To what extent does MHC binding translate to immunogenicity in humans?

Lee, Chloe H.; Antanaviciute, Agne; Buckley, Paul; Simmons, Alison; Koohy, Hashem

doi:10.1016/j.immuno.2021.100006

Cited by 9 publications

(25 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite broad application of these models in predicting peptide targets for CD8+ T cell responses in both cancer and infection settings, there is no systematic comparative analysis of features associated to cancer versus pathogenic peptide immunogenicity [ 21 ]. As discussed above in identifying immunogenic peptides in these two settings, we noticed inconsistent performances between cancer versus pathogenic scenarios as well as within cancer settings.…”

Section: Resultsmentioning

confidence: 99%

“…[ 23 ] demonstrated that dissimilarity of a peptide to the self-proteome is associated with neoantigen immunogenicity. However, our recent work [ 21 ] has demonstrated that dissimilarity to self is limited in discriminating immunogenic pathogenic peptides. Indeed, Koncz et al .…”

Section: Resultsmentioning

confidence: 99%

“…Compared with pathogenic peptides, which are substantially different from the human proteome, neoantigens often exhibit only a single point mutation from the corresponding wild-type self-peptide [ 20 ]. This high sequence similarity between cancer neoantigens and self-peptides is likely subject to immune tolerance; thus, the focus has been placed on identifying features that permit neoepitopes to escape, which may be less applicable for pathogenic epitopes [ 21 ].…”

Section: Introductionmentioning

confidence: 99%

“…[ 23 ] have shown that dissimilarity to the self-proteome permits neoepitopes to escape from immune tolerance. However, we have recently shown that dissimilarity to self is limited in distinguishing immunogenic peptides from pathogens [ 21 ], which indicate differences in the features required to predict immunogenic peptides from pathogens versus cancer. Complicating matters further, existing models are primarily trained on pathogenic epitopes, some of which are used then to predict immunogenic neoantigens.…”

Section: Introductionmentioning

confidence: 99%

“…However, we have recently observed that the fraction of presented peptides in humans which are immunogenic may be more variable (2.4–69% across different studies) than observed by Croft et al . in mice [ 21 ]. It is unclear how much of this variation is a species-specific effect, or perhaps due to differences between experimental assays.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Evaluating performance of existing computational models in predicting CD8+ T cell pathogenic epitopes and cancer neoantigens

Buckley

Lee

et al. 2022

Briefings in Bioinformatics

Self Cite

View full text Add to dashboard Cite

T cell recognition of a cognate peptide–major histocompatibility complex (pMHC) presented on the surface of infected or malignant cells is of the utmost importance for mediating robust and long-term immune responses. Accurate predictions of cognate pMHC targets for T cell receptors would greatly facilitate identification of vaccine targets for both pathogenic diseases and personalized cancer immunotherapies. Predicting immunogenic peptides therefore has been at the center of intensive research for the past decades but has proven challenging. Although numerous models have been proposed, performance of these models has not been systematically evaluated and their success rate in predicting epitopes in the context of human pathology has not been measured and compared. In this study, we evaluated the performance of several publicly available models, in identifying immunogenic CD8+ T cell targets in the context of pathogens and cancers. We found that for predicting immunogenic peptides from an emerging virus such as severe acute respiratory syndrome coronavirus 2, none of the models perform substantially better than random or offer considerable improvement beyond HLA ligand prediction. We also observed suboptimal performance for predicting cancer neoantigens. Through investigation of potential factors associated with ill performance of models, we highlight several data- and model-associated issues. In particular, we observed that cross-HLA variation in the distribution of immunogenic and non-immunogenic peptides in the training data of the models seems to substantially confound the predictions. We additionally compared key parameters associated with immunogenicity between pathogenic peptides and cancer neoantigens and observed evidence for differences in the thresholds of binding affinity and stability, which suggested the need to modulate different features in identifying immunogenic pathogen versus cancer peptides. Overall, we demonstrate that accurate and reliable predictions of immunogenic CD8+ T cell targets remain unsolved; thus, we hope our work will guide users and model developers regarding potential pitfalls and unsettled questions in existing immunogenicity predictors.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%