proteinmpnn

Files (2)

• SKILL.md
• temperature-guide.md · references/

SKILL.md (first lines)

# ProteinMPNN Sequence Design

## Prerequisites

| Requirement | Minimum | Recommended |
|-------------|---------|-------------|
| Python | 3.8+ | 3.10 |
| CUDA | 11.0+ | 11.7+ |
| GPU VRAM | 8GB | 16GB (T4) |
| RAM | 8GB | 16GB |

## How to run

> **First time?** See [Installation Guide](../../docs/installation.md) to set up Modal and biomodals.

### Option 1: Local installation (recommended)
```bash
git clone https://github.com/dauparas/ProteinMPNN.git
cd ProteinMPNN

python protein_mpnn_run.py \
  --pdb_path backbone.pdb \
  --out_folder output/ \
  --num_seq_per_target 16 \
  --sampling_temp "0.1"
```

**GPU**: T4 (16GB) sufficient | **Time**: ~50-100 sequences/minute

### Option 2: Modal (via LigandMPNN wrapper)
```bash
cd biomodals
modal run modal_ligandmpnn.py \
  --pdb-path backbone.pdb \
  --num-seq-per-target 16
```

Note: LigandMPNN includes ProteinMPNN functionality.

## Config Schema

### Core Parameters

| Parameter | Default | Range | Description |
|-----------|---------|-------|-------------|
| `--pdb_path` | required | path | Single PDB input |
| `--pdb_path_chains` | all | A,B | Chains to design (comma-sep) |
| `--out_folder` | required | path | Output directory |
| `--num_seq_per_target` | 1 | 1-1000 | Sequences per structure |
| `--sampling_temp` | "0.1" | "0.0001-1.0" | Temperature (string!) |
| `--seed` | 0 | int | Random seed |
| `--batch_size` | 1 | 1-32 | Batch size |

### Temperature Guide
```
0.1  -> Low diversity, high recovery (production)
0.2  -> Moderate diversity (default)
0.3  -> Higher diversity (exploration)
0.5+ -> Very diverse, lower quality
```

**IMPORTANT**: Temperature must be passed as a string, not float.

## Common mistakes

### Temperature Parameter
✅ **Correct**:
```bash
--sampling_temp "0.1"    # String with quotes
```

❌ **Wrong**:
```bash
--sampling_temp 0.1      # Float without quotes - may cause errors
--sampling_temp 0.1,0.2  # Multiple temps need proper format
```

### Fixed Positions JSONL
✅ **Correct**:
```json
{"A": [1, 2, 3, 10, 11], "B": [5, 6]}
```

❌ **Wrong**:
```json
{"A": "1,2,3,10,11"}     # String instead of list
{A: [1, 2, 3]}           # Missing quotes on key
{"A": [1,2,3,]}          # Trailing comma
```

### Chain Selection
✅ **Correct**:
```bash
--pdb_path_chains A,B    # No spaces
```

❌ **Wrong**:
```bash
--pdb_path_chains A, B   # Space after comma
--pdb_path_chains "A,B"  # Quotes may cause issues
```

### Amino Acid Biases
```bash
# Bias toward certain AAs (positive = favor)
--bias_AA_jsonl '{"A": {"A": 1.5, "W": -2.0}}'

# Omit specific AAs globally
--omit_AAs "CM"  # No cysteine or methionine

# Per-position omission
--omit_AA_jsonl '{"A": {"1": "C", "2": "CM"}}'
```

### Multi-Chain Design
```bash
# Design chains A and B together
--pdb_path_chains A,B

From BioTender-max/awesome-bio-agent-skills, MIT.

More from BioTender-max/awesome-bio-agent-skills