oteam/odesign-protein-rigid

ODesign

Pre-trained model for biomolecular interaction design through coordinate diffusion followed by sequence or element assignment.

Disclaimer

This is an UNOFFICIAL implementation of ODesign: A World Model for Biomolecular Interaction Design by Odin Zhang, Xujun Zhang, Haitao Lin, et al.

The OFFICIAL repository of ODesign is at The-Institute-for-AI-Molecular-Design/ODesign.

The team releasing ODesign did not write this model card for this model so this model card has been written by the OFoldX team.

Model Details

ODesign uses an all-atom diffusion model for biomolecular interaction design. Its variants support different designed molecules and either rigid or flexible receptor context. Inverse-folding models assign sequences or elements to the generated structures.

The protein routes use ProteinMPNN without ligand atom context and LigandMPNN otherwise. The released DNA route uses OInvFold, the RNA route uses gRNAde, and the provisional ligand route uses ligand OInvFold. Please refer to the Training Details section for more information on the training process.

Variants

• oteam/odesign-protein-flex: Protein backbone generation with flexible receptor context and MPNN sequence assignment.
• oteam/odesign-protein-rigid: Protein backbone generation with rigid receptor context and the same context-selected MPNN assignment.
• oteam/odesign-dna-rigid: DNA generation with rigid receptor context and OInvFold sequence assignment.
• oteam/odesign-rna-rigid: RNA generation with rigid receptor context and gRNAde sequence assignment.
• oteam/odesign-ligand-rigid: Provisional ligand coordinate generation with rigid context and element assignment.

Model Specification

Single Size	Pair Size	Pairformer Blocks	MSA Stages	Token Denoiser Blocks	Atom Encoder Blocks	Atom Decoder Blocks	Num Parameters (M)
384	128	48	4	24	3	3	355.687

The parameter count is for the coordinate-generation model.

Links

• Code: ofoldx.models.odesign
• Paper: ODesign: A World Model for Biomolecular Interaction Design
• Developed by: Odin Zhang, Xujun Zhang, Haitao Lin, et al.
• Model type: All-atom coordinate diffusion with sequence or element assignment
• Original Repository: The-Institute-for-AI-Molecular-Design/ODesign

Usage

The model file depends on the ofoldx library. You can install it using pip:

pip install ofoldx

Direct Use

Polymer Design

You can use this model directly with a pipeline for polymer design over instance B of a conditioning complex:

import ofoldx
from ofoldx.io import read_structure
from ofoldx.models.design import PositionSelection

structure, _ = read_structure("complex.cif")
pipeline = ofoldx.pipeline("design", model="oteam/odesign-protein-flex")
result = pipeline(structure, design=PositionSelection(instances=("B",)), num_backbones=2, num_sequences=4, seed=7)

Protein, DNA and RNA candidates contain canonical Structure predictions and the inverse folder's named scores. Without a design selection, every position of the route's monomer type is generated. For de novo generation, supply a typed PolymerSequence of unknown residues; the chemistry guide describes these inputs.

Rigid context remains fixed, while flexible context can move.

Provisional Ligand Design

You can use this model directly with a pipeline for provisional ligand design with 8–16 atom slots around an observed pocket:

import ofoldx
from ofoldx.io import read_structure

structure, _ = read_structure("pocket.cif")
pipeline = ofoldx.pipeline("design", model="oteam/odesign-ligand-rigid")
result = pipeline(structure, generated_atoms=range(8, 17), seed=7)

Ligand candidates contain ordered coordinates, predicted elements, per-atom probabilities and pair bond logits. Their bonds, formal charges and chemical-component identities remain undetermined; they are provisional designs.

Downstream Use

Alternative Inverse Folding

ODesignOInvFoldProteinPipeline and ODesignOInvFoldRnaPipeline combine ODesign generation with OInvFold for protein and RNA sequence design. The backbone component can also be used independently for coordinate generation. See the pipeline guide for component loading and staged execution.

Training Details

ODesign learns coordinate denoising together with geometric and chemical reconstruction objectives.

Training Data

ODesign was trained on PDB crystal structures released before 30 September 2021 and distilled structures predicted by AlphaFold 2 and OpenFold. The official repository provides the training data.

Training Procedure

ODesign learns to denoise structures with coordinate, bond and distance-based objectives.

Citation

@misc{zhang2025odesign,
  title={ODesign: A World Model for Biomolecular Interaction Design},
  author={Zhang, Odin and Zhang, Xujun and Lin, Haitao and others},
  year={2025},
  eprint={2510.22304},
  archivePrefix={arXiv},
  primaryClass={q-bio.BM}
}

The artifacts distributed in this repository are part of the OFoldX project. If OFoldX supports your research, please cite the OFoldX project as follows:

@software{chen2026ofoldx,
  author = {Chen, Zhiyuan},
  title  = {OFoldX},
  year   = {2026},
  url    = {https://github.com/OTeam-AI4S/OFoldX},
}

Contact

Please use GitHub issues of OFoldX for any questions or comments on the model implementation or model card.

Please contact the authors of the ODesign paper for questions or comments on the paper/model.

License

The model weights, implementation, configuration, and documentation distributed by OFoldX are licensed under the GNU Affero General Public License.

For additional terms and clarifications, please refer to our License FAQ.

SPDX-License-Identifier: AGPL-3.0-or-later