Co-optimization of therapeutic antibody affinity and specificity using machine learning models of sequence-activity data
Peer-reviewed| Поле | Значение |
|---|---|
| Authors | NR |
| Year | 2022 |
| Venue / status | Peer-reviewed journal; Nature Communications (Q1) |
| DOI | 10.1038/s41467-022-31457-3 |
| URL | https://doi.org/10.1038/s41467-022-31457-3 |
| Preprint URL | NR |
| Tools | Makowski et al. affinity–specificity co-optimization |
Задача и claimed novelty
| Поле | Значение |
|---|---|
| Main biological task | Multi-objective combinatorial mutation design for an existing therapeutic antibody |
| Claimed novelty | NR |
Архитектура
| Поле | Значение |
|---|---|
| Model architecture | Supervised models on display/deep-sequencing labels + Pareto-front proposal |
| Input representation | NR |
| Encoder / decoder | NR |
| Generative state | Combinatorial CDR mutation sets predicted to trade affinity against nonspecific binding |
| Conditioning | Eight positions in three heavy-chain CDRs in reported campaign |
| 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 | Co-optimized variants beyond sampled library; affinity gains trade against specificity |
| Tested designs | ~10^7 variant library; 4,000 sequences used for modeling; 125 clones characterized |
| Success criterion | Co-optimized variants beyond sampled library; affinity gains trade against specificity |
| Main results | Co-optimized variants beyond sampled library; affinity gains trade against specificity |
Limitations
Requires campaign-specific labelled data and library construction; not a transferable pretrained generator; one antibody case study.
Что статья показывает
- Co-optimized variants beyond sampled library; affinity gains trade against specificity
Чего статья не показывает
- Predicted confidence, likelihood, pLDDT, docking or developability scores are not experimental affinity.
Размеченные утверждения
| Утверждение | Тип | Локатор |
|---|---|---|
| Co-optimization of therapeutic antibody affinity and specificity using machine learning models of sequence-activity data | Author-reported | DOI/URL from landscape matrix |
| Publication status: Peer-reviewed journal; Nature Communications (Q1) | Author-reported | landscape matrix, publication status column |
Review queue
- Official repository/service not verified.
Source-grounded разбор статьи
Исследовательский вопрос
[MATRIX]Multi-objective combinatorial mutation design for an existing therapeutic antibody
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]Requires campaign-specific labelled data and library construction; not a transferable pretrained generator; one antibody case study.
Дополнительные ограничения интерпретации
NRПока Methods/SI не прочитаны, нельзя утверждать split leakage, точные гиперпараметры и полноту ablation.