Agent skill

binder-design

Guidance for choosing the right protein binder design tool. Use this skill when: (1) Deciding between BoltzGen, BindCraft, or RFdiffusion, (2) Planning a binder design campaign, (3) Understanding trade-offs between different approaches, (4) Selecting tools for specific target types. For specific tool parameters, use the individual tool skills (boltzgen, bindcraft, rfdiffusion, etc.).

View SKILL.md on GitHub Repository
Stars 2,009
Forks 275

Install this agent skill to your Project

npx add-skill https://github.com/FreedomIntelligence/OpenClaw-Medical-Skills/tree/main/skills/binder-design

SKILL.md

Binder Design Tool Selection

Decision tree

De novo binder design?
│
├─ Standard target → BoltzGen (recommended)
│   All-atom output (no separate ProteinMPNN step needed)
│   Better for ligand/small molecule binding
│   Single-step design (backbone + sequence + side chains)
│
├─ Need diversity/exploration → RFdiffusion + ProteinMPNN
│   Maximum backbone diversity
│   Two-step: backbone then sequence
│
├─ Integrated validation → BindCraft
│   Built-in AF2 validation
│   End-to-end pipeline
│
├─ Ligand binding → BoltzGen ✓
│   All-atom diffusion handles ligand context
│
├─ Peptide/nanobody → Germinal
│   VHH/nanobody design
│   Germline-aware optimization
│
└─ Antibody/Nanobody
    +-- VHH design --> germinal skill

Tool comparison

Tool Strengths Weaknesses Best For
BoltzGen All-atom, single-step, ligand-aware Higher GPU requirement Standard (recommended)
BindCraft End-to-end, built-in AF2 validation Less diverse Production campaigns
RFdiffusion High diversity, fast Requires ProteinMPNN Exploration, diversity
Germinal Nanobody/VHH design Specialized Antibody optimization

Recommended Pipeline: BoltzGen → Chai → QC

BoltzGen provides all-atom design with built-in side-chain packing:

Target → BoltzGen → Validate → Filter
 (pdb)  (all-atom)   (chai)     (qc)

1. Target preparation

bash
# Fetch structure from PDB
# Use pdb skill for guidance
  • Trim to binding region + 10A buffer
  • Remove waters and ligands
  • Renumber chains if needed

2. Hotspot selection

  • Choose 3-6 exposed residues
  • Prefer charged/aromatic residues
  • Cluster spatially (within 10-15A)

3. Design with BoltzGen (Recommended)

First, create a YAML config file (e.g., binder.yaml):

yaml
entities:
  - protein:
      id: B
      sequence: 70..100

  - file:
      path: target.cif
      include:
        - chain:
            id: A
      binding_types:
        - chain:
            id: A
            binding: 45,67,89

Then run:

bash
modal run modal_boltzgen.py \
  --input-yaml binder.yaml \
  --protocol protein-anything \
  --num-designs 50

Why BoltzGen?

  • All-atom output (no separate ProteinMPNN step needed)
  • Better for ligand/small molecule binding
  • Single-step design (backbone + sequence + side chains)

4. Alternative: RFdiffusion Pipeline

For maximum diversity or when backbone-only is preferred:

bash
# Step 1: Backbone generation
modal run modal_rfdiffusion.py \
  --pdb target.pdb \
  --contigs "A1-150/0 70-100" \
  --hotspot "A45,A67,A89" \
  --num-designs 500

# Step 2: Sequence design
modal run modal_ligandmpnn.py \
  --pdb-path backbone.pdb \
  --num-seq-per-target 16 \
  --sampling-temp 0.1

5. Validation

bash
modal run modal_chai1.py \
  --input-faa sequences.fasta \
  --out-dir predictions/

6. Filtering

Apply standard thresholds:

  • pLDDT > 0.80
  • ipTM > 0.50
  • PAE_interface < 10
  • scRMSD < 2.0 A

See protein-qc skill for details.

Number of designs

Stage Count Purpose
Backbone generation 500-1000 Diversity
Sequences per backbone 8-16 Sequence space
AF2 predictions All Validation
After filtering 50-200 Candidates
Experimental testing 10-50 Final selection

Common mistakes

Wrong hotspots

  • Using buried residues
  • Too many hotspots (over-constrain)
  • Wrong chain/residue numbers

Insufficient diversity

  • Too few designs generated
  • Low temperature in ProteinMPNN
  • Not exploring multiple backbones

Poor target preparation

  • Including full protein instead of binding region
  • Missing important structural features
  • Wrong protonation states

Timeline guide

Step Compute Time
RFdiffusion (500 designs) 2-4 hours
ProteinMPNN (8000 sequences) 1-2 hours
AF2 prediction (8000 sequences) 12-24 hours
Filtering and analysis 1-2 hours

Total: 1-2 days of compute

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

FreedomIntelligence/OpenClaw-Medical-Skills

vcf-annotator

Annotate VCF variants with VEP, ClinVar, gnomAD frequencies, and ancestry-aware context. Generates prioritised variant reports.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

chemist-analyst

Analyzes events through chemistry lens using molecular structure, reaction mechanisms, thermodynamics, kinetics, and analytical techniques (spectroscopy, chromatography, mass spectrometry). Provides insights on chemical processes, material properties, reaction pathways, synthesis, and analytical methods. Use when: Chemical reactions, material analysis, synthesis planning, process optimization, environmental chemistry. Evaluates: Molecular structure, reaction mechanisms, yield, selectivity, safety, environmental impact.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

bio-alignment-io

Read, write, and convert multiple sequence alignment files using Biopython Bio.AlignIO. Supports Clustal, PHYLIP, Stockholm, FASTA, Nexus, and other alignment formats for phylogenetics and conservation analysis. Use when reading, writing, or converting alignment file formats.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

sleep-analyzer

分析睡眠数据、识别睡眠模式、评估睡眠质量，并提供个性化睡眠改善建议。支持与其他健康数据的关联分析。

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

metabolomics-workbench-database

Access NIH Metabolomics Workbench via REST API (4,200+ studies). Query metabolites, RefMet nomenclature, MS/NMR data, m/z searches, study metadata, for metabolomics and biomarker discovery.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

bio-hi-c-analysis-matrix-operations

Balance, normalize, and transform Hi-C contact matrices using cooler and cooltools. Apply iterative correction (ICE), compute expected values, and generate observed/expected matrices. Use when normalizing or transforming Hi-C matrices.

2,009 275
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results