---
license: cc-by-4.0
datasets:
- ChenLiu1996/ImmunoStruct
tags:
- chemistry
- biology
- medical
- immunology
- immunogenicity
---
# ImmunoStruct
[ImmunoStruct enables multimodal deep learning for immunogenicity prediction](https://www.nature.com/articles/s42256-025-01163-y)
[](https://www.nature.com/articles/s42256-025-01163-y)
[](https://www.nature.com/articles/s42256-025-01163-y.pdf)

[](https://huggingface.co/datasets/ChenLiu1996/ImmunoStruct)
[](https://huggingface.co/ChenLiu1996/ImmunoStruct)
[](https://github.com/KrishnaswamyLab/ImmunoStruct)
[](https://www.linkedin.com/in/kevin-bijan-givechian-phd-36467ba3/)
[](https://www.linkedin.com/in/joao-felipe-rocha/)
[](https://www.linkedin.com/in/chenliu1996/)
[](https://scholar.google.com/citations?user=3rDjnykAAAAJ&sortby=pubdate)
[](https://x.com/KevinGivechian)
[](https://x.com/ChenLiu_1996)
[](https://x.com/KrishnaswamyLab)
Project leads: [Kevin Bijan Givechian](https://www.linkedin.com/in/kevin-bijan-givechian-phd-36467ba3/), [João Felipe Rocha](https://www.linkedin.com/in/joao-felipe-rocha/), [Chen Liu](https://www.linkedin.com/in/chenliu1996/).
Correspondence: `akiko.iwasaki@yale.edu`, `smita.krishnaswamy@yale.edu`.
Instructions on preparing everything and running training/inference is provided on [our official GitHub repository](https://github.com/KrishnaswamyLab/ImmunoStruct).
The multimodal datasets of IEDB, CEDAR and clinical validation are available at [our huggingface dataset repo](https://huggingface.co/datasets/ChenLiu1996/ImmunoStruct).
In short, the models can be downloaded using
`hf download ChenLiu1996/ImmunoStruct --local-dir ./`
If possible, we will try to provide better serving of the models, but no guarantee.
**NOTE: [Our official repository](https://github.com/KrishnaswamyLab/ImmunoStruct) also includes detailed instructions and working code on how to run protein folding using AlphaFold2.**
## Model details
We open-sourced two pre-trained models in the paper:
1. IEDB_model_seed1.pt
2. CEDAR_model_seed2.pt
The former can be used to run inference on the IEDB dataset.
The latter can be used to run inference on the CEDAR dataset and the clinical validation dataset.
## Citation
If you find ImmunoStruct helpful to your research, please cite our paper:
BibTeX:
```bibtex
@article{givechian2026immunostruct,
title={ImmunoStruct enables multimodal deep learning for immunogenicity prediction},
author={Givechian, Kevin Bijan and Rocha, Jo{\~a}o Felipe and Liu, Chen and Yang, Edward and Tyagi, Sidharth and Greene, Kerrie and Ying, Rex and Caron, Etienne and Iwasaki, Akiko and Krishnaswamy, Smita},
journal={Nature Machine Intelligence},
volume={8},
pages={70--83},
year={2026},
publisher={Nature Publishing Group UK London}
}
```
Nature format:
Givechian, K.B., Rocha, J.F., Liu, C. et al. ImmunoStruct enables multimodal deep learning for immunogenicity prediction. *Nat Mach Intell* 8, 70–83 (2026). https://doi.org/10.1038/s42256-025-01163-y