces sequence-similarity leakage, measures how performance changes with labelled training-set size, and reports residue-level model attributions. Applied to in-house polyreactivity and size-exclusion (SEC) data, it reaches mean AUC 0.959 and 0.933. Trained on those data alone, it transfers to a 246,293-antibody public library (AUC 0.950 with our deployed model) and ranks polyreactivity at a level similar to the best reported Ginkgo competition point estimate, without training on its data. Any laboratory can screen candidates before running assays, generate residue-level engineering hypotheses, and retrain DELPHI for a new assay.
Related
- Computational Prediction of Recombinant Expression Efficiency for Industrial Enzymes: An Integrated Co-Translational Sequence Analysis Platform
- kitAb: an automated framework for assay-specific antibody developability modeling and large-scale landscape analysis
- Archetype analysis of multiplexed antibody imaging reveals interpretable morphological programs in optical pooled screens
- 211 A discovery-to-therapeutics platform for functional antibody targeting of complex membrane-embedded tumor antigens: validation with claudin family members
- ToxINsight: An Integrated Web Platform for Sequence, Structure, and Ligand Binding Analysis of Snake Venom Toxins
BibTeX
@misc{nguyen2026interpretable,
title = {{An interpretable open platform for sequence-based antibody developability prediction}},
author = {H. N. Nguyen and D. Kothiwal and Y. Su and M. A. Kieu and R. Cao and H. Zhu and A. A. R. Teixeira},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.16.750421},
url = {https://www.biorxiv.org/content/10.64898/2026.09.16.750421v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - An interpretable open platform for sequence-based antibody developability prediction
AU - H. N. Nguyen
AU - D. Kothiwal
AU - Y. Su
AU - M. A. Kieu
AU - R. Cao
AU - H. Zhu
AU - A. A. R. Teixeira
PY - 2026
DA - 2026/09/16
PB - bioRxiv
DO - 10.64898/2026.09.16.750421
UR - https://www.biorxiv.org/content/10.64898/2026.09.16.750421v1
AB - Antibody developability is increasingly predictable from sequence, yet software and trained models are rarely made available. We present DELPHI, open software for training developability predictors from labelled antibody assay data, together with ready-to-run, retrainable models. DELPHI compares 25 language-model and classifier combinations under CDR H3-cluster cross-validation that reduces sequence-similarity leakage, measures how performance changes with labelled training-set size, and reports residue-level model attributions. Applied to in-house polyreactivity and size-exclusion (SEC) data, it reaches mean AUC 0.959 and 0.933. Trained on those data alone, it transfers to a 246,293-antibody public library (AUC 0.950 with our deployed model) and ranks polyreactivity at a level similar to the best reported Ginkgo competition point estimate, without training on its data. Any laboratory can screen candidates before running assays, generate residue-level engineering hypotheses, and retrain DELPHI for a new assay.
ER -