Giancarlo Croce

---

T cells play a crucial role in eliminating infected and
cancerous cells. In cancer, inducing new T cell responses
or enhancing pre-existing ones is revolutionizing
immunotherapy treatments. A T cell is activated when its T
cell receptor (TCR) detects a specific peptide, called
epitope, on the surface of infected or cancerous cells. Each
person carries billions of T cells with unique TCRs, allowing
the detection of a wide range of viral and cancerous
targets. Unfortunately, such large diversity makes it difficult
to experimentally determine which T cells recognize which
epitopes. Machine learning offers a promising solution to link TCRs
with their cognate epitopes. In this presentation, I will
introduce MixTCRpred, our tool for predicting TCR-epitope
interactions, which recently ranked among the most
accurate in an international benchmarking competition. I
will outline its applications, including T cell-based
diagnostics to detect diseases from T cells in blood
samples, and identifying novel T cells to target specific
cancer epitopes.

---

Ludwig Institute for Cancer Research
215
Contact: Denny Trimcev