StoPred Downloads
Last updated 2026-05-30
This page focuses on StoPred model downloads and runtime assets. For current code, visit GitHub.
download Core Downloads
StoPred Singularity container
Container image for local StoPred execution with the downloadable model releases.
Complete model collection
Current StoPred model releases and evaluation files.
Manuscript software package
Archive used for the manuscript. For current code, visit GitHub.
Manuscript training data
Training, validation, and test data used for the manuscript.
storage Model Releases
| Model name | Use case | Training cutoff | Folds | Archive size | Updated | Download |
|---|---|---|---|---|---|---|
default_20231001 |
Manuscript complex model | 2023-10-01 | 5 | 86.1 MB | 2026-05-30 | Download |
default_20260101 |
Default complex model | 2026-01-01 | 5 | 86.0 MB | 2026-05-30 | Download |
unk_single_20260101 |
Single-sequence unknown model | 2026-01-01 | 5 | 86.1 MB | 2026-05-30 | Download |
Model naming:
default_YYYYMMDDis a complex StoPred model. The date is the training cutoff date.default_20231001is the manuscript complex model.unk_single_YYYYMMDDshould only be used for single-entity inputs where it is unknown whether the protein is a monomer or a homomer. The default complex model assumes complex formation.
article Citation
Liu, Q., Peng, C., Zheng, W., Zhang, C., & Freddolino, L. (2025). StoPred: Accurate Stoichiometry Prediction for Protein Complexes Using Protein Language Models and Graph Attention. bioRxiv. https://doi.org/10.1101/2025.10.20.683515