Prediction of T-cell epitopes based on least squares support vector machines and amino acid properties

T-lymphocyte (T-cell) is a very important component in human immune system. It possesses a receptor (TCR) that is specific for the foreign epitopes which are in a form of short peptides bound to the major histocompatibility complex (MHC). When T-cell receives the message about the peptides bound to MHC, it makes the immune system active and results in the disposal of the immunogen. The antigenic determinants recognized and bound by the T-cell receptor is known as T-cell epitope. The accurate prediction of T-cell epitopes is crucial for vaccine development and clinical immunology. For the first time we developed new models using least squares support vector machine (LSSVM) and amino acid properties for T-cell epitopes prediction. A dataset including 203 short peptides (167 non-epitopes and 36 epitopes) was used as the input dataset and it was randomly divided into a training set and a test set. The models based on LSSVM and amino acid properties were evaluated using leave-one-out cross-validation method and the predictive ability of the test set, and obtained the results of 0.9875 and 0.9734 under the ROC curves, respectively. This result is more satisfactory than that were reported before. Especially, the accuracy of true positive gets a marked enhancement.     

Toward prediction of class II mouse major histocompatibility complex peptide binding affinity: in silico bioinformatic evaluation using partial least squares, a robust multivariate statistical technique

The accurate identification of T-cell epitopes remains a principal goal of bioinformatics within immunology. As the immunogenicity of peptide epitopes is dependent on their binding to major histocompatibility complex (MHC) molecules, the prediction of binding affinity is a prerequisite to the reliable prediction of epitopes. The iterative self-consistent (ISC) partial-least-squares (PLS)-based additive method is a recently developed bioinformatic approach for predicting class II peptide-MHC binding affinity. The ISC-PLS method overcomes many of the conceptual difficulties inherent in the prediction of class II peptide-MHC affinity, such as the binding of a mixed population of peptide lengths due to the open-ended class II binding site. The method has applications in both the accurate prediction of class II epitopes and the manipulation of affinity for heteroclitic and competitor peptides. The method is applied here to six class II mouse alleles (I-Ab, I-Ad, I-Ak, I-As, I-Ed, and I-Ek) and included peptides up to 25 amino acids in length. A series of regression equations highlighting the quantitative contributions of individual amino acids at each peptide position was established. The initial model for each allele exhibited only moderate predictivity. Once the set of selected peptide subsequences had converged, the final models exhibited a satisfactory predictive power. Convergence was reached between the 4th and 17th iterations, and the leave-one-out cross-validation statistical terms--q2, SEP, and NC--ranged between 0.732 and 0.925, 0.418 and 0.816, and 1 and 6, respectively. The non-cross-validated statistical terms r2 and SEE ranged between 0.98 and 0.995 and 0.089 and 0.180, respectively. The peptides used in this study are available from the AntiJen database (http://www.jenner.ac.uk/AntiJen). The PLS method is available commercially in the SYBYL molecular modeling software package. The resulting models, which can be used for accurate T-cell epitope prediction, will be made freely available online (http://www.jenner.ac.uk/MHCPred).     

Recovery of known T-cell epitopes by computational scanning of a viral genome

A new computational method (EpiDock) is proposed for predicting peptide binding to class I MHC proteins, from the amino acid sequence of any protein of immunological interest. Starting from the primary structure of the target protein, individual three-dimensional structures of all possible MHC-peptide (8-, 9- and 10-mers) complexes are obtained by homology modelling. A free energy scoring function (Fresno) is then used to predict the absolute binding free energy of all possible peptides to the class I MHC restriction protein. Assuming that immunodominant epitopes are usually found among the top MHC binders, the method can thus be applied to predict the location of immunogenic peptides on the sequence of the protein target. When applied to the prediction of HLA-A^*0201-restricted T-cell epitopes from the Hepatitis B virus, EpiDock was able to recover 92% of known high affinity binders and 80% of known epitopes within a filtered subset of all possible nonapeptides corresponding to about one tenth of the full theoretical list.The proposed method is fully automated and fast enough to scan a viral genome in less than an hour on a parallel computing architecture. As it requires very few starting experimental data, EpiDock can be used: (i) to predict potential T-cell epitopes from viral genomes (ii) to roughly predict still unknown peptide binding motifs for novel class I MHC alleles.     

A combination of epitope prediction and molecular docking allows for good identification of MHC class I restricted T-cell epitopes

In silico identification of T-cell epitopes is emerging as a new methodology for the study of epitope-based vaccines against viruses and cancer. In order to improve accuracy of prediction, we designed a novel approach, using epitope prediction methods in combination with molecular docking techniques, to identify MHC class I restricted T-cell epitopes. Analysis of the HIV-1 p24 protein and influenza virus matrix protein revealed that the present approach is effective, yielding prediction accuracy of over 80% with respect to experimental data. Subsequently, we applied such a method for prediction of T-cell epitopes in SARS coronavirus (SARS-CoV) S, N and M proteins. Based on available experimental data, the prediction accuracy is up to 90% for S protein. We suggest the use of epitope prediction methods in combination with 3D structural modelling of peptide-MHC-TCR complex to identify MHC class I restricted T-cell epitopes for use in epitope based vaccines like HIV and human cancers, which should provide a valuable step forward for the design of better vaccines and may provide in depth understanding about activation of T-cell epitopes by MHC binding peptides.     

Prediction of T-cell epitopes using biosupport vector machines

The immune system is concerned with the recognition and disposal of foreign or "non self" molecules or cells that enter the body of an immunologically competent individual. The generation of an immune response depends on the interaction of components, namely, the immunogen (nonself or foreign cell or molecule), antibody producing humoral immune system, and sensitized lymphocyte producing cellular immune system. An immunogen possesses surface structures referred to as epitopes; the precise pattern of each epitope enables an individual's immune system to recognize cells or molecules as self or immunogens. During the recognition process, the specific cells known as macrophages identify the epitope structures on the immunogen and save them in the form of short peptides 10-18 amino-acids-long known as immune dominant peptides (IDPs). IDPs are then bound with surface proteins on macrophages known as MHC protein complexes. The macrophages then present this IDP-MHC complex to a T cell that possesses a specific receptor that is specific for the foreign epitope on the IDP bound to MHC complex. This initiates an immune system cascade that results in the disposal of the immunogen. The study and accurate prediction of T-cell epitopes is, thus, very important for designing vaccines against pathogenic diseases. The present study applied the newly developed biosupport vector machine to the T-cell epitope data. This new algorithm introduces a biobasis function into the conventional support vector machines so that the nonnumerical attributes (amino acids) in protein sequences can be recognized without a feature extraction process, which often fails to properly code the biological content in protein sequences. The prediction accuracy of a 10-fold cross validation is 90.31%, compared with 87.86% using support vector machines reported as the best compared with other algorithms in an earlier study.     

A critical cross-validation of high throughput structural binding prediction methods for pMHC

T-cells recognize antigens via their T-cell receptors. The major histocompatibility complex (MHC) binds antigens in a specific way, transports them to the surface and presents the peptides to the TCR. Many in silico approaches have been developed to predict the binding characteristics of potential T-cell epitopes (peptides), with most of them being based solely on the amino acid sequence. We present a structural approach which provides insights into the spatial binding geometry. We combine different tools for side chain substitution (threading), energy minimization, as well as scoring methods for protein/peptide interfaces. The focus of this study is on high data throughput in combination with accurate results. These methods are not meant to predict the accurate binding free energy but to give a certain direction for the classification of peptides into peptides that are potential binders and peptides that definitely do not bind to a given MHC structure. In total we performed approximately 83,000 binding affinity prediction runs to evaluate interactions between peptides and MHCs, using different combinations of tools. Depending on the tools used, the prediction quality ranged from almost random to around 75% of accuracy for correctly predicting a peptide to be either a binder or a non-binder. The prediction quality strongly depends on all three evaluation steps, namely, the threading of the peptide, energy minimization and scoring.     

基于岭回归和SVM的高维特征选择与肽QSAR建模

岭回归估计权重绝对值在一定程度上体现了对应特征作用大小, 据此发展了基于岭回归(RR)和支持向量机(SVM)的高维特征选择算法. 对苦味二肽(BTT)和细胞毒性T淋巴细胞(CTL)表位9 肽两个肽体系, 以氨基酸的531 个物理化学性质参数直接表征肽结构, 各获得1062、4779 个初始特征; 对训练集, 初始特征以岭回归排序后序贯引入, 当SVM留一法交叉测试(LOOCV)的均方误差(MSE)显著上扬时终止, 最后以多轮末尾淘汰进一步精筛, 分别获得7、18个物理化学意义明确的保留特征. 基于保留特征与支持向量回归(SVR), 对训练集建立定量构效关系(QSAR)模型, 预测独立测试集, 其拟合精度、留一法交叉测试精度、独立预测精度均优于现有文献报道结果. 新方法运行速度快, 选取的特征物理化学意义明确, 解释性强, 在肽、蛋白质定量构效关系建模等高维数据回归预测领域有较广泛应用前景.     

New horizons in mouse immunoinformatics: reliable in silico prediction of mouse class I histocompatibility major complex peptide binding affinity

Quantitative structure-activity relationship (QSAR) analysis is a main cornerstone of modern informatic disciplines. Predictive computational models, based on QSAR technology, of peptide-major histocompatibility complex (MHC) binding affinity have now become a vital component of modern day computational immunovaccinology. Historically, such approaches have been built around semi-qualitative, classification methods, but these are now giving way to quantitative regression methods. The additive method, an established immunoinformatics technique for the quantitative prediction of peptide-protein affinity, was used here to identify the sequence dependence of peptide binding specificity for three mouse class I MHC alleles: H2-D(b), H2-K(b) and H2-K(k). As we show, in terms of reliability the resulting models represent a significant advance on existing methods. They can be used for the accurate prediction of T-cell epitopes and are freely available online ( http://www.jenner.ac.uk/MHCPred).     

SVM and SVR-based MHC-binding prediction using a mathematical presentation of peptide sequences

At present, there are a number of methods for the prediction of T-cell epitopes and major histocompatibility complex (MHC)-binding peptides. Despite numerous methods for predicting T-cell epitopes, there still exist limitations that affect the reliability of prevailing methods. For this reason, the development of models with high accuracy are crucial. An accurate prediction of the peptides that bind to specific major histocompatibility complex class I and II (MHC-I and MHC-II) molecules is important for an understanding of the functioning of the immune system and the development of peptide-based vaccines. Peptide binding is the most selective step in identifying T-cell epitopes. In this paper, we present a new approach to predicting MHC-binding ligands that takes into account new weighting schemes for position-based amino acid frequencies, BLOSUM and VOGG substitution of amino acids, and the physicochemical and molecular properties of amino acids. We have made models for quantitatively and qualitatively predicting MHC-binding ligands. Our models are based on two machine learning methods support vector machine (SVM) and support vector regression (SVR), where our models have used for feature selection, several different encoding and weighting schemes for peptides. The resulting models showed comparable, and in some cases better, performance than the best existing predictors. The obtained results indicate that the physicochemical and molecular properties of amino acids (AA) contribute significantly to the peptide-binding affinity.     

In silico epitope identification of unique multidrug resistance proteins from Salmonella Typhi for vaccine development

Typhoid fever is a multisystemic illness caused by Salmonella enterica serovars Typhi and is resistant to most antibiotics and drugs. The resistance is conferred through multidrug resistance (MDR) proteins, which efflux most antibiotics and other drugs. We predicted potential candidate B-cell and T-cell epitopes using bio- and immune-informatics tools in the 11 MDR proteins - EmrA, EmrB, EmrD, MdtA, MdtB, MdtC, MdtG, MdtH, MdtK, MdtL and TolC. The antigenic potential of the MDR proteins was calculated using VaxiJen server. The B-cell and T-cell epitopes of the MDR proteins were predicted using BCPred and ProPredI and ProPred respectively. The binding affinities of the predicted T-cell epitopes were estimated using T-epitope designer and MHCPred tools. 10, 7, 5, 12, 14, 21, 26, 3, 3 and 3 B-cell epitopes were identified in EmrA, EmrB, EmrD, TolC, MdtA, MdtB, MdtC, MdtG, MdtH and MdtL respectively. We predicted 9 T-cell epitopes - YVSRRAVQP (EmrA), FGVANAISI (EmrB), MVNSQVKQA and YQGGMVNSQ (TolC), WDRTNSHKL (MdtA), FLRNIPTAI (MdtB), YVEQLGVTG (MdtG), VKWMYAIEA (MdtH) and LAHTNTVTL (MdtL) capable of eliciting both humoral and adaptive immune responses. These T-cell epitopes specifically bind to HLA alleles - DRB1*0101 and DRB1*0401. This is the first report of epitope prediction in the MDR proteins of S. Typhi. Taken together, these results indicate the MDR proteins – EmrA, MdtA and TolC are the most suitable vaccine candidates for S. Typhi. The findings of our study on the MDR proteins prove to be useful in the development of peptide-based vaccine for the prevention and/or treatment of typhoid fever.     

In silico identification of outer membrane protein (Omp) and subunit vaccine design against pathogenic Vibrio cholerae

Virulence-related outer membrane proteins (Omps) are expressed in bacteria (Gram-negative) such as V. cholerae and are vital to bacterial invasion in to eukaryotic cell and survival within macrophages that could be best candidate for development of vaccine against V. cholerae. Applying in silico approaches, the 3-D model of the Omp was developed using Swiss model server and validated byProSA and Procheck web server. The continuous stretch of amino acid sequences 26 mer: RTRSNSGLLTWGDKQTITLEYGDPAL and 31 mer: FFAGGDNNLRGYGYKSISPQDASGALTGAKY having B-cell binding sites were selected from sequence alignment after B cell epitopes prediction by BCPred and AAP prediction modules of BCPreds. Further, the selected antigenic sequences (having B-cell epitopes) were analyzed for T-cell epitopes (MHC I and MHC II alleles binding sequence) by using ProPred 1 and ProPred respectively. The epitope (9 mer: YKSISPQDA) that binds to both the MHC classes (MHC I and MHC II) and covers maximum MHC alleles were identified. The identified epitopes can be useful in designing comprehensive peptide vaccine development against V. cholerae by inducing optimal immune response.     

T-cell epitopes of the La/SSB autoantigen: prediction based on the homology modeling of HLA-DQ2/DQ7 with the insulin-B peptide/HLA-DQ8 complex

T-cell epitopes are important components of the inappropriate response of the immune system to self-proteins in autoimmune diseases. In this study, the candidate T-cell epitopes of the La/SSB autoantigen, the main target of the autoimmune response in patients with Sjogren's Syndrome (SS), and Systemic Lupus Erythematosus (SLE) were predicted using as a template the HLA-DQ2 and DQ7 molecules, which are genetically linked to patients with SS and SLE. Modeling of DQ2 and DQ7 was based on the crystal structure of HLA-DQ8, an HLA molecule of high risk factor of type I diabetes, which is also an autoimmune disease. The quality and reliability of the modeled DQ2 and DQ7 was confirmed by the Ramachandran plot and the TINKER molecular modeling software. Common and/or similar candidate T-cell epitopes, obtained by comparing three different approaches the Taylor's sequence pattern, the TEPITOPE quantitative matrices, and the MULTIPRED artificial neural network, were subjected to homology modeling with the crystal structure of the insulin-B peptide complexed with HLA-DQ8, and the best superposed candidate epitopes were placed into the modeled HLA-DQ2 and DQ7 binding grooves to perform energy minimization calculations. Six T-cell epitopes were predicted for HLA-DQ7 and nine for HLA-DQ2 covering parts of the amino-terminal and the central regions of the La/SSB autoantigen. Residues corresponding to the P1, P4, and P9 pockets of the HLA-DQ2 and DQ7 binding grooves experience very low SASA because they are less exposed to the microenvironment of the groove. The proposed T-cell epitopes complexed with HLA-DQ2/DQ7 were further evaluated for their binding efficiency according to their potential interaction energy, binding affinity, and IC50 values. Our approach constitutes the ground work for a rapid and reliable experimentation concerning the T-cell epitope mapping of autoantigens, and could lead to the development of T-cell inhibitors as immunotherapeutics in autoimmune diseases.     

Studies on peptides. CLXVII. Solid-phase syntheses and immunological properties of fragment peptides related to human malaria circumsporozoite protein

A glycine-linked tetramer of Asn-Ala-Asn-Pro, a tandem repeated sequence of malaria circumsporozoite (CS) protein, was synthesized by the Boc-based solid phase method, followed by deprotection with 1 M trimethylsilyl trifluoromethanesulfonate-thioanisole in trifluoroacetic acid. In addition, three tetramer-related peptides were similarly synthesized, i.e., a 34-residue peptide [linked with TH, a proposed T-cell epitope of CS, at the C-terminus of the tetramer], a 46-residue peptide and a 59-residue peptide [linked with HA or HA', two proposed T-cell epitopes of influenza hemagglutinin protein, at the N-terminus of the above 34-residue peptide]. Their immunological properties were examined by enzyme-linked immunosorbent assay, for which three different congenic strains of mouse were used to raise the specific antibodies. Despite conjugation of T-cell epitopes to the tetramer, the mice of low-responder strains to the tetramer failed to produce any antibody specific to the tetramer. However, with the aid of recombinant interleukin 2 as an adjuvant, the low-responder mice produced antibody with relatively high titers.     

Immunology of O-glycosylated proteins: approaches to the design of a MUC1 glycopeptide-based tumor vaccine

Until about 1990 there was general consent about the assumption that only protein and peptide antigens have the capacity of CD4(+) or CD8(+) T-cell stimulation. Since about ten years evidence is now accumulating that carbohydrate-peptide epitopes do play a role in classical MHC-mediated immune responses. This holds true for glycopeptides, where the glycan chain is short and not located at an "anchor residue" needed for MHC interaction. T-cell recognition of O-glycosylated peptides is potentially of high biomedical significance, because it can mediate the immune protection against microorganisms, the vaccination in anti-tumor therapies, but also some aspects of autoimmunity. The epithelial type 1 transmembrane mucin MUC1 is established as a marker for monitoring recurrence of breast cancer and is a promising target for immunotherapeutic strategies to treat cancer by active specific immunization. Natural human immune responses to the tumor-associated glycoforms of the mucin indicate that antibody reactivities are more directed to glycopeptide than to non-glycosylated peptide epitopes. To overcome the weak immunogenicity of the natural target, heavily O-glycosylated MUC1, the question was addressed whether O-linked glycans remain intact during processing in the MHC class II pathway and interfere with endosomal processing and peptide presentation. Attempts were made to define on a biochemical level the structural requirements for an efficient endosomal proteolysis catalyzed by cathepsin L in antigen-presenting cells. Evidence based on work with CD4(+) T-hybridomas confirms that O-glycopeptides can be effectively presented to T-cells and that glycans can form integral parts of the TCR defined epitopes. Similar approaches are currently followed in the MHC class I pathway which aim at the identification of immunogenic glycopeptides generated by immunoproteasomes.     

A totally synthetic, self-assembling, adjuvant-free MUC1 glycopeptide vaccine for cancer therapy

In the development of vaccines for epithelial tumors, the key targets are MUC1 proteins, which have a variable number of tandem repeats (VNTR) bearing tumor-associated carbohydrate antigens (TACAs), such as Tn and STn. A major obstacle in vaccine development is the low immunogenicity of the short MUC1 peptide. To overcome this obstacle, we designed, synthesized, and evaluated several totally synthetic self-adjuvanting vaccine candidates with self-assembly domains. These vaccine candidates aggregated into fibrils and displayed multivalent B-cell epitopes under mild conditions. Glycosylation of Tn antigen on the Thr residue of PDTRP sequence in MUC1 VNTR led to effective immune response. These vaccines elicited a high level antibody response without any adjuvant and induced antibodies that recognized human breast tumor cells. These vaccines appeared to act through a T-cell independent pathway and were associated with the activation of cytotoxic T cells. These fully synthetic, molecularly defined vaccine candidates had several features that hold promise for anticancer therapy.     

3D-Epitope-Explorer (3DEX): localization of conformational epitopes within three-dimensional structures of proteins

Neutralizing antibodies often recognize conformational, discontinuous epitopes. Linear peptides mimicking such conformational epitopes can be selected from phage display peptide libraries by screening with the respective antibodies. However, it is difficult to localize these "mimotopes" within the three-dimensional (3D) structures of the target proteins. Knowledge of conformational epitopes of neutralizing antibodies would help to design antigens able to elicit protective immune responses. Therefore, we provide here a software that allows to localize linear peptide sequences within 3D structures of proteins. The 3D-Epitope-Explorer (3DEX) software allows to map conformational epitopes in 3D protein structures based on an algorithm that takes into account the physicochemical neighborhood of C(alpha)- or C(beta)-atoms of individual amino acids. A given amino acid of a peptide sequence is localized within the protein and the software searches within predefined distances for the amino acids neighboring that amino acid in the peptide. Surface exposure of the amino acids can also be taken into consideration. The procedure is then repeated for the remaining amino acids of the peptide. The introduction of a joker function allows to map peptide mimotopes, which do not necessarily have 100% sequence homology to the protein. Using this software we were able to localize mimotopes selected from phage displayed peptide libraries with polyclonal antibodies from HIV-positive patient plasma within the 3D structure of gp120, the exterior glycoprotein of HIV-1. We also analyzed two recently published peptide sequences corresponding to known conformational epitopes to further confirm the integrity of 3DEX.     

多肽序列的结构特征与其MHC限制性

从多肽序列的一级结构出发,基于多肽序列中氨基酸侧链间距离和氨基酸侧链的电性特征,构建了多肽序列特征矢量,简称ζ矢量,选取文献中主要组织相容性复合物(MHC)中的14个A^d限制性和14个非A^d限制性多肽序列作为训练集建立辅助性T细胞(helperTlymphocytc,Th)表位预测的定量模型,为了检验该预报系统的精确性,进行了随机抽样检验和交互检验.结果表明,该预报系统具有稳定性好、预测能力强的特点,该方法可用于人的MHCⅠ和Ⅱ类表位的预测、蛋白质抗原免疫识别、亚单位疫苗分子设计及研制.     

Prediction of protein structural class based on symmetrical recurrence quantification analysis

Protein structural class prediction for low similarity sequences is a significant challenge and one of the deeply explored subjects. This plays an important role in drug design, folding recognition of protein, functional analysis and several other biology applications. In this paper, we worked with two benchmark databases existing in the literature (1) 25PDB and (2) 1189 to apply our proposed method for predicting protein structural class. Initially, we transformed protein sequences into DNA sequences and then into binary sequences. Furthermore, we applied symmetrical recurrence quantification analysis (the new approach), where we got 8 features from each symmetry plot computation. Moreover, the machine learning algorithms such as Linear Discriminant Analysis (LDA), Random Forest (RF) and Support Vector Machine (SVM) are used. In addition, comparison was made to find the best classifier for protein structural class prediction. Results show that symmetrical recurrence quantification as feature extraction method with RF classifier outperformed existing methods with an overall accuracy of 100% without overfitting.     

Synthesis and biological evaluation of two chemically modified peptide epitopes for the class I MHC protein HLA-B*2705

The T-cell receptor of a CD8(+) T-cell recognises peptide epitopes bound by class I major histocompatibility complex (MHC) glycoproteins presented in a groove on their upper surface. Within the groove of the MHC molecule are 6 pockets, two of which mostly display a high degree of specificity for binding amino acids capable of making conserved and energetically favourable contacts with the MHC. One type of MHC molecule, HLA-B*2705, preferentially binds peptides containing an arginine at position 2. In an effort to increase the affinity of peptides for HLA-B*2705, potentially leading to better immune responses to such a peptide, we synthesised two modified epitopes where the amino acid at position 2 involved in anchoring the peptide to the class I molecule was replaced with the alpha-methylated beta,gamma-unsaturated arginine analogue 2-(S)-amino-5-guanidino-2-methyl-pent-3-enoic acid. The latter was prepared via a multi-step synthetic sequence, starting from alpha-methyl serine, and incorporated into dipeptides which were fragment-coupled to resin-bound heptameric peptides yielding the target nonameric sequences. Biological characterisation indicated that the modified peptides were poorer than the native peptides at stabilising empty class I MHC complexes, and cells sensitised with these peptides were not recognised as well by cognate CD8(+) T-cells, where available, compared to those sensitised with the native peptide. We suggest that the modifications made to the peptide have decreased its ability to bind to the peptide binding groove of HLA-B*2705 molecules which may explain the decrease in recognition by cytotoxic T-cells when compared to the native peptide.