diffdock

Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.

1.54x

Quality

66%

Does it follow best practices?

Impact

94%

1.54x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/diffdock/SKILL.md

Evaluation results

88%

46%

Virtual Screening Pipeline for Kinase Inhibitor Discovery

Batch virtual screening pipeline

Criteria

Without context

With context

Batch CSV validation script

100%

Correct CSV column names

100%

Results analysis script usage

100%

Export to CSV

100%

ESM embedding precomputation

Confidence is not affinity

100%

Confidence score thresholds

80%

100%

Downstream scoring recommendation

100%

Canonical SMILES in CSV

100%

Batch size parameter

100%

40%

Optimizing DiffDock for a Challenging Macrocyclic Inhibitor

Custom configuration for challenging docking

Criteria

Without context

With context

Config from template

50%

100%

Increased torsion temperature

100%

Increased samples per complex

100%

Increased inference steps

100%

Ensemble CSV for both conformations

53%

100%

Uses --protein_ligand_csv flag

87%

100%

Torsion temperature rationale

100%

Samples rationale

100%

No affinity claims

100%

96%

13%

Post-Docking Analysis and Candidate Prioritization Report

Results interpretation and affinity pipeline

Criteria

Without context

With context

Uses analyze_results.py

100%

Three-tier classification

100%

Correct High confidence compounds

100%

Correct Low confidence compounds

100%

Confidence is not potency

100%

Large peptide flag

100%

Large ligand caveat

100%

Downstream affinity tool

100%

Review multiple top poses

37%

50%

Export to CSV

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: b58ad7e

Evaluated: 27 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Virtual Screening Pipeline for Kinase Inhibitor Discovery Optimizing DiffDock for a Challenging Macrocyclic Inhibitor Post-Docking Analysis and Candidate Prioritization Report

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.