# HLAIIPred: cross-attention mechanism for modeling the interaction of HLA class II molecules with peptides

## TL;DR

HLAIIPred is a deep learning model that predicts how HLA class II molecules interact with peptides, improving predictions for immunogenicity and cancer immunotherapy.

## Contribution

HLAIIPred introduces a cross-attention Transformer-based model for HLA class II peptide prediction with improved accuracy and insights into core peptide residues.

## Key findings

- HLAIIPred outperforms existing models with a 16% increase in predicting peptides presented by less frequent alleles.
- The model identifies core peptide residues involved in HLAII interactions and improves immunogenicity prediction for therapeutic antibodies.
- HLAIIPred prioritizes neoantigens for cancer immunotherapy with high accuracy.

## Abstract

We introduce HLAIIPred, a deep learning model to predict peptides presented by class II human leukocyte antigens (HLAII) on the surface of antigen presenting cells. HLAIIPred is trained using a Transformer-based neural network and a dataset comprising of HLAII-presented peptides identified by mass spectrometry. In addition to predicting peptide presentation, the model can also provide important insights into peptide-HLAII interactions by identifying core peptide residues that form such interactions. We evaluate the performance of HLAIIPred on three different tasks, peptide presentation in monoallelic samples, immunogenicity prediction of therapeutic antibodies, and neoantigen prioritization for cancer immunotherapy. Additionally, we created a dataset of biotherapeutics HLAII peptides presented by human dendritic cells. This data is used to develop screening strategies to predict the unwanted immunogenic segments of therapeutic antibodies by HLAII presentation models. HLAIIPred demonstrates superior or equivalent performance when compared to the latest models across all evaluated benchmark datasets. We achieve a 16% increase in prediction of presented peptides compared to the second-best model on a set of unseen peptides presented by less frequent alleles. The model improves clinical immunogenicity prediction, identifies epitopes in therapeutic antibodies and prioritize neoantigens with high accuracy.

HLAIIPred is an end-to-end, context-free deep learning model that predicts peptide-HLAII presentation using only sequence and allele input. It outperforms existing models, predicts core residues, and identifies immunogenic hotspots in antibodies.