---
name: diffdock-molecular-docking
description: Diffusion-based molecular docking to predict 3D ligand–protein binding poses (blind docking) with confidence scoring; use when you need pose prediction for drug discovery or virtual screening.
license: MIT
author: AIPOCH
---
> **Source**: [https://github.com/aipoch/medical-research-skills](https://github.com/aipoch/medical-research-skills)

# DiffDock Molecular Docking

## When to Use

- **Blind docking** when you have a protein structure (PDB) and a ligand (SMILES) but no known binding site.
- **Pose prediction** to generate multiple plausible 3D binding conformations and rank them.
- **Virtual screening support** to quickly evaluate candidate ligands by predicted binding poses and confidence.
- **Drug discovery workflows** where you need automated docking outputs (SDF poses + scores) for downstream analysis.
- **Batch/advanced docking** when running many ligand–protein pairs or using alternative inputs (e.g., sequence-based workflows; see `references/workflows_examples.md`).

## Key Features

- **Diffusion generative sampling** to produce diverse ligand binding poses.
- **Confidence model scoring** to rank predicted poses.
- **Simple CLI inference** for single protein–ligand docking.
- **Batch/advanced workflows** documented in `references/workflows_examples.md`.
- **Structured outputs** including ranked SDF pose files and a confidence score report.

## Dependencies

- Python (version not specified)
- PyTorch (version not specified)
- PyTorch Geometric / PyG (version not specified)
- RDKit (version not specified)
- ESM (version not specified)

## Example Usage

### 1) Verify the Environment

```bash
python scripts/setup_check.py
```

### 2) Run Standard Inference (Single Docking)

Dock a single ligand (SMILES) to a protein structure (PDB) and write results to an output directory:

```bash
python scripts/inference_runner.py \
  --protein ./data/protein.pdb \
  --ligand "CC(=O)Oc1ccccc1C(=O)O" \
  --out_dir ./results
```

**Arguments**
- `--protein`: Path to the protein PDB file.
- `--ligand`: Ligand SMILES string.
- `--out_dir`: Output directory (default: `results/`).

### 3) Outputs

After inference, the tool produces:

- **Ranked SDF pose files** (e.g., `rank1.sdf`, `rank2.sdf`, ...), each containing a predicted 3D binding pose.
- **Confidence score report**: `confidence_scores.txt`, listing the score for each ranked pose.

## Implementation Details

- **Pose generation**: Uses a diffusion-based generative model to sample multiple candidate ligand poses relative to the protein target.
- **Ranking**: A separate confidence model assigns a score to each sampled pose; poses are sorted by this score and saved as `rank*.sdf`.
- **Parameterization**:
  - For the complete CLI argument list and defaults, see `references/parameters_reference.md`.
  - For confidence interpretation, known limitations, and expected accuracy/scope, see `references/confidence_and_limitations.md`.
- **Advanced workflows**: Batch processing and alternative input configurations are documented in `references/workflows_examples.md`.

---
Source attribution: aipoch
Original project: https://github.com/aipoch/medical-research-skills/blob/686e09dfdb45d9fa1a08664f06bb93769b3c9d91/scientific-skills/Evidence%20Insight/diffdock-molecular-docking/SKILL.md
License: MIT
License information: https://opensource.org/license/mit
Original upstream instructions; downloaded from Mega AI Hub.

## Upstream license notice

MIT License

Copyright (c) 2026 AIpoch

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.



# Mega AI Hub resource metadata

Creator / publisher: aipoch
Original project: https://github.com/aipoch/medical-research-skills/blob/686e09dfdb45d9fa1a08664f06bb93769b3c9d91/scientific-skills/Evidence%20Insight/diffdock-molecular-docking/SKILL.md
Attribution basis: Source repository record

Fields of work: Education & Research, Healthcare & Life Sciences
Roles: Clinical Research, Medical Writer, Researcher
Specialties: API
Tools: Python, Pytorch, Rdkit
Country applicability: Country not specified
Classification basis: Source catalog labels
Instruction SHA-256: 2dfdcc3812f63a2c75785c11a3f8493e18ed9e559d3870b14f1af4e8a4dbd691