vibtcr is a Python package which implements the Mutlimodal Variational Information
Bottleneck (MVIB) and the Attentive Variational Information Bottleneck (AVIB), with a focus on
TCR-peptide interaction prediction.
Paper: "Attentive Variational Information Bottleneck for TCR–peptide interaction prediction", F. Grazioli, P. Machart, A. Mösch, K. Li, L.V. Castorina, N. Pfeifer, M.R. Min, Bioinformatics 2022
cd vibtcr
pip install .
Remark: vibtcr requires a different version of PyTorch than tcrmodels. It's recommended to install them in different environments.
vibtcr
│ README.md
│ ...
│
└───data/
│ │ alpha-beta-splits/ (all TCR data, split in two disjoint sets: alpha+beta, beta-only)
│ │ ergo2-paper/ (data used in the ERGO II paper - contains VDJDB and McPAS)
│ │ mhc/ (the NetMHCIIpan-4.0 data)
│ │ nettcr2-paper/ (data used in the NetTCR2.0 paper - contains IEDB, VDJDB and MIRA)
│ │ vdjdb/ (complete VDJdb data from 5th of September)
│
└───notebooks/
│ │ notebooks.classification/ (TCR-peptide experiments with AVIB/MVIB)
| │ notebooks.mouse/ (identifying mouse TCRs as suitable OOD dataset)
│ │ notebooks.ood/ (out-of-distribution detection experiments with AVIB)
│ │ notebooks.regression/ (peptide-MHC BA regression with AVIB)
│
└───tcrmodels/ (Python package which wraps SOTA ML-based TCR models)
│
└───vibtcr/ (Python package which implements MVIB and AVIB for TCR-peptide interaction prediction)
tcrmodels wraps state-of-the-art ML-based TCR prediction models.
So far, it includes:
cd tcrmodels
pip install .
pip install torch===1.4.0 torchvision===0.5.0 -f https://download.pytorch.org/whl/torch_stable.html
tcrmodels requires Python 3.6
Springer I, Tickotsky N and Louzoun Y (2021), Contribution of T Cell Receptor Alpha and Beta CDR3, MHC Typing, V and J Genes to Peptide Binding Prediction. Front. Immunol. 12:664514. DOI: https://doi.org/10.3389/fimmu.2021.664514
Montemurro, A., Schuster, V., Povlsen, H.R. et al. NetTCR-2.0 enables accurate prediction of TCR-peptide binding by using paired TCRα and β sequence data. Commun Biol 4, 1060 (2021). DOI: https://doi.org/10.1038/s42003-021-02610-3
For vibtcr, we provide a non-commercial license, see LICENSE.txt
If you find this work useful, please cite:
@article{10.1093/bioinformatics/btac820,
author = {Grazioli, Filippo and Machart, Pierre and Mösch, Anja and Li, Kai and Castorina, Leonardo V and Pfeifer, Nico and Min, Martin Renqiang},
title = "{Attentive Variational Information Bottleneck for TCR–peptide interaction prediction}",
journal = {Bioinformatics},
volume = {39},
number = {1},
year = {2022},
month = {12},
issn = {1367-4811},
doi = {10.1093/bioinformatics/btac820},
url = {https://doi.org/10.1093/bioinformatics/btac820},
note = {btac820},
eprint = {https://academic.oup.com/bioinformatics/article-pdf/39/1/btac820/48493569/btac820.pdf},
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent