PXDesign: Fast, Modular, and Accurate De Novo Design of Protein Binders

PXDesign achieves nanomolar binder hit rates of 17–82% across six of seven diverse protein targets, surpassing prior methods such as AlphaProteo. This experimental success rate is enabled by advances in both binder generation and filtering. We develop both a diffusion-based generative model (PXDesign-d) and a hallucination-based approach (PXDesign-h), each showing strong in silico performance that outperforms existing models. Beyond generation, we systematically analyze confidence-based filtering and ranking strategies from multiple structure predictors, comparing their accuracy, efficiency, and complementarity on datasets spanning de novo binders and mutagenesis. Finally, we validate the full design process experimentally, achieving high hit rates and multiple nanomolar binders.

To support future research and broaden community adoption, we release the full PXDesign pipeline ( https://github.com/bytedance/PXDesign ), provide public access to PXDesign through a dedicated web server ( https://protenix-server.com ), and make all designed binder sequences available at the project page ( https://protenix.github.io/pxdesign ).