---
tags:
  - Biology
  - Protein
  - Protein Design
license: agpl-3.0
library_name: ofoldx
pipeline_tag: other
---

# PXDesign

Pre-trained model for protein backbone generation and binder evaluation through Preview and Extended workflows.

## Disclaimer

This is an UNOFFICIAL implementation of [PXDesign](https://github.com/bytedance/PXDesign) by ByteDance Research.

The OFFICIAL repository of PXDesign is at [bytedance/PXDesign](https://github.com/bytedance/PXDesign).

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

## Model Details

PXDesign-d is an all-atom diffusion model for de novo and target-conditioned protein backbone generation. The complete
workflow assigns a ProteinMPNN sequence to each backbone, evaluates the complex and unbound binder with AlphaFold 2,
and filters and ranks the candidates.

Extended evaluation additionally uses Protenix base or the structure-conditioned Protenix mini model, following the
released [PXDesignBench](https://github.com/bytedance/PXDesignBench) composition. Please refer to the
[Training Details](#training-details) section for more information on the training process.

### Variants

- **PXDesign backbone**: Independently callable PXDesign-d coordinate generation through `PxDesignBackbonePipeline`.
- **PXDesign Preview**: The complete workflow with 100 backbones by default, ProteinMPNN assignment, AlphaFold 2 evaluation and released selection rules.
- **PXDesign Extended**: The complete workflow with 500 backbones by default, full-target Protenix routing, two-step screening and 20-step reruns for candidates passing basic screening.

### Model Specification

| Single Size | Pair Size | Token Denoiser Blocks | Atom Encoder Blocks | Atom Decoder Blocks | Num Parameters (M) |
| ----------- | --------- | --------------------- | ------------------- | ------------------- | ------------------ |
| 384         | 128       | 16                    | 4                   | 4                   | 139.028            |

The parameter count is for PXDesign-d.

### Links

- **Code**: [ofoldx.models.pxdesign](https://github.com/OTeam-AI4S/OFoldX/tree/main/ofoldx/models/pxdesign)
- **Weights**: [oteam/pxdesign](https://huggingface.co/oteam/pxdesign)
- **Developed by**: ByteDance Research
- **Model type**: Target-conditioned all-atom protein backbone diffusion model
- **Original Repository**: [bytedance/PXDesign](https://github.com/bytedance/PXDesign)

## Usage

The model file depends on the [`ofoldx`](https://ofoldx.valtech.ai/) library. You can install it using pip:

```bash
pip install ofoldx
```

### Direct Use

#### Preview Design

You can use this model directly with a pipeline for 100-residue binder design with Preview evaluation:

```python
import ofoldx
from ofoldx.io import read_structure
from ofoldx.models.pxdesign import from_binder

structure, _ = read_structure("target.cif")
pipeline = ofoldx.pipeline("design", model="oteam/pxdesign")
result = pipeline(structure, placement=from_binder(100, [("target", [(0, 150)])]), seed=7)
```

The result retains all candidates and ranked selection views. `from_binder` names the target instances and zero-based
component spans. The complete repository contains six child pipelines; Preview uses the backbone generator,
ProteinMPNN and two AlphaFold 2 evaluations.

#### Extended Design

You can use this model directly with a pipeline for binder design with Extended evaluation and full-target MSA evidence:

```python
import ofoldx
from ofoldx.io import read_structure
from ofoldx.models.pxdesign import from_binder
from ofoldx.msa import read_msa

structure, _ = read_structure("target.cif")
msa = {"target": read_msa("target.a3m", entity_id="1")}
pipeline = ofoldx.pipeline("design", model="oteam/pxdesign")
result = pipeline(
    structure,
    placement=from_binder(100, [("target", [(0, 150)])]),
    msa=msa,
    mode="extended",
    seed=7,
)
best = result.ranked_candidates[0]
```

Here `target` is the target instance ID and `1` is its entity ID; use the IDs in your input structure.

Extended additionally uses both Protenix children and reruns every candidate passing basic screening, including strict
passes, for 20 diffusion transitions. Returned structures come from those reruns.

Both modes retain admitted target heavy atoms, including Se/SE, and preserve MSE rather than substituting MET as in the
upstream input convention. This can change the checkpoint's feature distribution; observation masks control usable
geometry without changing chemical membership.

### Downstream Use

#### Backbone Generation

The `backbone` component is independently callable through `PxDesignBackbonePipeline` and returns structures in
generation order without sequence assignment or evaluation. The [pipeline guide](https://github.com/OTeam-AI4S/OFoldX/blob/main/ofoldx/pipelines/README.md) describes
component loading and staged calls; [`PxDesignPipeline`][ofoldx.models.pxdesign.PxDesignPipeline] documents the six
component roles and the chemical matching used in evaluation.

## Training Details

PXDesign-d learns protein backbone generation with an EDM coordinate-denoising objective.

### Training Data

PXDesign-d is trained on [PDB](https://www.rcsb.org/) structures up to 1 May 2021 and predicted structures from
[AlphaFold DB](https://alphafold.ebi.ac.uk/) and [MGnify](https://www.ebi.ac.uk/metagenomics/).

### Training Procedure

PXDesign-d is trained in two stages, first emphasizing monomer structures and then target-conditioned complexes.

OFoldX exposes the differentiable one-step denoiser for training and fine-tuning; it does not reproduce PXDesign-d's
full training pipeline.

## Citation

Please cite the PXDesign work and repository described by the [upstream project](https://github.com/bytedance/PXDesign).

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

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

## Contact

Please use GitHub issues of [OFoldX](https://github.com/OTeam-AI4S/OFoldX/issues) for any questions or comments on the model implementation or model card.

Please contact the authors of the [PXDesign project](https://github.com/bytedance/PXDesign) 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](https://github.com/OTeam-AI4S/OFoldX/blob/main/license.md).

For additional terms and clarifications, please refer to our
[License FAQ](https://github.com/OTeam-AI4S/OFoldX/blob/main/license-faq.md).

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

[ofoldx.models.pxdesign.PxDesignPipeline]: https://ofoldx.valtech.ai/models/pxdesign/#ofoldx.models.pxdesign.PxDesignPipeline