LigandMPNN

Peer-reviewed Prospective flag

Summary card

Поле	Значение
Год / version	2025 / NR
Last verified	2026-08-13
Что генерирует	Sequence + side-chain conformations
Форматы	Protein–ligand and biomolecular complexes
De novo или optimization	Design/redesign on fixed backbone
DOI	10.1038/s41592-025-02626-1
Paper	https://doi.org/10.1038/s41592-025-02626-1
GitHub / repo	https://github.com/dauparas/LigandMPNN
Service / docs	NR
Лицензия	MIT
Доступность / commercial-use	Code + weights

Задача и generated object

Used in these tasks: Fixed-backbone inverse folding, Дизайн функционального сайта

Part of these workflows: Fixed-backbone inverse folding

Поле	Значение
Required input	Protein backbone plus ligand/metal/nucleic-acid atomic context
Conditioning	Design masks; fixed residues; non-protein atoms; chain context
Fixed	NR
Editable	NR
Generated	Sequence + side-chain conformations
Single mutations	Yes
Combinatorial mutations	Yes
Multi-objective	No built-in Pareto optimizer

Architecture and operating principle

Author-reported (matrix)

Поле	Значение
Base architecture	NR
Model type	2 Fixed-backbone inverse-folding generator
Representation	NR
Generation unit	Sequence + side-chain conformations
Conditioning	Design masks; fixed residues; non-protein atoms; chain context
Output	Sequence + side-chain conformations
Sampling	NR
Internal scorer	Conditional sequence probability; side-chain confidence
Structure generation	No global 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

Design masks; fixed residues; non-protein atoms; chain context

Generation mechanism

ProteinMPNN-family graph model with non-protein atomic context and side-chain packing

Sampling: NR

Training objectives

Losses: NR

Mathematical details: NR

Integrated evaluation

Conditional sequence probability; side-chain confidence

Architectural limitations

Requires a credible backbone and ligand pose.
does not generate backbone or prove catalysis/binding.

Uses these architecture components: Inverse folding, Graph neural networks

Generator versus evaluation

Component	Role	Architecture class	Integrated or external	Required or optional	Influences generation or post hoc	Runtime included or excluded
LigandMPNN	primary_generator	ProteinMPNN-family graph model with non-protein atomic context and side-chain packing	integrated	required	generation	included
Conditional sequence probability; side-chain confidence	integrated_scorer	NR	integrated	optional	post_hoc	unknown
Folding, ligand-pose/interaction validation, experiments	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	3–4 prospective experimental validation
Retrospective	NR
Prospective binding	Experimental structural/functional evidence for general proteins
Prospective function	Experimental structural/functional evidence for general proteins
Developability	NR
Tested design count	Multiple ligand/enzyme/interface campaigns; aggregate count: NR
Success criterion	Experimental structural/functional evidence for general proteins
Independent validation	NR
Summary	Experimental structural/functional evidence for general proteins

Limitations and failure modes

Requires a credible backbone and ligand pose.
does not generate backbone or prove catalysis/binding.

Deployment: Limited internal pilot

Reproducibility, license, commercial use

Поле	Значение
Repo	https://github.com/dauparas/LigandMPNN
License	MIT
Commercial-use status	Code + weights

Alternative tools: NR

Complementary: NR

Related: NR

Not directly comparable with

EvoDiff, FrameDiff, ProGen / ProGen2, 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

Atomic context-conditioned protein sequence design using LigandMPNN

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

Детальный source-grounded разбор

1. Что именно делает инструмент

[MATRIX]Вход: Protein backbone plus ligand/metal/nucleic-acid atomic context

[MATRIX]Generated object: Sequence + side-chain conformations

[MATRIX]Фиксируется: NR. Изменяется: NR.

[MATRIX]De novo или parent-based: Design/redesign on fixed backbone.

2. Архитектура в одном абзаце

[MATRIX]ProteinMPNN-family graph model with non-protein atomic context and side-chain packing

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: Sequence + side-chain conformations. Structure generation: No global 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): Experimental structural/functional evidence for general proteins

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): 3–4 prospective experimental validation. Prospective binding: Experimental structural/functional evidence for general proteins. Function: Experimental structural/functional evidence for general proteins. Developability: NR.

Requires a credible backbone and ligand pose.
does not generate backbone or prove catalysis/binding.

15. Reproducibility

NRРепозиторий: https://github.com/dauparas/LigandMPNN. Лицензия: MIT. Commit: NR, пока страница не имеет отдельного analysis YAML.

Модульная схема

Модуль	Класс модели	Вход	Выход	Обучается или frozen	Generator или evaluator
LigandMPNN	ProteinMPNN-family graph model with non-protein atomic context and side-chain packing	NR	NR	NR	primary_generator
Conditional sequence probability; side-chain confidence	NR	NR	NR	NR	integrated_scorer
Folding, ligand-pose/interaction validation, experiments	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
LigandMPNN	primary_generator	integrated	required	generation
Conditional sequence probability; side-chain confidence	integrated_scorer	integrated	optional	post_hoc
Folding, ligand-pose/interaction validation, experiments	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
Experimental structural/functional evidence for general proteins	ProteinMPNN-family graph model with non-protein atomic context and side-chain packing	5. prospective functional validation	Multiple ligand/enzyme/interface campaigns; aggregate count: NR	NR	Experimental structural/functional evidence for general proteins	[MATRIX] landscape matrix 2026-08-13	Не читать как полный разбор primary paper.