Главная / Инструменты / Structural evolution (ESM-IF1 complex)

Structural evolution (ESM-IF1 complex)

structural evolution ESM-IF1 complex evolution

Peer-reviewed Prospective flag Direct antibody flag

Summary card

Поле	Значение
Год / version	2024 / NR
Last verified	2026-08-13
Что генерирует	Ranked sequence mutations preserving complex structure
Форматы	Protein complexes and antibody–antigen complexes
De novo или optimization	Optimizes existing complex
DOI	10.1126/science.adk8946
Paper	https://doi.org/10.1126/science.adk8946
GitHub / repo	https://github.com/varun-shanker/structural-evolution
Service / docs	NR
Лицензия	MIT
Доступность / commercial-use	Code/scripts

Задача и generated object

Used in these tasks: Предложение single mutations, Affinity maturation, Structure-aware дизайн antibody sequence

Part of these workflows: Zero-shot mutation proposal

Поле	Значение
Required input	Parent complex structure/backbone
Conditioning	Complex geometry; selected mutable positions
Fixed	NR
Editable	NR
Generated	Ranked sequence mutations preserving complex structure
Single mutations	Yes
Combinatorial mutations	Limited/derived combinations; primary strength is ranked substitutions
Multi-objective	No explicit multi-objective loop

Architecture and operating principle

Author-reported (matrix)

Поле	Значение
Base architecture	NR
Model type	6 Mutation generator / optimization workflow
Representation	NR
Generation unit	Ranked sequence mutations preserving complex structure
Conditioning	Complex geometry; selected mutable positions
Output	Ranked sequence mutations preserving complex structure
Sampling	NR
Internal scorer	Structure-conditioned sequence likelihood
Structure generation	No structure prediction; uses supplied complex backbone

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

Complex geometry; selected mutable positions

Generation mechanism

Inverse-folding likelihood on complete complex backbone to rank substitutions

Sampling: NR

Training objectives

Losses: NR

Mathematical details: NR

Integrated evaluation

Structure-conditioned sequence likelihood

Architectural limitations

Requires credible complex structure.
likelihood targets structural compatibility, not direct affinity.
epistasis/multiobjective scope limited.

Uses these architecture components: Inverse folding, Zero-shot mutation scoring

Generator versus evaluation

Component	Role	Architecture class	Integrated or external	Required or optional	Influences generation or post hoc	Runtime included or excluded
ESM-IF1 conditional mutation proposal	primary_generator	Inverse-folding likelihood on complete complex backbone to rank substitutions	integrated	required	generation	included
Structure-conditioned sequence likelihood	integrated_scorer	NR	integrated	optional	post_hoc	unknown
Complex-quality validation and functional assays	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	4 prospective functional validation
Retrospective	Multiple protein/antibody systems; exact count by system: paper-level extraction required
Prospective binding	Improved antibody potency/resilience to viral escape reported
Prospective function	Improved antibody potency/resilience to viral escape reported
Developability	NR
Tested design count	Multiple protein/antibody systems; exact count by system: paper-level extraction required
Success criterion	Improved antibody potency/resilience to viral escape reported
Independent validation	NR
Summary	Improved antibody potency/resilience to viral escape reported

Limitations and failure modes

Requires credible complex structure.
likelihood targets structural compatibility, not direct affinity.
epistasis/multiobjective scope limited.

Deployment: Limited internal pilot

Reproducibility, license, commercial use

Поле	Значение
Repo	https://github.com/varun-shanker/structural-evolution
License	MIT
Commercial-use status	Code/scripts

Alternative tools: NR

Complementary: NR

Related: NR

Not directly comparable with

EVOLVEpro, FrameDiff, Makowski et al. affinity–specificity co-optimization, RFantibody, RFdiffusion, RFdiffusion All-Atom (RFdiffusionAA), tFold System

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

Unsupervised evolution of protein and antibody complexes with a structure-informed language model

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

Детальный source-grounded разбор

1. Что именно делает инструмент

[MATRIX]Вход: Parent complex structure/backbone

[MATRIX]Generated object: Ranked sequence mutations preserving complex structure

[MATRIX]Фиксируется: NR. Изменяется: NR.

[MATRIX]De novo или parent-based: Optimizes existing complex.

2. Архитектура в одном абзаце

[MATRIX]Inverse-folding likelihood on complete complex backbone to rank substitutions

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: Ranked sequence mutations preserving complex structure. Structure generation: No structure prediction; uses supplied complex backbone.

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): Improved antibody potency/resilience to viral escape reported

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): 4 prospective functional validation. Prospective binding: Improved antibody potency/resilience to viral escape reported. Function: Improved antibody potency/resilience to viral escape reported. Developability: NR.

Requires credible complex structure.
likelihood targets structural compatibility, not direct affinity.
epistasis/multiobjective scope limited.

15. Reproducibility

NRРепозиторий: https://github.com/varun-shanker/structural-evolution. Лицензия: MIT. Commit: NR, пока страница не имеет отдельного analysis YAML.

Модульная схема

Модуль	Класс модели	Вход	Выход	Обучается или frozen	Generator или evaluator
ESM-IF1 conditional mutation proposal	Inverse-folding likelihood on complete complex backbone to rank substitutions	NR	NR	NR	primary_generator
Structure-conditioned sequence likelihood	NR	NR	NR	NR	integrated_scorer
Complex-quality validation and functional assays	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
ESM-IF1 conditional mutation proposal	primary_generator	integrated	required	generation
Structure-conditioned sequence likelihood	integrated_scorer	integrated	optional	post_hoc
Complex-quality validation and functional assays	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
Improved antibody potency/resilience to viral escape reported	Inverse-folding likelihood on complete complex backbone to rank substitutions	5. prospective functional validation	Multiple protein/antibody systems; exact count by system: paper-level extraction required	NR	Improved antibody potency/resilience to viral escape reported	[MATRIX] landscape matrix 2026-08-13	Не читать как полный разбор primary paper.