К содержанию
Меню
Protein AbGen
документация
Задачи
Workflows
Инструменты
Architecture Guide
Сравнение
Вход
Главная / выбор по данным
Scope и правила
Evidence
Задачи
Оптимизация существующих белков
Affinity maturation
Combinatorial mutation design
Few-shot active learning
ML-guided directed evolution
Multi-objective optimization
Off-target optimization
Оптимизация polyspecificity
Предложение single mutations
Оптимизация specificity
Антитела и antibody-like форматы
Structure-aware дизайн antibody sequence
Antigen-conditioned antibody design
Генерация CDR backbone
Генерация CDR sequence
De novo antibody design
De novo VHH / nanobody design
Framework redesign
Humanization
Joint CDR sequence–structure design
Downstream engineering (не генерация)
CHO expression engineering
Codon optimization
Developability assessment
Expression prediction
Manufacturability assessment
Secretion engineering
Общий белковый дизайн
De novo генерация backbone
De novo non-antibody binder design
Генерация полной 3D structure
Дизайн функционального сайта
Генерация белковых sequences
Fixed-backbone inverse folding
Joint sequence–structure generation
Motif scaffolding
Target-conditioned binder generation
Workflows
Antibody inverse folding / masked CDR redesign
Antibody sequence infilling and CDR diversification
Antigen-conditioned CDR co-design
Antigen-conditioned CDR-H3 generation (PALM-H3)
Target-conditioned de novo antibody (RFantibody)
De novo backbone → inverse folding
Epitope-specific mAb CDR generation (tFold)
Few-shot / labelled mutation optimization
Fixed-backbone inverse folding
Framework humanization
Joint / multimodal sequence–structure generation
Target-conditioned miniprotein binder (BindCraft)
Sequence-first de novo generation
All-atom universal binder generation (BoltzGen)
Zero-shot mutation proposal
Инструменты
AbDiffuser
AlphaProteo
AntiFold
BindCraft
BoltzGen
Chroma
DiffAb
dyMEAN
Efficient evolution with ESM-1v
ESM-IF1
ESM3
EvoDiff
EVOLVEpro
FlowDesign
FrameDiff
HuAbDiffusion
HuDiff
Humatch
IgLM
LigandMPNN
Makowski et al. affinity–specificity co-optimization
PALM-H3
ProGen / ProGen2
ProteinGenerator
ProteinMPNN
RFantibody
RFdiffusion All-Atom (RFdiffusionAA)
RFdiffusion
Structural evolution (ESM-IF1 complex)
tFold System
Architecture
Все концепции
Active learning
Autoregressive generation
Autoregressive structure generation
Bayesian optimization
Coordinate diffusion
Few-shot mutation optimization
Flow matching
Frame diffusion
Geometric transformers
Graph neural networks
Invariant models
Inverse folding
Joint sequence–structure diffusion
Joint sequence–structure generation
Masked-token infilling
Multi-objective optimization
Protein language models
SE(3)-equivariant networks
Sequence diffusion
Torsion diffusion
Zero-shot mutation scoring
Справочники
Papers
Glossary
Compare architectures
Главная
/
Glossary
/ Few-shot
Few-shot
Малопримерный режим
Оптимизация по малому числу experimental labels, обычно десятки на раунд.
Где встречается
EVOLVEpro
Few-shot active learning
ML-guided directed evolution
Active learning
Few-shot mutation optimization