Deep generative selection models of T and B cell receptor repertoires with soNNia

TL;DR

This paper introduces a machine learning framework combining biophysical models and deep generative models to analyze T and B cell receptor repertoires, identifying features associated with immune functions and pathogen recognition.

Contribution

It develops a novel approach integrating biophysical receptor generation models with deep learning to classify lymphocyte subtypes and identify functional receptor features.

Findings

Successfully classifies CD4+ and CD8+ T-cells using sequence data

Identifies receptor features linked to immune function and pathogen targeting

Shows simple classifiers can perform comparably to complex models

Abstract

Subclasses of lymphocytes carry different functional roles to work together to produce an immune response and lasting immunity. Additionally to these functional roles, T and B-cell lymphocytes rely on the diversity of their receptor chains to recognize different pathogens. The lymphocyte subclasses emerge from common ancestors generated with the same diversity of receptors during selection processes. Here we leverage biophysical models of receptor generation with machine learning models of selection to identify specific sequence features characteristic of functional lymphocyte repertoires and subrepertoires. Specifically using only repertoire level sequence information, we classify CD4$^+$ and CD8$^+$ T-cells, find correlations between receptor chains arising during selection and identify T-cells subsets that are targets of pathogenic epitopes. We also show examples of when simple linear classifiers do as well as more complex machine learning methods.