> ## Documentation Index > Fetch the complete documentation index at: https://proto.evodesign.org/docs/llms.txt > Use this file to discover all available pages before exploring further. # FAMPNN Inverse Folding > FAMPNN structure-conditioned protein sequence design with full-atom sidechain co-generation.
FAMPNN Inverse Folding
**License:** FAMPNN is open source and free for academic and commercial use under an MIT license. Please refer to [the license](https://github.com/richardshuai/fampnn/blob/main/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.

Go to Tool Page
evo-design/proto-language/proto\_language/generator/fampnn\_generator.py
View source
```bibtex theme={null} @inproceedings{widatalla2025fampnn, title={Sidechain conditioning and modeling for full-atom protein sequence design with {FAMPNN}}, author={Widatalla, Talal and Shuai, Richard W. and Hie, Brian L. and Huang, Po-Ssu}, booktitle={Proceedings of the 42nd International Conference on Machine Learning}, year={2025}, series={PMLR}, volume={267}, address={Vancouver, Canada} } ```
Copy citation
Generator contributors
Protein sequence generator using FAMPNN. ## API Reference
ConfigFAMPNNGeneratorConfig Source
Configuration for FAMPNN inverse-folding generation. FAMPNN jointly designs amino-acid identities and sidechain conformations conditioned on an input backbone. The generator writes the designed protein sequence to the assigned segment and attaches FAMPNN's full-atom output structure to `proposal.structure`. Structure(s) with optional chain, fixed-position, and fixed-sidechain selections. When sampling a multi-chain structure, write only this chain's sequence to the target segment. FAMPNN checkpoint variant for sequence design. Sampling temperature; lower is greedier and higher is more diverse. Number of sequences to process simultaneously on GPU. Number of iterative unmasking steps for sequence design. If true, skip sidechain generation during sampling. Repack sidechains after the final sequence is determined. Only condition on sidechains below this predicted-error threshold during design. Number of sidechain diffusion denoising steps. Step scale for sidechain diffusion. Device for model inference. Whether to print FAMPNN progress logs.
## Usage ```python python icon="python" theme={null} from proto_language.generator import FAMPNNGenerator, FAMPNNGeneratorConfig from proto_language.core import Segment config = FAMPNNGeneratorConfig( # Configure parameters here ) generator = FAMPNNGenerator(config) segment = Segment(length=100, sequence_type="protein") generator.assign(segment) generator.sample() ``` ## Metadata | Property | Value | | ------------------------ | ----------------- | | Key | `fampnn` | | Class | `FAMPNNGenerator` | | Category | `inverse_folding` | | Input Type | `structure` | | Uses GPU | `True` | | Supported Sequence Types | `protein` | | Allows Empty Start | `False` |