一键导入
molclaw-fpocket-toolkit-base
Detect binding pockets with fpocket_toolkit and return parsed pocket descriptors and run artifacts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Detect binding pockets with fpocket_toolkit and return parsed pocket descriptors and run artifacts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | molclaw-fpocket-toolkit-base |
| description | Detect binding pockets with fpocket_toolkit and return parsed pocket descriptors and run artifacts. |
| license | MIT license |
| metadata | {"skill-author":"PJLab"} |
Note:
drugsda-file-transfer before execution.drugsda-fix_pdb before execution.The description of tool fpocket_toolkit.
Detect pockets with fpocket_toolkit and store every run under the tool_result/fpocket_result directory.
Args:
pdb_file (str): Input PDB/mmCIF file to scan (required).
top_n (int): Limit returned pockets to the top N (0 = all). Default: 0.
min_druggability (float|None): Filter out pockets below this threshold. Default: None.
verbose (bool): Enable verbose descriptor parsing during the run. Default: False.
Return:
status (str): 'success' or 'error'.
msg (str): Human-readable narrative about the run.
run_dir (str): Absolute directory storing this run.
output_dir (str): Path where fpocket preserved its outputs.
pockets (List[Dict[str, Any]]): Parsed pocket descriptors, including scores and centers.
pocket_count (int): Number of pockets returned.
output_files (Dict[str, str]): Preserved fpocket output files.
exported (Dict[str, str]|None): Export metadata when requested.
files (Dict[str, str]): All files created under the run_dir.
How to use tool fpocket_toolkit :
response = await client.session.call_tool(
"fpocket_toolkit",
arguments={
"pdb_file": "/path/to/input.pdb",
"top_n": 0,
"min_druggability": None,
"verbose": False,
}
)
result = client.parse_result(response)
pockets = result.get("pockets")
# 1) Main mode: analyze full structure and return all pockets (README example)
{
"pdb_file": "/path/to/input.pdb",
"top_n": 0,
"min_druggability": None,
"verbose": False,
}
# 2) Variant mode: return top 3 pockets above a druggability threshold
{
"pdb_file": "/path/to/input.pdb",
"top_n": 3,
"min_druggability": 0.5,
"verbose": True,
}
When fpocket pocket dimensions are used for downstream docking (QuickVina, etc.), enforce a minimum of 25.0 Å per dimension. If any pocket dimension returned by fpocket is less than 25 Å, override it to 25.0 Å before passing to the docking tool.
# After fpocket pocket detection:
size_x = max(25.0, pocket["size_x"])
size_y = max(25.0, pocket["size_y"])
size_z = max(25.0, pocket["size_z"])
Predict the ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties of the input molecules.
Predict binding affinity between target protein sequence and small molecule SMILES using Boltz-2.
Predict protein structures with Chai-1 from sequence or FASTA input and return model scoring summaries.
Chroma toolkit skill covering chroma_monomer for single-chain generation, chroma_complex for multi-chain assembly generation, and chroma_symmetry for symmetry-constrained protein design.
Retrieve SMILES strings from PubChem database using compound names.
Generate entirely new drug-like molecules from scratch without any starting molecule, using REINVENT4's de novo prior.