DeepLigand: accurate prediction of MHC class I ligands using peptide embedding

Abstract

Motivation The computational modeling of peptide display by class I major histocompatibility complexes (MHCs) is essential for peptide-based therapeutics design. Existing computational methods for peptide-display focus on modeling the peptide-MHC-binding affinity. However, such models are not able to characterize the sequence features for the other cellular processes in the peptide display pathway that determines MHC ligand selection. Results We introduce a semi-supervised model, DeepLigand that outperforms the state-of-the-art models in MHC Class I ligand prediction. DeepLigand combines a peptide language model and peptide binding affinity prediction to score MHC class I peptide presentation. The peptide language model characterizes sequence features that correspond to secondary factors in MHC ligand selection other than binding affinity. The peptide embedding is learned by pre-training on natural ligands, and can discriminate between ligands and non-ligands in the absence of binding affinity prediction. Although conventional affinity-based models fail to classify peptides with moderate affinities, DeepLigand discriminates ligands from non-ligands with consistently high accuracy. Availability and implementation We make DeepLigand available at https://github.com/gifford-lab/DeepLigand. Supplementary information Supplementary data are available at Bioinformatics online.

Current challenges for epitope-agnostic TCR interaction prediction and a new perspective derived from image classification

“…Using this type of interaction map as the model input differs from the prevailing approach seen in sequence-based prediction models, not only in the field of TCR-epitope prediction, but also for the modeling of other molecular interactions. The interacting molecules are usually supplied to a model as separate or concatenated inputs (1, 3,5,6,15,18,20,25,26). Those types of models need to learn an internal representation for each molecule separately, before being combined again in deeper layers.…”

Deep Learning in Proteomics

“…[141] During the past few years, a number of deep learningbased methods have been developed that outperform traditional machine learning methods, including shallow neural networks, for peptide-MHC binding prediction ( Table 4). Among these algorithms, 14 (ConvMHC, [142] HLA-CNN, [143] DeepMHC, [144] DeepSeqPan, [145] MHCSeqNet, [146] MHCflurry, [147] DeepHLApan, [148] ACME, [149] EDGE, [137] CNN-NF, [150] DeepNeo, [151] DeepLigand, [152] MHCherryPan, [153] and DeepAttentionPan [141] ) are specific for MHC class I binding prediction, three (DeepSeqPanII, [154] MARIA, [138] and NeonMHC2 [139] ) are specific for MHC class II binding prediction, and four (AI-MHC, [155] MHCnuggets, [156] PUFFIN, [157] and USMPep [158] ) can make predictions for both classes. All four types of peptide encoding approaches illustrated in Figure 1 are used in these tools, with one-hot encoding and BLOSUM matrix encoding being the most frequently used methods (Table 4).…”

“…A new network initiated with weights transferred from the first step is further trained with immunopeptidomics data when available. In DeepLigand, 152 two modules are combined to predict MHC-I peptide presentation rather than binding affinity prediction. The first module is a pan-specific binding affinity prediction module based on a deep residual network while the second one is a peptide embedding module based on a deep language model (ELMo [160] ).…”

Uncovering the Tumor Antigen Landscape: What to Know about the Discovery Process

“…Zhao et al [ 185 ] developed a convolutional neural network that uses different peptide properties, such as the order of the sequence, the hydropathy index, polarity, and the length of peptide needed to perform the prediction, as these are the key factors in determining the binding to the HLA molecule. Zeng and Gifford developed a deep learning-based method that consists of a binding affinity prediction module and a peptide embedding module [ 186 ]. The latter applies a deep language model to embed each peptide into a vector representation.…”