drug-pose-validation

Validate ligand poses with PoseBusters and output JSON plus filtered SDF.

144|21|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill drug-pose-validation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: drug-pose-validation
Source: https://github.com/learningmatter-mit/AtomisticSkills/tree/main/.agents/skills/drug-pose-validation
Command: npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill drug-pose-validation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires posebusters, rdkit, and includes scripts (resource) components.

What problem does it solve?

It eliminates wasted compute on chemically invalid or sterically impossible ligand poses by validating pose geometry and clashes before downstream refinement.

Core Features & Use Cases

  • PoseBusters-based plausibility checks: bond lengths/angles, planarity/flatness, stereochemistry, and internal ligand clashes using PoseBusters.
  • Optional receptor-aware clash screening: when a receptor PDB is provided, it also filters protein–ligand steric clashes using PoseBusters in dock mode.
  • Deterministic outputs for triage: generates a per-pose JSON validation report, a human-readable summary, and an SDF containing only valid poses.

Quick Start

Run pose validation on your docked ligand poses by converting them to SDF if needed and then executing the validate_poses.py script with the poses SDF, an optional receptor PDB, and an output directory.

Frequently Asked Questions about drug-pose-validation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I filter docked ligand poses for steric clashes and invalid stereochemistry?

PoseBusters validates docked ligand poses for steric clashes and invalid stereochemistry by checking bond lengths, planarity, and internal clashes. It outputs a JSON report and an SDF file containing only the physically plausible poses passing these configured tests.

What is physical plausibility validation in drug discovery docking workflows?

Physical plausibility validation in drug discovery prevents wasted compute on chemically invalid docked ligand poses. It checks generated docking geometries for internal clashes and invalid stereochemistry before passing them to expensive molecular simulation or refinement stages.

Do I need a receptor PDB file to run pose validation on my SDF inputs?

You do not need a receptor PDB to run pose validation, as SDF inputs can be screened for ligand-only plausibility. Providing a receptor PDB optionally enables receptor-aware protein-ligand steric clash filtering using PoseBusters in dock mode.

How do I use PoseBusters to triage ligand poses before molecular simulation?

To use PoseBusters to triage ligand poses before molecular simulation, execute the validation script with your poses SDF and optional receptor PDB. This generates a per-pose JSON validation report and an SDF filtered to only valid poses for downstream analysis.

Can I validate generated ligand poses without receptor context?

Yes, you can validate generated ligand poses without receptor context by applying ligand-only pose screening. This mode checks internal ligand clashes, bond lengths, angles, and stereochemistry using PoseBusters to ensure physical plausibility before any refinement steps.