Model code and data for paper "Learning antibody sequence constraints from allelic inclusion", Jagota et. al., 2024. BioRxiv preprint: https://www.biorxiv.org/content/10.1101/2024.10.22.619760v1
We recommend using a GPU. Allelic inclusion models train on a GPU in a few minutes, and memory requirements are low.
Training data for mouse models is not included on GitHub but is available on Zenodo at https://zenodo.org/records/13948324.
Required Python packages and version used:
- torch (2.2.1)
- numpy (1.26.4)
- pandas (2.1.4)
- tqdm (4.65.0)
- scipy (1.11.4)
- sklearn (1.3.0)