USMPep: Universal Sequence Models for Major Histocompatibility Complex Binding Affinity Prediction
J. Vielhaben, M. Wenzel, W. Samek, N. Strodthoff
USMPep is a simple recurrent neural network that achieves state-of-the-art MHC class I binding prediction with a single architecture and hyperparameters, and shows solid performance for MHC class II despite limited data.
bioRxiv · Oct 24, 2019Immunology