LigandMPNN Inverse Folding

LigandMPNN structure-conditioned protein sequence design with ligand awareness

License: LigandMPNN is open source and free for academic and commercial use under an MIT license. Please refer to the license for full terms.

This generator is open source. Any third-party models, product names, or trademarks referenced are the property of their respective owners, and Proto is not affiliated with them.

---

Tools Used Tools Used Source Source Cite Cite

Go to Tool Page

proto-bio/proto-language/proto_language/generator/ligandmpnn_generator.py

View source

@ARTICLE{Dauparas2025-eg,
  title     = "Atomic context-conditioned protein sequence design using
               {LigandMPNN}",
  author    = "Dauparas, Justas and Lee, Gyu Rie and Pecoraro, Robert and An,
               Linna and Anishchenko, Ivan and Glasscock, Cameron and Baker,
               David",
  journal   = "Nat. Methods",
  publisher = "Springer Science and Business Media LLC",
  volume    =  22,
  number    =  4,
  pages     = "717--723",
  doi       = "10.1038/s41592-025-02626-1",
  month     =  apr,
  year      =  2025,
  language  = "en"
}

Copy citation

Protein sequence generator using LigandMPNN inverse folding model.

This generator uses LigandMPNN to design protein sequences that are predicted to fold into a given 3D backbone structure while considering ligand context. Unlike ProteinMPNN, LigandMPNN is aware of non-protein atoms (ligands, cofactors, metal ions) in the structure.

LigandMPNN is particularly effective for:

• Designing enzymes with specific active site geometries
• Optimizing binding pockets around ligands
• Creating sequences for cofactor-dependent proteins
• Redesigning protein-ligand interfaces

API Reference

ConfigLigandMPNNGeneratorConfig Source

Configuration object for LigandMPNNGenerator.

This class defines configuration parameters for the LigandMPNN generator, which uses the LigandMPNN inverse folding model to design protein sequences conditioned on a given 3D backbone structure and ligand context.

LigandMPNN extends ProteinMPNN to be aware of non-protein atoms (ligands, cofactors, metal ions), making it particularly effective for:

• Enzyme active site design
• Binding pocket optimization
• Cofactor-dependent protein design

structure_inputs · array

Structure(s) with optional chains_to_redesign and fixed_positions constraints.

temperature · number · default: "0.1"

Randomness of sampling (0-1). Near 0 is deterministic; near 1 is proportional to model probs.

excluded_amino_acids · array

Single-letter amino-acid codes to forbid in the designed sequence.

batch_size · integer · default: "1"

Number of sequences to process simultaneously on GPU

device · string · default: "cuda"

GPU device for inference (e.g. ‘cuda’ or ‘cuda:0’).

verbose · boolean · default: "False"

Whether to print status messages during execution.

Usage

python

>>> from proto_language.generator import LigandMPNNGenerator, LigandMPNNGeneratorConfig
>>> from proto_language.core import Segment
>>> config = LigandMPNNGeneratorConfig(
...     structure_inputs="/path/to/enzyme_with_ligand.pdb",
...     temperature=0.1,
... )
>>> gen = LigandMPNNGenerator(config)
>>> segment = Segment(length=100, sequence_type="protein")
>>> gen.assign(segment)
>>> gen.sample()  # Generates num_proposals sequences from the backbone

Metadata

Property	Value
Key	ligandmpnn
Class	LigandMPNNGenerator
Category	inverse_folding
Input Type	structure
Uses GPU	True
Supported Sequence Types	protein
Allows Empty Start	False