This application explores the vast diversity of T-cell receptor (TCR) interaction with antigens, based on their epitopes.

TCR expression in human T-cells results from the somatic recombination of V and J genes. This process leads to a theoretically extensive repertoire of TCRs that could potentially recognize various antigens from pathogens. However, even with this theoretical diversity, the human immune system would still not be able to create an immune response against every existing pathogen. To overcome this limitation, T-cells are believed to be cross-reactive, meaning that a single T-cell can recognize many different peptides presented on MHC molecules. The same antigens can also be recognized by multiple TCRs, based on distinct antigen features. This cross-reactivity is critical for providing an adequate immune response to different pathogens.

However, while cross-reactivity is favorable in immune responses against pathogens, it can also have unintended consequences. For instance, it might play a role in the development of auto-immune diseases, or it can lead to off-target effects in cancer immunotherapies. Further understanding TCR cross-reactivity might thus improve numerous applications, including predicting viral escape, cancer neoantigen immunogenicity, autoimmunity, and off-target toxicity of T-cell-based therapies

Existing computational methods are able to predict all TCRs that bind a given epitope. But this application focuses on the opposite, predicting all epitopes that a given TCR binds. It is build on a comprehensive experimental mutational scan database on TCR activation, visualizing the predictions of how peptide mutations affect TCR activation.

The application provides an overview over the similarity space of mutated epitopes, visualizes the binding activity of cognate epitopes and their single amino acid mutations and illustrates the binding activity of an epitope and its mutations to one or multiple TCRs.

In summary, this application provides valuable insights into the cross-reactivity of TCRs and how it influences the immune response to various antigens.

Epitope similarity by diferent distance functions

The Epitope Cluster function clusters mutated epitopes according to their mutual sequence similarities, while showing their binding affinity. This information is based on user-defined distance functions and weights assigned to mutations at specific positions within the peptides.

TCR-pMHC normalized binding activity

The Peptide Binding Activity tab displays a heatmapto visualize the normalized binding activity of a user-choosen petide and all its single amino acid mutations to one or multiple TCRs. All possible single amino acid mutations per peptide position are displayed in alphabetical amino acid order as one block of the heatmap.

Alluvium plot displaing TCR-pMHC binding

The Mutant-TCR Interaction tab reveals which index peptides and their mutations are bound by multiple TCRs and which are recognized by only one TCR, emphasizing how different TCRs specific for the same index peptide recognize different mutants of it.