This repo is under active development: full installation instructions coming soon

📄 Paper: CIRPIN: Learning Circular Permutation-Invariant Representations to Uncover Putative Protein Homologs

Work In Progres

Colab Notebook CIRPIN

1. Install prerequisites

   CIRPIN requires:

   • Progres
   • TM-align
   • Foldcomp

2. Updating model weights

   To run CIRPIN, load CIRPIN/trained_models/CIRPIN_model/CIRPIN_model_5k_cp_epoch301.pt

   To run Progres, load CIRPIN/trained_models/Progres_model/Progres_trained_model.pt

3. Embedded AFDB-ClustR

   The full AFDB-ClustR embedded using CIRPIN/Progres is available for download at link

1. Datasets of CPs found SCOPe 40%: CIRPIN/scope40

2. Datasets of CPs found in AFDB-ClustR: link

License: MIT License - See LICENSE file for details
Citation: If you use CIRPIN in your research, please cite:

@article {Kolodziej2025.11.18.689110,
	author = {Kolodziej, Aiden R and Abulnaga, S. Mazdak and Ovchinnikov, Sergey},
	title = {CIRPIN: Learning Circular Permutation-Invariant Representations to Uncover Putative Protein Homologs},
	elocation-id = {2025.11.18.689110},
	year = {2025},
	doi = {10.1101/2025.11.18.689110},
	publisher = {Cold Spring Harbor Laboratory},
	URL = {https://www.biorxiv.org/content/early/2025/11/18/2025.11.18.689110},
	journal = {bioRxiv}
}

Questions or Collaboration: [email protected]