LigandMPNN (Claude Skill)

Designs protein sequences for a fixed backbone while accounting for bound ligands, metals, and nucleic acids using LigandMPNN, driven as a Claude skill.

   
Type Claude Skill
Supplier Baker Lab, UW Institute for Protein Design
Availability GA — Claude Science research skill
Pricing Free / OSS (MIT)
Capabilities Read/Write — local inference; writes designed sequences (FASTA) and scores
Verified works · 2026-07-20
Security cleared · 2026-07-20 — provenance matches Baker Lab, MIT, featured Claude Science skill, local inference no credentials

How to install

  • Claude Science — enable the built-in LigandMPNN research skill (Anthropic-hosted; not published to the public anthropics/life-sciences marketplace).
  • Run the model yourself — the upstream model is open source:
    git clone https://github.com/dauparas/LigandMPNN
    

    Follow the repo README for environment setup and model weights.

What it does

Runs LigandMPNN to design amino-acid sequences for a target backbone (PDB), conditioning on non-protein context (ligands, metals, nucleotides). Local inference only.

Primary use cases: Ligand-binding-site redesign, enzyme active-site design, metalloprotein sequence design

Notes

Claude Science: Featured as a research skill in Anthropic’s Claude Science. Its inclusion there is an independent signal of quality and trustworthiness for life-science work.

Local (your compute). Same family as ProteinMPNN with explicit non-protein atom context.

Sources


Installed this tool?

Share feedback — install path, OS, errors, workarounds. The form opens with this tool pre-selected and a link back to this page.