Dataset Information

Improving MHC class I antigen-processing predictions using representation learning and cleavage site-specific kernels.

ABSTRACT: In this work, we propose a new deep-learning model, MHCrank, to predict the probability that a peptide will be processed for presentation by MHC class I molecules. We find that the performance of our model is significantly higher than that of two previously published baseline methods: MHCflurry and netMHCpan. This improvement arises from utilizing both cleavage site-specific kernels and learned embeddings for amino acids. By visualizing site-specific amino acid enrichment patterns, we observe that MHCrank's top-ranked peptides exhibit enrichments at biologically relevant positions and are consistent with previous work. Furthermore, the cosine similarity matrix derived from MHCrank's learned embeddings for amino acids correlates highly with physiochemical properties that have been experimentally demonstrated to be instrumental in determining a peptide's favorability for processing. Altogether, the results reported in this work indicate that MHCrank demonstrates strong performance compared with existing methods and could have vast applicability in aiding drug and vaccine development.

SUBMITTER: Lawrence PJ

PROVIDER: S-EPMC9499997 | biostudies-literature | 2022 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Improving MHC class I antigen-processing predictions using representation learning and cleavage site-specific kernels.

Lawrence Patrick J PJ Ning Xia X

Cell reports methods 20220919 9

In this work, we propose a new deep-learning model, MHCrank, to predict the probability that a peptide will be processed for presentation by MHC class I molecules. We find that the performance of our model is significantly higher than that of two previously published baseline methods: MHCflurry and netMHCpan. This improvement arises from utilizing both cleavage site-specific kernels and learned embeddings for amino acids. By visualizing site-specific amino acid enrichment patterns, we observe th ...[more]

PMID: 36160050

Similar Datasets

Project description:The well described conventional antigen-processing pathway is accountable for most peptides that end up in MHC class I molecules at the cell surface. These peptides experienced liberation by the proteasome and transport by the peptide transporter TAP. However, there are multiple roads that lead to Rome, illustrated by the increasing number of alternative processing pathways that have been reported during last years. Interestingly, TAP-deficient individuals do not succumb to viral infections, suggesting that CD8 T cell immunity is sufficiently supported by alternative TAP-independent processing pathways. To date, a diversity of viral and endogenous TAP-independent peptides have been identified in the grooves of different MHC class I alleles. Some of these peptides are not displayed by normal TAP-positive cells and we therefore called them TEIPP, for "T-cell epitopes associated with impaired peptide processing." TEIPPs are hidden self-antigens, are derived from normal housekeeping proteins, and are processed via unconventional processing pathways. Per definition, TEIPPs are presented via TAP-independent pathways, but recent data suggest that part of this repertoire still depend on proteasome and metalloprotease activity. An exception is the C-terminal peptide of the endoplasmic reticulum (ER)-membrane-spanning ceramide synthase Trh4 that is surprisingly liberated by the signal peptide peptidase (SPP), the proteolytic enzyme involved in cleaving leader sequences. The intramembrane cleaving SPP is thereby an important contributor of TAP-independent peptides. Its family members, like the Alzheimer's related presenilins, might contribute as well, according to our preliminary data. Finally, alternative peptide routing is an emerging field and includes processes like the unfolded protein response, the ER-associated degradation, and autophagy-associated vesicular pathways. These data convince us that there is a world to be discovered in the field of unconventional antigen processing.

Dataset Information

Improving MHC class I antigen-processing predictions using representation learning and cleavage site-specific kernels.

Publications

Improving MHC class I antigen-processing predictions using representation learning and cleavage site-specific kernels.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets