Detect binding pockets in PDB or mmCIF protein structures with fpocket.
npx skills add https://github.com/InternScience/MolClaw --skill molclaw-fpocket
Please help me install this Agent Skill. Skill: molclaw-fpocket Source: https://github.com/InternScience/MolClaw/tree/main/skills/L1_tools/molclaw-fpocket Command: npx skills add https://github.com/InternScience/MolClaw --skill molclaw-fpocket
Detect and characterize binding pockets in protein structures to identify optimal sites for small-molecule docking and virtual screening, removing manual pocket selection and standardizing pocket descriptors.
Use molclaw-fpocket to detect and rank pockets from protein.pdb and return parsed pocket descriptors with a druggability filter of 0.2 and top_n set to 5.
High-intent search queries and answers about installing and using this skill.
To detect binding pockets in a protein structure for docking, you can use automated pocket detection tools that accept PDB or mmCIF inputs and output parsed pocket descriptors like scores, centers, volumes, and residue contacts.
Druggability filtering for binding pockets works by applying a numerical threshold to rank and narrow candidate sites, ensuring only pockets meeting a specific druggability score are exported for downstream virtual screening.
Yes, you can use mmCIF inputs for protein pocket detection alongside standard PDB files, allowing you to identify small-molecule binding sites and extract their coordinates for virtual screening workflows.
The minimum docking box size needed when preparing protein pockets is 25.0 Å per axis, which is enforced to prevent undersized search volumes during small-molecule docking and virtual screening.
To rank and select top protein binding pockets for hit-site prioritization, apply top-N ranking and druggability thresholding utilities to filter detected sites and export the best pocket coordinates and metadata.
When detecting binding pockets for structural biology workflows, the output includes parsed, structured pocket descriptors such as druggability scores, pocket centers, volumes, residue contacts, and atom-level counts.
InternSciencePredicts ligand binding pockets in PDB proteins with confidence scores and docking boxes.
InternSciencePerforms GPU-accelerated molecular docking and generates pose files with affinity values.
ANakaVisualize protein-ligand binding sites with PyMOL and generate publication-ready figures.
yf8578Run molecular dynamics simulations with OpenMM and analyze trajectories using MDAnalysis.
