Binder Design Tool Selection Agent Skill

Binder Design Tool Selection

Plain-language role: Use this skill to choose the right binder-design-tool-selection method, not to run the design model itself.

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)   (chai1-structure-prediction)     (qc)

1. Target preparation

# 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):

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:

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:

# 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

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-design-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

Templates and Demo

• Planning template: templates/binder-design-tool-selection/target-brief.md
• Minimal walkthrough: examples/minimal-binder-campaign/README.md
• Example filled brief: examples/minimal-binder-campaign/target-brief.md

Inputs

• A design objective such as de novo binder generation, ligand binding, or nanobody optimization.
• Target context including structure availability, hotspot knowledge, and diversity requirements.
• Compute and timeline constraints that affect tool choice.

Outputs

• A recommended tool choice or tool combination for the target and campaign goal.
• A staged workflow covering target preparation, generation, validation, and filtering.
• Suggested handoffs into skills such as pdb, boltzgen, rfdiffusion, chai1-structure-prediction, and protein-design-qc.

Next Step

Use pdb to prepare the target, then execute the chosen design path with boltzgen, bindcraft, or rfdiffusion.