nf-proteindesign¶

Overview¶

nf-proteindesign is a Nextflow pipeline for high-throughput protein design using Boltzgen, an all-atom generative diffusion model. Design proteins, peptides, and nanobodies to bind various biomolecular targets with a comprehensive suite of downstream analysis modules.

Analysis Modules¶

Key Features¶

• :material-parallel: Parallel Processing: Run multiple design specifications simultaneously
• YAML-Based Design: Complete control with custom design specifications
• Comprehensive Analysis: Six optional analysis modules for quality assessment
• Sequence Optimization: ProteinMPNN + Boltz-2 validation workflow
• Container Support: Full Docker compatibility
• :material-gpu: GPU Acceleration: Optimized for NVIDIA GPU execution
• Organized Outputs: Structured results with unified reporting

:material-pipeline: Pipeline Workflow¶

graph TB
    A[Samplesheet<br/>Design YAMLs] --> B{Boltzgen<br/>Precomputed?}
    B -->|No| C[Run Boltzgen Design]
    B -->|Yes| D[Use Precomputed]
    C --> E[Budget Designs<br/>CIF + NPZ]
    D --> E

    E --> F{ProteinMPNN<br/>Enabled?}
    F -->|No| Z[Boltzgen Outputs Only]
    F -->|Yes| G[Sequence Optimization<br/>Parallel per Design]

    G --> H{Boltz-2<br/>Enabled?}
    H -->|No| Y[MPNN Sequences Only]
    H -->|Yes| I[Prepare Sequences<br/>Split + Target]

    I --> J[Boltz-2 Structure<br/>Prediction]
    J --> K[Boltz-2 Structures<br/>CIF + NPZ]

    K --> L{Analysis<br/>Modules?}
    L -->|IPSAE| M[Interface Scoring]
    L -->|PRODIGY| N[Binding Affinity]
    L -->|Foldseek| O[Structural Search]

    M --> P{Consolidate?}
    N --> P
    O --> P
    P -->|Yes| Q[Unified CSV + HTML<br/>Report]

    Q --> R[Final Results]
    K --> R
    Z --> R
    Y --> R

    style C fill:#9C27B0,stroke:#9C27B0,color:#fff
    style G fill:#8E24AA,stroke:#8E24AA,color:#fff
    style J fill:#7B1FA2,stroke:#7B1FA2,color:#fff
    style Q fill:#6A1B9A,stroke:#6A1B9A,color:#fff

    classDef note fill:#FFF3E0,stroke:#FF9800,color:#000
    class L note

Quick Start¶

Get started with nf-proteindesign in minutes:

# 1. Install Nextflow (>=23.04.0)
curl -s https://get.nextflow.io | bash

# 2. Run the pipeline
nextflow run seqeralabs/nf-proteindesign \
    -profile docker \
    --input samplesheet.csv \
    --outdir results

What Can You Design?¶

The pipeline leverages Boltzgen's capabilities to design:

• Proteins: Full-length protein binders targeting specific interfaces
• Peptides: Short peptide sequences for tight binding
• Nanobodies: Compact antibody fragments for therapeutic applications
• Multi-target Binders: Design to multiple targets simultaneously

All with the flexibility to specify: - Binding site residues - Designed chain type (protein, peptide, nanobody) - Chain length constraints - Custom diffusion parameters

Documentation Structure¶

Computing Requirements¶

Contributing¶

We welcome contributions! The pipeline is designed with modularity and extensibility in mind.

License¶

This project is licensed under the MIT License - see the LICENSE file for details.