Immuno-Oncology Therapies and Precision Medicine: Personal Tumor-Specific Neoantigen Prediction by Machine Learning

Saturday, September 16, 2017 - 11:20am - 11:50am
Keller 3-180
Yi-Hsiang Hsu (Harvard Medical School)
The fast moving of the cancer immunotherapy field has generated tremendous excitement regarding new therapeutic strategies and will likely change the paradigm of therapeutic interventions for cancer. Neoantigens, generated by tumor-specific DNA alterations that result in the formation of novel protein sequences and only in cancer cells, represent an optimal target for the immune system and make possible a new class of highly personalized vaccines with the potential for significant efficacy with reduced side effects. Although many somatic mutations could be presented as potential neoantigens, not all mutations can be processed for T-cell recognition. Therefore, the accurate identification of neoantigens with immunogenicity is critical in the development of a patient-specific cancer treatment. Reliable predictions of immunogenic peptides are necessary in cancer immunotherapy design and can minimize the experimental effort needed to identify epitopes. In this talk, I will present our work using a Convolutional Neural Network (CNN) model to predict tumor-specific neoantigens by identifying features of peptide sequences, MHC sequences and their interactions as well as TCR recognition pattern prediction.