White Paper: How Reliable Are Antigen-Specificity Annotations for T-Cell Receptors?

The Evidence Behind Computational TCR Epitope Annotation

As the T-cell therapy and immunomonitoring industries mature, the question is no longer whether computational TCR annotation works, but how well, and how to trust it.

This white paper provides a comprehensive overview of the current state of the art in computational TCR-epitope specificity annotation. It addresses the critical challenges of database matching, seen epitope prediction, and the frontier of unseen epitope models, and demonstrates how ImmuneWatch DETECT has become the standard for reliable, scalable TCR annotation.

What you'll learn:

• Why exact TCR database matching is unreliable, and how modern algorithms solve four fundamental challenges: match ambiguity, database false positives, cross-reactivity, and TCR publicity bias.
• How ImmuneWatch DETECT achieved the highest classification score in the independent IMMREP23 international benchmark — and how it has been validated through external experiments, MLOps best practices, patient-level confirmation, and 14+ peer-reviewed user publications.
• Where the field stands on unseen epitope prediction, and how ImmuneWatch's hybrid modelling and data generation strategy is closing the gap toward fully generalisable TCR annotation.

Whether you are developing cell therapies, running clinical trials with immune monitoring endpoints, or exploring TCR-based diagnostics, this white paper will equip you with the scientific evidence to confidently integrate computational TCR annotation into your pipeline.