Repository compiling the notebooks used for the investigation of my M4R Project at Imperial College London 📚

🔗 BioRxiv article

My investigation focused on the analysis of different unsupervised techniques in order to accurately cluster similar TCR sequences (i.e. sequences with a similar antigen specificity). 🫁🫀

Please note that the notebooks were used as exploratory tools 🔍.

Legend: The name of the notebooks start with a number that indicates the order in which these techniques and methods are discussed in the research paper. Note that "A" corresponds to extra/appendix notebooks not directly relevant to the analysis.

The automate.py file contains some essential functions that automate the pipeline for model building and results plotting. This simplifies readability and usage of our different modelling techniques.

Here is a link to a web application that shows how perform the different deep learning methods:

🔗 Website of the demo

🔗 GitHub repo of the demo