Alterlab Ligandmpnn logo

Alterlab Ligandmpnn

Community
AlterLab-IEU
alterlab-ligandmpnn

Design protein sequences around bound ligands, metals, and nucleic acids with LigandMPNN (Dauparas 2023) — inverse folding that conditions on non-protein context, so binding-pocket and metal-site residues are chosen to fit the actual ligand. Use when designing a small-molecule or metal binding pocket, redesigning residues that contact a ligand/ion/nucleic acid, or doing enzyme active-site design where the substrate matters. For backbone sequence design with NO ligand/metal context prefer alterlab-proteinmpnn; to GENERATE a backbone or scaffold a functional site prefer alterlab-rfdiffusion; to validate a design by refolding prefer alterlab-alphafold; to co-fold or dock the ligand prefer alterlab-boltz or alterlab-diffdock. Part of the AlterLab Academic Skills suite.

Overview

PublisherAlterLab-IEU
RepositoryAlterLab-Academic-Skills
Skill namealterlab-ligandmpnn
Stars
84
Forks
16
Bundled files
2
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by AlterLab-IEU on GitHub. Read the source before you install it.

Installation

Install the Alterlab Ligandmpnn AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/AlterLab-IEU/AlterLab-Academic-Skills/tree/main/skills/bioinformatics/alterlab-ligandmpnn
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/AlterLab-IEU/AlterLab-Academic-Skills.git /tmp/AlterLab-Academic-Skills
mkdir -p .claude/skills
cp -r /tmp/AlterLab-Academic-Skills/skills/bioinformatics/alterlab-ligandmpnn .claude/skills/alterlab-ligandmpnn
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Alterlab Ligandmpnn in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Alterlab Ligandmpnn on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Alterlab Ligandmpnn is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

LigandMPNN (ligand-aware sequence design)

Overview

LigandMPNN (Dauparas et al. 2023; dauparas/LigandMPNN) extends ProteinMPNN's inverse folding to condition on non-protein context — small-molecule ligands, metal ions, and nucleic acids. Because the model sees the ligand/metal atoms, the residues it designs for a binding pocket or metal site are chosen to complement what is actually bound, which plain ProteinMPNN (protein-atoms-only) cannot do.

Use it whenever the design target is a site that contacts a ligand or ion. For sequence design of a backbone with no bound context, use alterlab-proteinmpnn.

When to Use This Skill

Use this skill when the user wants to:

  • Design a small-molecule binding pocket so the residues fit the ligand.
  • Design a metal-coordinating site (e.g. Zn/Fe) with the ion in context.
  • Redesign residues that contact a ligand, ion, or nucleic acid.
  • Do enzyme active-site design where the substrate/cofactor should guide the choice.

Does NOT Trigger

ScenarioUse instead
Sequence design for a backbone with no ligand/metal contextalterlab-proteinmpnn
Generate a backbone or scaffold a functional motifalterlab-rfdiffusion
Validate a design by refoldingalterlab-alphafold
Co-fold the protein WITH the ligand from scratchalterlab-boltz
Dock a ligand into a fixed pocket (pose, not sequence)alterlab-diffdock

Core Capabilities

1. Ligand-aware pocket design

bash
# dauparas/LigandMPNN — clone, fetch weights once, then run run.py
git clone https://github.com/dauparas/LigandMPNN && cd LigandMPNN
bash get_model_params.sh "./model_params"

python run.py \
  --model_type ligand_mpnn \
  --checkpoint_ligand_mpnn "./model_params/ligandmpnn_v_32_010_25.pt" \
  --pdb_path complex_with_ligand.pdb \
  --out_folder out/ \
  --seed 111 \
  --batch_size 8 --number_of_batches 4

The input PDB must contain the ligand/metal atoms (HETATM). LigandMPNN designs pocket residues that fit that context; use --redesigned_residues/--fixed_residues to target only the site. Output FASTA headers carry overall_confidence and ligand_confidence (0–1, higher = more confident), averaged over the redesigned residues only.

2. Metal-site and nucleic-acid context

Provide the coordinating ion or the nucleic-acid chain in the structure so the model conditions on it — critical for metalloenzyme and DNA/RNA-binding designs.

3. Site-focused redesign

Restrict design to the residues within a shell of the ligand (redesign the pocket, keep the scaffold). Residues are named directly by chain+index — --redesigned_residues "A23 A24 B42D" or the complement, --fixed_residues "C1 C2 …". Insertion codes are preserved (B42D), because LigandMPNN parses structures with ProDy and keeps the original numbering rather than renumbering. Add --ligand_mpnn_use_side_chain_context 1 to condition on the side-chain atoms of the fixed residues as well as the backbone.

4. In the design pipeline

LigandMPNN provides the sequence step when the functional site involves a ligand: scaffold or generate the site with alterlab-rfdiffusion, design the pocket sequence here, then validate by refolding (alterlab-alphafold) and — if you need a pose/affinity — co-fold with alterlab-boltz or dock with alterlab-diffdock.

Resources

  • references/ligandmpnn_usage.md — install/pinning, model types, HETATM/context input, site-restricted design, and pipeline integration. Loaded on demand.

Part of the AlterLab Academic Skills suite.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Alterlab Ligandmpnn AI skill do?

Design protein sequences around bound ligands, metals, and nucleic acids with LigandMPNN (Dauparas 2023) — inverse folding that conditions on non-protein context, so binding-pocket and metal-site residues are chosen to fit the actual ligand. Use when designing a small-molecule or metal binding pocket, redesigning residues that contact a ligand/ion/nucleic acid, or doing enzyme active-site design where the substrate matters. For backbone sequence design with NO ligand/metal context prefer alterlab-proteinmpnn; to GENERATE a backbone or scaffold a functional site prefer alterlab-rfdiffusion;...

Why use Alterlab Ligandmpnn on TypingMind?

Because you install it once and use it with any model. Alterlab Ligandmpnn is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Alterlab Ligandmpnn in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/AlterLab-IEU/AlterLab-Academic-Skills/tree/main/skills/bioinformatics/alterlab-ligandmpnn. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Alterlab Ligandmpnn?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Alterlab Ligandmpnn?

As many as you like. As long as a model supports skills, you can use Alterlab Ligandmpnn with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Alterlab Ligandmpnn AI skill free?

Yes. It is published on GitHub by AlterLab-IEU under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇