HuAbDiffusion: humanized antibody generation from CDRs
Peer-reviewed| Поле | Значение |
|---|---|
| Authors | NR |
| Year | 2025 |
| Venue / status | Peer-reviewed journal; Briefings in Bioinformatics; Q1 status NR |
| DOI | 10.1093/bib/bbaf658 |
| URL | https://doi.org/10.1093/bib/bbaf658 |
| Preprint URL | NR |
| Tools | HuAbDiffusion |
Задача и claimed novelty
| Поле | Значение |
|---|---|
| Main biological task | Humanized variable-region generation from CDRs |
| Claimed novelty | NR |
Архитектура
| Поле | Значение |
|---|---|
| Model architecture | Discrete language diffusion |
| Input representation | NR |
| Encoder / decoder | NR |
| Generative state | Whole humanized V-region sequences |
| Conditioning | CDRs retained; human sequence prior |
| Sampling | NR |
Training
| Поле | Значение |
|---|---|
| Objectives / losses | NR |
| Dataset | NR |
| Preprocessing | NR |
Inference
NR
Baselines и retrospective benchmarks
| Поле | Значение |
|---|---|
| Baselines | NR |
| Retrospective | NR |
Prospective experimental validation
| Поле | Значение |
|---|---|
| Prospective | Paper reports retained or increased affinity for selected humanized antibodies |
| Tested designs | Evaluated on 22 mAbs; exact prospective subset and per-variant count: NR here |
| Success criterion | Paper reports retained or increased affinity for selected humanized antibodies |
| Main results | Paper reports retained or increased affinity for selected humanized antibodies |
Limitations
Service opacity and no verified source; humanization metrics do not equal immunogenicity; experimental subset details need extraction.
Что статья показывает
- Paper reports retained or increased affinity for selected humanized antibodies
Чего статья не показывает
- Predicted confidence, likelihood, pLDDT, docking or developability scores are not experimental affinity.
Размеченные утверждения
| Утверждение | Тип | Локатор |
|---|---|---|
| HuAbDiffusion: humanized antibody generation from CDRs | Interpretation | DOI/URL from landscape matrix |
| Publication status: Peer-reviewed journal; Briefings in Bioinformatics; Q1 status NR | Author-reported | landscape matrix, publication status column |
Review queue
- Bibliographic title must be verified against the PDF.
Source-grounded разбор статьи
Исследовательский вопрос
[MATRIX]Humanized variable-region generation from CDRs
Утверждения и поддержка evidence
[MATRIX]Claimed novelty: NR
[INTERPRETATION]Статья не «доказывает» общее преимущество метода вне своего экспериментального дизайна.
Что представленный evidence не поддерживает
- Predicted confidence, likelihood, pLDDT, docking or developability scores are not experimental affinity.
Ограничения, признанные авторами
[MATRIX]Service opacity and no verified source; humanization metrics do not equal immunogenicity; experimental subset details need extraction.
Дополнительные ограничения интерпретации
NRПока Methods/SI не прочитаны, нельзя утверждать split leakage, точные гиперпараметры и полноту ablation.