Optimization of therapeutic antibodies by predicting antigen specificity from antibody sequence via deep learning
Source check: Needs source check. Not independent scientific confirmation.
Finding: Neural networks trained on about 10^4 trastuzumab variants screened for HER2 specificity were used to screen about 10^8 computational variants and identify optimised HER2-specific variants.
Technology: Deep neural networks trained on antibody-library deep-sequencing/screening data
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Predicted antigen specificity from sequence to select optimised variants.
- Test and result
- Abstract reports prediction-led identification of variants; downstream validation details not assessed.
- Limitation / still unknown
- Developability and clinical relevance not assessed.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://europepmc.org/search?query=DOI:10.1038/s41551-021-00699-9
Needs source check · editorial details
The source has not been fully checked or contains conflicting information. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 17 Nov 2026. Review scope: Europe PMC metadata and abstract only; Methods not read..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.