amulety 1.1

Python package to create embeddings of BCR amino acid sequences.

AMULETY

Amulety stands for Adaptive imMUne receptor Language model Embedding Tool. It is a Python command line tool to embed B-cell receptor (antibody) and T-cell Receptor amino acid sequences using pre-trained protein or antibody language models. So far only BCR embeddings are supported but TCR support is planned for future releases. The package also has functionality to translate nucleotide sequences to amino acids with IgBlast.

Here is the list of currently supported embeddings:

Model	Command	Embedding Dimension	Reference
AntiBERTa2	antiberta2	1024	doi:10.1016/j.patter.2022.100513
AntiBERTy	antiberty	512	doi:10.48550/arXiv.2112.07782
BALM-paired	balm_paired	1024	doi:10.1016/j.patter.2024.100967
ESM2 (650M parameter)	esm2	1280	doi:10.1126/science.ade2574
User-specified model	custommodel	Configurable

Installation

You can install AMULETY using pip:

pip install amulety

Usage

To print the usage help for the AMULETY package then type:

amulety --help

The full usage documentation can also be found on the readthedocs usage page.

Contact

For help and questions please contact the Immcantation Group.

Authors

Mamie Wang (aut,cre) Gisela Gabernet (aut,cre) Steven Kleinstein (aut,cph)

Citing

If you use this package, please cite the pre-print:

AMULETY: A Python package to embed adaptive immune receptor sequences. Meng Wang, Yuval Kluger, Steven H. Kleinstein, Gisela Gabernet. BioRXiv 2025. DOI: https://doi.org/10.1101/2025.03.21.644583

To cite the paper comparing the embedding methods on BCR sequences, please cite:

Supervised fine-tuning of pre-trained antibody language models improves antigen specificity prediction. Meng Wang, Jonathan Patsenker, Henry Li, Yuval Kluger, Steven H. Kleinstein. BioRXiv 2024. DOI: https://doi.org/10.1101/2024.05.13.593807.

License

This project is licensed under the terms of the GPL v3 license. See the LICENSE file for details.