Annika Nelde

This paper introduces Immunotype, a deep learning model that directly types HLA alleles from immunopeptidomics data with 87.2% accuracy.

The paper introduces t2pmhc, a structure-informed graph neural network for predicting TCR-pMHC binding, showing improved generalization to unseen peptides compared to existing methods.

This study uses ribosome profiling to identify conserved HIV-1 alternative open reading frames (ARFs) that encode T cell antigens, revealing new targets for vaccination.