Machine learning analysis of the T cell receptor repertoire identifies sequence features of self-reactivity

Abstract

The Tcell receptor (TCR) determines specificity and affinity for both foreign and self-peptides presented by the major histocompatibility complex (MHC). Although the strength of TCR interactions with self-pMHC impacts Tcell function, it has been challenging to identify TCR sequence features that predict Tcell fate. To discern patterns distinguishing TCRs from naive CD4+ Tcells with low versus high self-reactivity, we used data from 42 mice to train a machine learning (ML) algorithm that identifies population-level differences between TCRβ sequence sets. This approach revealed that weakly self-reactive Tcell populations were enriched for longer CDR3β regions and acidic amino acids. We tested our ML predictions of self-reactivity using retrogenic mice with fixed TCRβ sequences. Extrapolating our analyses to independent datasets, we predicted high self-reactivity for regulatory Tcells and slightly reduced self-reactivity for Tcells responding to chronic infections. Our analyses suggest a potential trade-off between TCR repertoire diversity and self-reactivity. A record of this paper's transparent peer review process is included in the supplemental information.