Predicting MHC binding affinity using deep learning. This repository was created as part of the recruitment process for an internship position at Insta Deep. The deadline was short so it is more focused on the method than the results.
The code was written in Python 3.11.4. Poetry was used to manage the dependencies. And a makefile was used to create shortcuts for the most common commands.
To install the project:
make install
Then you need to put the data in the data
folder at the root of the project.
Data visualization and analysis is done in the notebook data_exploration.ipynb
.
The method, the model and the results are explained in main.ipynb
.