skills.rest
InternScience
InternScienceOFFICIAL
Explore all skills by this author →
GitHub

molclaw-boltz2-affinity

Predict protein-ligand binding affinity from a protein sequence and SMILES using Boltz-2.

28|2|Updated Mar 31, 2026
Search Tags:#drug-discovery#protein-ligand#chemoinformatics#boltz2#binding-affinity#lead-optimization#molecular-prediction
One-click install
npx skills add https://github.com/InternScience/MolClaw --skill molclaw-boltz2-affinity
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: molclaw-boltz2-affinity
Source: https://github.com/InternScience/MolClaw/tree/main/skills/L1_tools/molclaw-boltz2-affinity
Command: npx skills add https://github.com/InternScience/MolClaw --skill molclaw-boltz2-affinity

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Predicts the binding affinity between a target protein sequence and a small molecule SMILES using Boltz-2, enabling rapid in silico evaluation during early drug discovery.

Core Features & Use Cases

  • Predicts binding probability (binary) and the affinity value (log10 IC50) for protein–ligand pairs using Boltz-2.
  • Supports end-to-end workflows: retrieve the target protein sequence if needed, run Boltz-2 affinity prediction, and obtain outputs including affinity_probability_binary, affinity_pred_value, and complex_cif_file.
  • Useful in hit discovery and lead optimization pipelines to rank ligands by predicted binding strength.

Quick Start

Provide a target protein sequence (or retrieve it with the integrated skill) and a SMILES string for the ligand, then run the Boltz-2 affinity predictor to obtain the affinity metrics.

Frequently Asked Questions about molclaw-boltz2-affinity

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

FAQPage Schema
How do I predict protein-ligand binding affinity using a SMILES string and protein sequence?▼

To predict protein-ligand binding affinity, provide a target protein sequence and a small molecule SMILES string. The tool calculates the binding probability and predicted log10 IC50 value using the Boltz-2 model for rapid in silico evaluation.

What outputs does Boltz-2 generate for protein-ligand binding prediction?▼

Boltz-2 protein-ligand binding prediction outputs include affinity_probability_binary, affinity_pred_value, and a complex_cif_file. The CIF file containing the predicted complex structure is downloaded as a mandatory post-step.

Can I retrieve a target protein sequence directly within a binding affinity prediction workflow?▼

Yes, you can retrieve a target protein sequence directly within the workflow. The prediction process supports an integrated retrieval step before running the Boltz-2 affinity predictor on your small molecule SMILES.

When should I use in silico binding affinity prediction in drug discovery?▼

Use in silico binding affinity prediction during hit discovery and lead optimization pipelines. It enables rapid evaluation and ranking of small molecule ligand candidates by their predicted binding strength to the target protein.

What format is required for small molecule ligands in Boltz-2 affinity prediction?▼

Small molecule ligands must be provided as a SMILES string for Boltz-2 affinity prediction. The target protein counterpart must be supplied as a standard protein sequence to execute the prediction successfully.

Related Skills

Explore More Data & Analytics Skills→
PKU-YuanGroupPKU-YuanGroup
288

boltz

Predict protein, nucleic-acid, and small-molecule complex structures with Boltz-2.

Official
Intermediate
ANakaANaka
3

binding-site-visualization

Visualize protein-ligand binding sites with PyMOL and generate publication-ready figures.

Community
Advanced
Zeyuyang-0420Zeyuyang-0420

diffdock

Predict 3D binding poses of small molecule ligands to protein targets.

Community
Advanced
ovachieverovachiever
52

diffdock

Predict protein-ligand docking poses and confidence scores from PDB structures or sequences.

Community
Advanced