HuAbDiffusion

HuAbDiffusion YabXnization

Peer-reviewed Prospective flag Direct antibody flag

Summary card

Поле	Значение
Год / version	2025 / NR
Last verified	2026-08-13
Что генерирует	Whole humanized V-region sequences
Форматы	Monoclonal antibody variable domains
De novo или optimization	Framework generation/humanization
DOI	10.1093/bib/bbaf658
Paper	https://doi.org/10.1093/bib/bbaf658
GitHub / repo	NR
Service / docs	https://www.genscript.com/yabXnization-heterogenization-tool.html
Лицензия	Commercial web service terms; source code NR
Доступность / commercial-use	Service/access request; no verified open implementation

Задача и generated object

Used in these tasks: Humanization, Framework redesign

Part of these workflows: Framework humanization

Поле	Значение
Required input	Three CDR sequences
Conditioning	CDRs retained; human sequence prior
Fixed	NR
Editable	NR
Generated	Whole humanized V-region sequences
Single mutations	Implicit
Combinatorial mutations	Yes
Multi-objective	No explicit multi-objective optimizer

Architecture and operating principle

Author-reported (matrix)

Поле	Значение
Base architecture	NR
Model type	7 Humanization/framework redesign generator
Representation	NR
Generation unit	Whole humanized V-region sequences
Conditioning	CDRs retained; human sequence prior
Output	Whole humanized V-region sequences
Sampling	NR
Internal scorer	Pretrained language-model narrowing/ranking
Structure generation	No public integrated structure predictor

Input representation

Поле	Значение
Node features	NR
Edge / pair features	NR

Encoder и decoder

Поле	Значение
Encoder	NR
Decoder	NR

Latent or generative state

Conditioning mechanism

CDRs retained; human sequence prior

Generation mechanism

Discrete language diffusion

Sampling: NR

Training objectives

Losses: NR

Mathematical details: NR

Integrated evaluation

Pretrained language-model narrowing/ranking

Architectural limitations

Service opacity and no verified source.
humanization metrics do not equal immunogenicity.
experimental subset details need extraction.

Uses these architecture components: Sequence diffusion

Generator versus evaluation

Component	Role	Architecture class	Integrated or external	Required or optional	Influences generation or post hoc	Runtime included or excluded
HuAbDiffusion	primary_generator	Discrete language diffusion	integrated	required	generation	included
Pretrained language-model narrowing/ranking	integrated_scorer	NR	integrated	optional	post_hoc	unknown
Binding retention, structure, developability/immunogenicity assessment	mandatory_external_dependency	NR	external	required	post_hoc	excluded
Experimental binding/function assays	experimental_validation	NR	external	required	post_hoc	excluded

Training data и biases

Поле	Значение
Training data	NR
Preprocessing	NR
Negative examples	NR
Fine-tuning	NR
Label requirements	NR
Training hardware	NR
Training time	NR

Inference hardware и runtime

Поле	Значение
Candidate count	NR
Batch size	NR
Generation steps	NR
Recycles	NR
Size	NR
Reported runtime	NR
Runtime boundary	NR
CPU	NR
GPU	NR
VRAM	NR
RAM	NR
Disk	NR
Databases	NR

Experimental evidence

Поле	Значение
Evidence class	2–3 retrospective + prospective binding claim
Retrospective	Evaluated on 22 mAbs; exact prospective subset and per-variant count: NR here
Prospective binding	Paper reports retained or increased affinity for selected humanized antibodies
Prospective function	NR
Developability	NR
Tested design count	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
Independent validation	NR
Summary	Paper reports retained or increased affinity for selected humanized antibodies

Limitations and failure modes

Service opacity and no verified source.
humanization metrics do not equal immunogenicity.
experimental subset details need extraction.

Deployment: Service evaluation only

Reproducibility, license, commercial use

Поле	Значение
Repo	NR
License	Commercial web service terms; source code NR
Commercial-use status	Service/access request; no verified open implementation

Alternative tools: NR

Complementary: NR

Related: NR

Not directly comparable with

AlphaProteo, BindCraft, BoltzGen, FrameDiff, RFdiffusion, RFdiffusion All-Atom (RFdiffusionAA)

Сравнение допустимо только внутри совместимых категорий. Runtime и quality AntiFold vs RFantibody нельзя читать как одну шкалу: разные starting points и generated objects.

Depends on these external modules

Primary paper и полный список источников

HuAbDiffusion: humanized antibody generation from CDRs

Source matrix: landscape xlsx, last verified 2026-08-13.

Review queue

Bibliographic title must be verified against the PDF.

Детальный source-grounded разбор

1. Что именно делает инструмент

[MATRIX]Вход: Three CDR sequences

[MATRIX]Generated object: Whole humanized V-region sequences

[MATRIX]Фиксируется: NR. Изменяется: NR.

[MATRIX]De novo или parent-based: Framework generation/humanization.

2. Архитектура в одном абзаце

[MATRIX]Discrete language diffusion

4. Математическая модель

NRТочные уравнения, losses и sampling solver будут внесены после чтения paper/code. Не подставляйте родовой UCB/transformer/diffusion, если он не подтверждён для этого инструмента.

Поля NR: conditional probability, update equations, sampling algorithm

5. Физическое устройство состояния

[MATRIX]Seed generated state / output: Whole humanized V-region sequences. Structure generation: No public integrated structure predictor.

6. Физико-химическая интерпретация

[INTERPRETATION]Координаты, frames, PLM embeddings или residue tokens не являются расчётом ΔG, электростатики или растворителя, пока это не сделано внешней energy function. Likelihood / pLDDT / docking score ≠ экспериментальный binding.

7. Training

[MATRIX]Данные: NR. Fine-tuning: NR. Negative examples: NR.

8. Inference

[MATRIX]Sampling: NR. Generation steps: NR. Runtime boundary: NR.

10. Почему, по мнению авторов, метод эффективен

[MATRIX]Author-reported summary (матрица): Paper reports retained or increased affinity for selected humanized antibodies

NRCausal support не классифицирован, пока не прочитаны ablation и experimental sections.

12. Ablation analysis

NRAblation в этом выпуске не разобрана.

13. What the model does not represent

Явный растворитель и ионная сила — обычно отсутствуют в генераторе.
Протонирование, гликозилирование, кофакторы — обычно отсутствуют, если не оговорено.
Конформационная энтропия и долгая динамика — не моделируются одним forward pass.
Экспрессия, агрегация, иммуногенность — не равны likelihood генератора.
Истинная binding free energy — не равна internal score.

14. Evidence and limitations

[MATRIX]Evidence class (матрица): 2–3 retrospective + prospective binding claim. Prospective binding: Paper reports retained or increased affinity for selected humanized antibodies. Function: NR. Developability: NR.

Service opacity and no verified source.
humanization metrics do not equal immunogenicity.
experimental subset details need extraction.

15. Reproducibility

NRРепозиторий: NR. Лицензия: Commercial web service terms; source code NR. Commit: NR, пока страница не имеет отдельного analysis YAML.

Модульная схема

Модуль	Класс модели	Вход	Выход	Обучается или frozen	Generator или evaluator
HuAbDiffusion	Discrete language diffusion	NR	NR	NR	primary_generator
Pretrained language-model narrowing/ranking	NR	NR	NR	NR	integrated_scorer
Binding retention, structure, developability/immunogenicity assessment	NR	NR	NR	NR	mandatory_external_dependency
Experimental binding/function assays	NR	NR	NR	NR	experimental_validation

Generator versus evaluator

Компонент	Роль	Integrated или external	Required или optional	Влияет на generation или только ранжирует после
HuAbDiffusion	primary_generator	integrated	required	generation
Pretrained language-model narrowing/ranking	integrated_scorer	integrated	optional	post_hoc
Binding retention, structure, developability/immunogenicity assessment	mandatory_external_dependency	external	required	post_hoc
Experimental binding/function assays	experimental_validation	external	required	post_hoc

Claim–evidence matrix

Утверждение	Предполагаемый механизм	Тип evidence	Датасет / эксперимент	Comparator	Метрика / результат	Источник	Оговорка
Paper reports retained or increased affinity for selected humanized antibodies	Discrete language diffusion	5. prospective functional validation	Evaluated on 22 mAbs; exact prospective subset and per-variant count: NR here	NR	Paper reports retained or increased affinity for selected humanized antibodies	[MATRIX] landscape matrix 2026-08-13	Не читать как полный разбор primary paper.