De novo backbone → inverse folding

Практический сценарий

Нужна новая topology или motif scaffold, затем sequence.

Вход

Pipeline diagram

Backbone generation
→
Sequence design
→
Self-consistency
→
Experimental tests
Mermaid diagram
flowchart TD
    A[Motif or length or target] --> B[3D diffusion generator]
    B --> C[Backbone proposals]
    C --> D[Inverse-folding model]
    D --> E[Sequence-structure candidates]
    E --> F[Independent folding]
    F --> G[Experimental validation]

Стадии pipeline

СтадияАрхитектурный классЧто происходитГенерация или оценкаGeneratorПромежуточный объектАльтернативы
Backbone generationSE(3)-equivariant diffusionГенерируются backbone coordinatesgenerationrfdiffusionBackbonesFrameDiff, RFdiffusion All-Atom (RFdiffusionAA)
Sequence designGraph inverse foldingПодбирается sequencegenerationproteinmpnnSequence–backbone pairsESM-IF1
Self-consistencyStructure predictorAF2/RF проверяет designabilityevaluationNRFiltered designsAlphaFold2 / ColabFold
Experimental testsNRBinding/function/structureexperimentNRHitsNR

Hardware и runtime

GPU practically required for diffusion. Runtime boundary: generation vs AF2 filter must be separated. Exact hours NR.

Evidence level

RFdiffusion: prospective general protein evidence. Vanilla model fails on antibody loops.

Точки отказа

Критерии решения

Связанные задачи

De novo генерация backbone, Motif scaffolding, Fixed-backbone inverse folding

Связанные инструменты

RFdiffusion, ProteinMPNN, FrameDiff, RFdiffusion All-Atom (RFdiffusionAA)

Архитектуры

Frame diffusion, Inverse folding, SE(3)-equivariant networks

Papers

De novo design of protein structure and function with RFdiffusion, Robust deep learning-based protein sequence design using ProteinMPNN