OpenParable
Skip to content
← Back to skills

Molclaw Hdock Tool

Apps & Automation
molclaw-hdock-tool

Molclaw Hdock Tool is an Agent Skill for apps & automation published in the InternScience/Agents-A1 repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Run HDOCKlite docking for protein complexes and return run directories with ranked models.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/InternScience/Agents-A1/blob/HEAD/evaluation/Tools/molbench/skills/L1_tools/molclaw-hdock-tool/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/molclaw-hdock-tool, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

HDOCK Protein Docking

Usage

Note:

  • Local files are not directly accessible by the server. Please upload them to the server using drugsda-file-transfer before execution.
  • For PDB file inputs, it is recommended to preprocess them using drugsda-fix_pdb before execution.

2. HDOCK Protein Docking

The description of tool hdock_tool.

Run HDOCKlite protein docking and return the unique run directory, key files, and summary metrics for structure-based screening workflows.
Args:
    receptor (str): Receptor PDB file path.
    ligand (str): Ligand or partner PDB file path.
    nmax (int): Number of docking models to generate (default 100).
    no_complex (bool): Disable complex structure generation (default False).
    angle (int): Rotation sampling interval in degrees (default 15).
    rsite (str|None): Optional receptor binding-site residue file.
    lsite (str|None): Optional ligand binding-site residue file.
Return:
    status (str): success, partial_success, or error execution status.
    msg (str): Human-readable execution summary.
    output_dir (str): Unique run directory under tool_result/hdock_tool_result.
    receptor (str): Resolved receptor input path.
    ligand (str): Resolved ligand input path.
    nmax (int): Effective model count upper bound used.
    no_complex (bool): Effective no-complex flag used.
    angle (int): Effective angle parameter used.
    rsite (str|None): Effective receptor site file used.
    lsite (str|None): Effective ligand site file used.
    output_files (dict): Key generated file paths such as Hdock.out, topN.pdb, and best models.
    metrics (dict): Summary metrics including generated model count and best docking score when available.

Scoring Interpretation (HDOCK)

  • HDOCK Docking Score is a relative ranking score. Values are usually negative, and more negative means better predicted binding.
  • The score combines shape complementarity (FFT search), electrostatic interactions, and desolvation-like energy terms.
  • model_1.pdb is the top-ranked pose generated by createpl; model_2.pdb to model_10.pdb are sorted from better to worse by docking score.
  • Do not interpret HDOCK score as an absolute binding free energy in kcal/mol. Use it for within-run pose ranking and candidate prioritization.

How to use tool hdock_tool :

response = await client.session.call_tool(
    "hdock_tool",
    arguments={
        "receptor": "/path/to/receptor.pdb",
        "ligand": "/path/to/ligand.pdb",
        "nmax": 10,
        "angle": 15
    }
)
result = client.parse_result(response)
key_output = result["output_dir"]

Example parameter sets

# 1) Main mode: basic docking run (from README/test flow)
{
    "receptor": "/path/to/receptor.pdb",
    "ligand": "/path/to/ligand.pdb",
    "nmax": 10,
    "angle": 15,
    "no_complex": False
}

# 2) Variant mode: light sampling with complex generation disabled and defined sites
{
    "receptor": "/path/to/receptor.pdb",
    "ligand": "/path/to/ligand.pdb",
    "nmax": 5,
    "angle": 10,
    "no_complex": True,
    "rsite": "/path/to/rsite.txt",
    "lsite": "/path/to/lsite.txt"
}

# 3) Variant mode: exhaustive sampling with default complex outputs
{
    "receptor": "/path/to/receptor.pdb",
    "ligand": "/path/to/ligand.pdb",
    "nmax": 100,
    "angle": 15
}

⚠ Mandatory Output File Download (L3 Principle 14)

After calling this tool, you MUST download all output structure files from the MCP server to the local workspace using server_file_to_base64. A tool call is NOT considered complete until its output files have been downloaded and verified locally (ls -la <file> — size must be > 0).

import base64, os
response = await client.session.call_tool(
    "server_file_to_base64",
    arguments={"file_path": result["output_file"]}  # or relevant output field
)
dl = client.parse_result(response)
local_path = "stepNN_descriptive_name.ext"
with open(local_path, "wb") as f:
    f.write(base64.b64decode(dl["base64_string"]))
assert os.path.getsize(local_path) > 0, f"Download failed: {local_path}"

Download policy: All structure output files are Category A (user-critical) — essential for user verification, downstream analysis, and reproducibility. When in doubt, download. Over-collection is always preferred over under-collection.

Specific files to download from HDOCK output: All ranked docked complex PDB models (typically model_1.pdb through model_10.pdb). At minimum, download the top-ranked model. These are Category A files essential for downstream ProLIF protein-protein analysis.