--- library_name: "ofoldx" tags: - "biology" - "biomolecular-design" - "protein" - "rna" - "dna" - "pipeline" - "proteinmpnn" - "design-generation" - "protein-design" artifact_kind: "pipeline" repo_id: "oteam/proteinmpnn-noise002" license: "mit" pipeline_tag: "other" task: "design_generation" model-index: - name: "proteinmpnn-noise002" results: [] widget: - pipeline_tag: "other" task: "design_generation" example_title: "Backbone sequence design" text: "input_structure: backbone.cif\ndesign_chains: A" input_format: "structure_path" - pipeline_tag: "other" task: "design_generation" example_title: "Binder design" text: "target_structure: target.cif\ntarget_chains: A\ndesign_chains: B" input_format: "structure_path" --- # proteinmpnn-noise002 OFoldX `pipeline` artifact for biomolecular design generation, using the `proteinmpnn` architecture. ## Disclaimer This model card was generated by the OFoldX team for an OFoldX `pipeline` artifact. The upstream model authors did not write this card unless explicitly stated otherwise. OFoldX is pre-alpha research software. Check the source checkpoint, upstream release, and local validation before using the artifact for scientific or operational decisions. ## Model Details ProteinMPNN sequence-design model for protein backbones, including soluble and membrane variants. Converted ProteinMPNN-family sequence-design checkpoint for fixed-backbone inverse folding. ### Model Provenance - **Upstream Project**: ProteinMPNN / LigandMPNN release - **Source Release**: [https://github.com/dauparas/LigandMPNN](https://github.com/dauparas/LigandMPNN) - **Primary Paper**: [Robust deep learning-based protein sequence design using ProteinMPNN](https://doi.org/10.1126/science.add2187) - **Upstream License**: MIT for upstream ProteinMPNN and LigandMPNN code/model parameters ### Model Specification | Field | Value | | ----- | ----- | | Repository | `oteam/proteinmpnn-noise002` | | Artifact Kind | `pipeline` | | Task | `design_generation` | | Architecture | `proteinmpnn` | | Entrypoint | `ofoldx.pipelines.design.DesignPipeline` | > [!NOTE] > Checkpoint metadata: `k_neighbors=48`; the `noiseXXX` suffix identifies the training-noise variant. ### Links - **Hub repository**: [oteam/proteinmpnn-noise002](https://huggingface.co/oteam/proteinmpnn-noise002) - **Upstream paper**: [Robust deep learning-based protein sequence design using ProteinMPNN](https://doi.org/10.1126/science.add2187) - **Upstream repository**: [ProteinMPNN / LigandMPNN release](https://github.com/dauparas/ProteinMPNN) - **Source checkpoint release**: [https://github.com/dauparas/LigandMPNN](https://github.com/dauparas/LigandMPNN) - **Code**: [`ofoldx/pipelines/design.py`](https://github.com/OTeam-AI4S/OFoldX/tree/main/ofoldx/pipelines/design.py) - **Project repository**: [https://github.com/OTeam-AI4S/OFoldX](https://github.com/OTeam-AI4S/OFoldX) - **Issues**: [https://github.com/OTeam-AI4S/OFoldX/issues](https://github.com/OTeam-AI4S/OFoldX/issues) ## Usage The artifact depends on the [`ofoldx`](https://github.com/OTeam-AI4S/OFoldX) library. Install it with pip: ```bash pip install ofoldx ``` ### Pipeline Usage Load the artifact from `oteam/proteinmpnn-noise002` with the OFoldX task pipeline. Use `AutoModel` or `AutoProcessor` only when you need lower-level control: ```python from ofoldx.pipelines import Pipeline pipeline = Pipeline.from_pretrained("oteam/proteinmpnn-noise002") ``` When a matching processor is available, load it with `AutoProcessor.from_pretrained(...)` and pass the processed batch to the model. ### Interface - **Task**: `design_generation` - **Artifact kind**: `pipeline` - **Architecture**: `proteinmpnn` - **Runtime files**: `manifest.json`, `config.json`, and `model.safetensors` when present ## Training Details OFoldX did not train these weights. This repository contains a converted checkpoint and OFoldX runtime metadata for loading it. ### Training Data ProteinMPNN was trained for fixed-backbone protein sequence design on PDB biounits with a 2021-08-02 data snapshot. OFoldX does not redistribute the training set. ### Training Procedure Upstream ProteinMPNN trains graph-neural inverse-folding models with 48-neighbor checkpoints and multiple Gaussian backbone-noise variants. OFoldX converts released ProteinMPNN-family checkpoints into `model.safetensors`; it does not run ProteinMPNN training. ## Evaluation OFoldX conversion reports and contract tests validate artifact structure and checkpoint loading. Task-level scientific evaluation should be checked against the corresponding upstream model release or paper. ## Limitations - This artifact is distributed for research use. - Inputs must match the model-specific processor and expected biomolecular representation. - OFoldX is pre-alpha, so APIs and artifact metadata may still change before a stable release. ## Citation Please cite the upstream ProteinMPNN / LigandMPNN release work for the source checkpoint. If OFoldX supports your work, please also cite or link the OFoldX project repository. ```bibtex @article{dauparas2022robust, author = {Dauparas, Justas and Anishchenko, Ivan and Bennett, Nathaniel and Bai, Hua and Ragotte, Robert J. and Milles, Lukas F. and Wicky, Basile I. M. and Courbet, Alexis and de Haas, Rob J. and Bethel, Neville and others}, title = {Robust deep learning-based protein sequence design using ProteinMPNN}, journal = {Science}, volume = {378}, number = {6615}, pages = {49--56}, year = {2022}, doi = {10.1126/science.add2187} } ``` ## Contact Please use [OFoldX GitHub issues](https://github.com/OTeam-AI4S/OFoldX/issues) for questions or comments about this model card. ## License The Hub `license` metadata, when present, reflects the source checkpoint or upstream project license. The OFoldX project license is not yet finalized. The source checkpoint is associated with the upstream license noted above: MIT for upstream ProteinMPNN and LigandMPNN code/model parameters. Review both OFoldX and upstream terms before redistribution or production use.
