tattabio/flashppi2

FlashPPI: Linear-time prediction of proteome-scale microbial protein interactions

NEW: FlashPPI2 is the latest version of FlashPPI which includes training on high-confidence AFDB heterodimer complexes, improving AUPRC by 17% of the tattabio/ecoli_pdb_benchmark

The original FlashPPI model is available at tattabio/flashppi.

Web Server

FlashPPI is integrated into seqhub.org. You can upload a FASTA and interactively explore whole-proteome networks and contact maps. Explore an example network here.

Usage

See the github repo https://github.com/TattaBio/flashppi for usage and inference scripts.

License

The model code and inference scripts in the GitHub repo are licensed under the Apache License 2.0.

The FlashPPI model weights are hosted on Hugging Face and released under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license. The weights are freely available for academic and research purposes.

Citing

If you use FlashPPI or our datasets in your research, please cite:

@article{
  doi:10.1073/pnas.2610619123,
  author = {Andre Cornman  and Matt Tranzillo  and Nicolo G. Zulaybar  and Imane Bouzit  and Yunha Hwang },
  title = {Linear-time prediction of proteome-scale microbial protein interactions},
  journal = {Proceedings of the National Academy of Sciences},
  volume = {123},
  number = {25},
  pages = {e2610619123},
  year = {2026},
  doi = {10.1073/pnas.2610619123},
  URL = {https://www.pnas.org/doi/abs/10.1073/pnas.2610619123},
  eprint = {https://www.pnas.org/doi/pdf/10.1073/pnas.2610619123},
}