Publication Date: 12/26/2022
Reference: 39(1): 1-8, 2022
Authors: Filippo Grazioli, NEC Laboratories Europe; Pierre Machart, NEC Laboratories Europe; Anja Mosch, NEC Laboratories Europe; Kai Li, NEC Laboratories America, Inc.; Leonardo V. Castorina, University of Edinburgh; Nico Pfeifer, University of Tubingen; Martin Renqiang Min, NEC Laboratories America, Inc.
Abstract: MotivationWe present a multi-sequence generalization of Variational Information Bottleneck and call the resulting model Attentive Variational Information Bottleneck (AVIB). Our AVIB model leverages multi-head self-attention to implicitly approximate a posterior distribution over latent encodings conditioned on multiple input sequences. We apply AVIB to a fundamental immuno-oncology problem: predicting the interactions between T-cell receptors (TCRs) and peptides.ResultsExperimental results on various datasets show that AVIB significantly outperforms state-of-the-art methods for TCR–peptide interaction prediction. Additionally, we show that the latent posterior distribution learned by AVIB is particularly effective for the unsupervised detection of out-of-distribution amino acid sequences.
Publication Link: https://academic.oup.com/bioinformatics/article/39/1/btac820/6960920