POPISK: T-cell reactivity prediction using support vector machines and string kernels

Abstract

BackgroundAccurate prediction of peptide immunogenicity and characterization of relation between peptide sequences and peptide immunogenicity will be greatly helpful for vaccine designs and understanding of the immune system. In contrast to the prediction of antigen processing and presentation pathway, the prediction of subsequent T-cell reactivity is a much harder topic. Previous studies of identifying T-cell receptor (TCR) recognition positions were based on small-scale analyses using only a few peptides and concluded different recognition positions such as positions 4, 6 and 8 of peptides with length 9. Large-scale analyses are necessary to better characterize the effect of peptide sequence variations on T-cell reactivity and design predictors of a peptide's T-cell reactivity (and thus immunogenicity). The identification and characterization of important positions influencing T-cell reactivity will provide insights into the underlying mechanism of immunogenicity.ResultsThis work establishes a large dataset by collecting immunogenicity data from three major immunology databases. In order to consider the effect of MHC restriction, peptides are classified by their associated MHC alleles. Subsequently, a computational method (named POPISK) using support vector machine with a weighted degree string kernel is proposed to predict T-cell reactivity and identify important recognition positions. POPISK yields a mean 10-fold cross-validation accuracy of 68% in predicting T-cell reactivity of HLA-A2-binding peptides. POPISK is capable of predicting immunogenicity with scores that can also correctly predict the change in T-cell reactivity related to point mutations in epitopes reported in previous studies using crystal structures. Thorough analyses of the prediction results identify the important positions 4, 6, 8 and 9, and yield insights into the molecular basis for TCR recognition. Finally, we relate this finding to physicochemical properties and structural features of the MHC-peptide-TCR interaction.ConclusionsA computational method POPISK is proposed to predict immunogenicity with scores which are useful for predicting immunogenicity changes made by single-residue modifications. The web server of POPISK is freely available at http://iclab.life.nctu.edu.tw/POPISK.

West Nile Virus T-Cell Ligand Sequences Shared with Other Flaviviruses: a Multitude of Variant Sequences as Potential Altered Peptide Ligands

“…A total of 76 (38%) of the 200 positive HLA-restricted ELISpot responses corresponded to at least one predicted nonamer epitope with a concordant predicted HLA restriction (see Table S1). This relatively low extent of correspondence among the discovered HLA-restricted immunogenic peptides and predicted HLA binding motifs is comparable to those reported for other pathogens (27) and can be attributed, at least in part, to the fact that HLA binding prediction does not account for the pivotal role of the TCR in the immunogenicity of a peptide (67).…”

miHA-Match: Computational detection of tissue-specific minor histocompatibility antigens

“…Additional factors are antigen processing and the availability of a T-cell receptor that matches the peptide: HLA complex. While good prediction methods for HLA binding are available, the prediction of antigen processing and T-cell reactivity leave room for improvement (Tung et al, 2011). We therefore restrict the epitope prediction to the prediction of HLA binding in the default setting.…”

PAAQD: Predicting immunogenicity of MHC class I binding peptides using amino acid pairwise contact potentials and quantum topological molecular similarity descriptors

“…The immunogenicity does not only depend on the particular allele of HLA and the type of TCR in host immune system but is also governed by negative T-cell selection (central tolerance). The central tolerance is defined as the property of the whole proteome and cannot casually be learned by machine learning approaches (Van Regenmortel, 2001;Kanduc, 2005;Tung et al, 2011).…”

“…The PAAQD's performance was evaluated by using the IMMA2 dataset published by Tung et al (2011) and compared with the two existing T-cell reactivity predictors, POPI and POPISK. We evaluated the importance of positions by removing features corresponding to specific position of nonapeptides.…”

“…POPI used 23 informative physicochemical properties collected from the AAindex database (Kawashima et al, 2008) for encoding peptides and applied the support vector machine (SVM) as a classifier. The second predictor is POPISK (Tung et al, 2011). POPISK used the SVM with string kernels.…”