BEST: improved prediction of B-cell epitopes from antigen sequences

Accurate identification of immunogenic regions in a given antigen chain is a difficult and actively pursued problem. Although accurate predictors for T-cell epitopes are already in place, the prediction of the B-cell epitopes requires further research. We overview the available approaches for the prediction of B-cell epitopes and propose a novel and accurate sequence-based solution. Our BEST (B-cell Epitope prediction using Support vector machine Tool) method predicts epitopes from antigen sequences, in contrast to some method that predict only from short sequence fragments, using a new architecture based on averaging selected scores generated from sliding 20-mers by a Support Vector Machine (SVM). The SVM predictor utilizes a comprehensive and custom designed set of inputs generated by combining information derived from the chain, sequence conservation, similarity to known (training) epitopes, and predicted secondary structure and relative solvent accessibility. Empirical evaluation on benchmark datasets demonstrates that BEST outperforms several modern sequence-based B-cell epitope predictors including ABCPred, method by Chen et al. (2007), BCPred, COBEpro, BayesB, and CBTOPE, when considering the predictions from antigen chains and from the chain fragments. Our method obtains a cross-validated area under the receiver operating characteristic curve (AUC) for the fragment-based prediction at 0.81 and 0.85, depending on the dataset. The AUCs of BEST on the benchmark sets of full antigen chains equal 0.57 and 0.6, which is significantly and slightly better than the next best method we tested. We also present case studies to contrast the propensity profiles generated by BEST and several other methods.     

Stepwise prediction of conformational discontinuous B-cell epitopes using the Mapitope algorithm

Mapping the epitope of an antibody is of great interest, since it contributes much to our understanding of the mechanisms of molecular recognition and provides the basis for rational vaccine design. Here we present Mapitope, a computer algorithm for epitope mapping. The algorithm input is a set of affinity isolated peptides obtained by screening phage display peptide-libraries with the antibody of interest. The output is usually 1-3 epitope candidates on the surface of the atomic structure of the antigen. We have systematically tested the performance of Mapitope by assessing the effect of the algorithm parameters on the final prediction. Thus, we have examined the effect of the statistical threshold (ST) parameter, relating to the frequency distribution and enrichment of amino acid pairs from the isolated peptides and the D (distance) and E (exposure) parameters which relate to the physical parameters of the antigen. Two model systems were analyzed in which the antibody of interest had previously been co-crystallized with the antigen and thus the epitope is a given. The Mapitope algorithm successfully predicted the epitopes in both models. Accordingly, we formulated a stepwise paradigm for the prediction of discontinuous conformational epitopes using peptides obtained from screening phage display libraries. We applied this paradigm to successfully predict the epitope of the Trastuzumab antibody on the surface of the Her-2/neu receptor in a third model system.     

Identification of conformational B-cell Epitopes in an antigen from its primary sequence

Background

One of the major challenges in the field of vaccine design is to predict conformational B-cell epitopes in an antigen. In the past, several methods have been developed for predicting conformational B-cell epitopes in an antigen from its tertiary structure. This is the first attempt in this area to predict conformational B-cell epitope in an antigen from its amino acid sequence.

Results

All Support vector machine (SVM) models were trained and tested on 187 non-redundant protein chains consisting of 2261 antibody interacting residues of B-cell epitopes. Models have been developed using binary profile of pattern (BPP) and physiochemical profile of patterns (PPP) and achieved a maximum MCC of 0.22 and 0.17 respectively. In this study, for the first time SVM model has been developed using composition profile of patterns (CPP) and achieved a maximum MCC of 0.73 with accuracy 86.59%. We compare our CPP based model with existing structure based methods and observed that our sequence based model is as good as structure based methods.

Conclusion

This study demonstrates that prediction of conformational B-cell epitope in an antigen is possible from is primary sequence. This study will be very useful in predicting conformational B-cell epitopes in antigens whose tertiary structures are not available. A web server CBTOPE has been developed for predicting B-cell epitope http://www.imtech.res.in/raghava/cbtope/.     

Machine learning approaches for prediction of linear B-cell epitopes on proteins

Identification and characterization of antigenic determinants on proteins has received considerable attention utilizing both, experimental as well as computational methods. For computational routines mostly structural as well as physicochemical parameters have been utilized for predicting the antigenic propensity of protein sites. However, the performance of computational routines has been low when compared to experimental alternatives. Here we describe the construction of machine learning based classifiers to enhance the prediction quality for identifying linear B-cell epitopes on proteins. Our approach combines several parameters previously associated with antigenicity, and includes novel parameters based on frequencies of amino acids and amino acid neighborhood propensities. We utilized machine learning algorithms for deriving antigenicity classification functions assigning antigenic propensities to each amino acid of a given protein sequence. We compared the prediction quality of the novel classifiers with respect to established routines for epitope scoring, and tested prediction accuracy on experimental data available for HIV proteins. The major finding is that machine learning classifiers clearly outperform the reference classification systems on the HIV epitope validation set.     

Computational strategies for protein conformational ensemble detection

Protein function is constrained by the three-dimensional structure but is delineated by its dynamics. This framework must satisfy specificity of function along with adaptability to changing environments and evolvability under external constraints. The accessibility of the available regions of the energy landscape for a set of conditions and shifts in the populations upon their modulation have effects propagating across scales, from biomolecular interactions, to organisms, to populations. Developing the ability to detect and juggle protein conformations supplemented by a physics-based understanding has implications for not only in vivo problems but also for resistance impeding drug discovery and bionano-sensor design.     

Selection and combination of machine learning classifiers for prediction of linear B-cell epitopes on proteins

Recently, new machine learning classifiers for the prediction of linear B-cell epitopes were presented. Here we show the application of Receiver Operator Characteristics (ROC) convex hulls to select optimal classifiers as well as possibilities to improve the post test probability (PTP) to meet real world requirements such as high throughput epitope screening of whole proteomes. The major finding is that ROC convex hulls present an easy to use way to rank classifiers based on their prediction conservativity as well as to select candidates for ensemble classifiers when validating against the antigenicity profile of 10 HIV-1 proteins. We also show that linear models are at least equally efficient to model the available data when compared to multi-layer feed-forward neural networks.     

A meta-learning approach for B-cell conformational epitope prediction

Background

One of the major challenges in the field of vaccine design is identifying B-cell epitopes in continuously evolving viruses. Various tools have been developed to predict linear or conformational epitopes, each relying on different physicochemical properties and adopting distinct search strategies. We propose a meta-learning approach for epitope prediction based on stacked and cascade generalizations. Through meta learning, we expect a meta learner to be able integrate multiple prediction models, and outperform the single best-performing model. The objective of this study is twofold: (1) to analyze the complementary predictive strengths in different prediction tools, and (2) to introduce a generic computational model to exploit the synergy among various prediction tools. Our primary goal is not to develop any particular classifier for B-cell epitope prediction, but to advocate the feasibility of meta learning to epitope prediction. With the flexibility of meta learning, the researcher can construct various meta classification hierarchies that are applicable to epitope prediction in different protein domains.

Results

We developed the hierarchical meta-learning architectures based on stacked and cascade generalizations. The bottom level of the hierarchy consisted of four conformational and four linear epitope prediction tools that served as the base learners. To perform consistent and unbiased comparisons, we tested the meta-learning method on an independent set of antigen proteins that were not used previously to train the base epitope prediction tools. In addition, we conducted correlation and ablation studies of the base learners in the meta-learning model. Low correlation among the predictions of the base learners suggested that the eight base learners had complementary predictive capabilities. The ablation analysis indicated that the eight base learners differentially interacted and contributed to the final meta model. The results of the independent test demonstrated that the meta-learning approach markedly outperformed the single best-performing epitope predictor.

Conclusions

Computational B-cell epitope prediction tools exhibit several differences that affect their performances when predicting epitopic regions in protein antigens. The proposed meta-learning approach for epitope prediction combines multiple prediction tools by integrating their complementary predictive strengths. Our experimental results demonstrate the superior performance of the combined approach in comparison with single epitope predictors.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-014-0378-y) contains supplementary material, which is available to authorized users.     

Mapping of linear B-cell epitopes of hepatitis B surface antigen

Eight monoclonal antibodies directed against the surface protein of hepatitis B virus (HBV) were tested using an epitope-mapping system (Pepscan) for characterizing antigenic domains. Four different amino acid sequences corresponding to linear epitopes were identified: one in pre-S1 corresponding to the sequence 29–36, two in pre-S2 corresponding to overlapping sequences 134–141 and 137–144, and one in the S region of the protein corresponding to the amino acid sequence 117–126.     

Drug response prediction by ensemble learning and drug-induced gene expression signatures

Chemotherapeutic response of cancer cells to a given compound is one of the most fundamental information one requires to design anti-cancer drugs. Recently, considerable amount of drug-induced gene expression data has become publicly available, in addition to cytotoxicity databases. These large sets of data provided an opportunity to apply machine learning methods to predict drug activity. However, due to the complexity of cancer drug mechanisms, none of the existing methods is perfect. In this paper, we propose a novel ensemble learning method to predict drug response. In addition, we attempt to use the drug screen data together with two novel signatures produced from the drug-induced gene expression profiles of cancer cell lines. Finally, we evaluate predictions by in vitro experiments in addition to the tests on data sets. The predictions of the methods, the signatures and the software are available from http://mtan.etu.edu.tr/drug-response-prediction/.     

Drug-target interaction prediction via class imbalance-aware ensemble learning

Background

Multiple computational methods for predicting drug-target interactions have been developed to facilitate the drug discovery process. These methods use available data on known drug-target interactions to train classifiers with the purpose of predicting new undiscovered interactions. However, a key challenge regarding this data that has not yet been addressed by these methods, namely class imbalance, is potentially degrading the prediction performance. Class imbalance can be divided into two sub-problems. Firstly, the number of known interacting drug-target pairs is much smaller than that of non-interacting drug-target pairs. This imbalance ratio between interacting and non-interacting drug-target pairs is referred to as the between-class imbalance. Between-class imbalance degrades prediction performance due to the bias in prediction results towards the majority class (i.e. the non-interacting pairs), leading to more prediction errors in the minority class (i.e. the interacting pairs). Secondly, there are multiple types of drug-target interactions in the data with some types having relatively fewer members (or are less represented) than others. This variation in representation of the different interaction types leads to another kind of imbalance referred to as the within-class imbalance. In within-class imbalance, prediction results are biased towards the better represented interaction types, leading to more prediction errors in the less represented interaction types.

Results

We propose an ensemble learning method that incorporates techniques to address the issues of between-class imbalance and within-class imbalance. Experiments show that the proposed method improves results over 4 state-of-the-art methods. In addition, we simulated cases for new drugs and targets to see how our method would perform in predicting their interactions. New drugs and targets are those for which no prior interactions are known. Our method displayed satisfactory prediction performance and was able to predict many of the interactions successfully.

Conclusions

Our proposed method has improved the prediction performance over the existing work, thus proving the importance of addressing problems pertaining to class imbalance in the data.

     

Analysis and prediction of protective continuous B-cell epitopes on pathogen proteins

Background

The application of peptide based diagnostics and therapeutics mimicking part of protein antigen is experiencing renewed interest. So far selection and design rationale for such peptides is usually driven by T-cell epitope prediction, available experimental and modelled 3D structure, B-cell epitope predictions such as hydrophilicity plots or experience. If no structure is available the rational selection of peptides for the production of functionally altering or neutralizing antibodies is practically impossible. Specifically if many alternative antigens are available the reduction of required synthesized peptides until one successful candidate is found is of central technical interest. We have investigated the integration of B-cell epitope prediction with the variability of antigen and the conservation of patterns for post-translational modification (PTM) prediction to improve over state of the art in the field. In particular the application of machine-learning methods shows promising results.

Results

We find that protein regions leading to the production of functionally altering antibodies are often characterized by a distinct increase in the cumulative sum of three presented parameters. Furthermore the concept to maximize antigenicity, minimize variability and minimize the likelihood of post-translational modification for the identification of relevant sites leads to biologically interesting observations. Primarily, for about 50% of antigen the approach works well with individual area under the ROC curve (AROC) values of at least 0.65. On the other hand a significant portion reveals equivalently low AROC values of < = 0.35 indicating an overall non-Gaussian distribution. While about a third of 57 antigens are seemingly intangible by our approach our results suggest the existence of at least two distinct classes of bioinformatically detectable epitopes which should be predicted separately. As a side effect of our study we present a hand curated dataset for the validation of protectivity classification. Based on this dataset machine-learning methods further improve predictive power to a class separation in an equilibrated dataset of up to 83%.

Conclusion

We present a computational method to automatically select and rank peptides for the stimulation of potentially protective or otherwise functionally altering antibodies. It can be shown that integration of variability, post-translational modification pattern conservation and B-cell antigenicity improve rational selection over random guessing. Probably more important, we find that for about 50% of antigen the approach works substantially better than for the overall dataset of 57 proteins. Essentially as a side effect our method optimizes for presumably best applicable peptides as they tend to be likely unmodified and as invariable as possible which is answering needs in diagnosis and treatment of pathogen infection. In addition we show the potential for further improvement by the application of machine-learning methods, in particular Random Forests.     

Prediction of continuous B-cell epitopes in an antigen using recurrent neural network

B-cell epitopes play a vital role in the development of peptide vaccines, in diagnosis of diseases, and also for allergy research. Experimental methods used for characterizing epitopes are time consuming and demand large resources. The availability of epitope prediction method(s) can rapidly aid experimenters in simplifying this problem. The standard feed-forward (FNN) and recurrent neural network (RNN) have been used in this study for predicting B-cell epitopes in an antigenic sequence. The networks have been trained and tested on a clean data set, which consists of 700 non-redundant B-cell epitopes obtained from Bcipep database and equal number of non-epitopes obtained randomly from Swiss-Prot database. The networks have been trained and tested at different input window length and hidden units. Maximum accuracy has been obtained using recurrent neural network (Jordan network) with a single hidden layer of 35 hidden units for window length of 16. The final network yields an overall prediction accuracy of 65.93% when tested by fivefold cross-validation. The corresponding sensitivity, specificity, and positive prediction values are 67.14, 64.71, and 65.61%, respectively. It has been observed that RNN (JE) was more successful than FNN in the prediction of B-cell epitopes. The length of the peptide is also important in the prediction of B-cell epitopes from antigenic sequences. The webserver ABCpred is freely available at www.imtech.res.in/raghava/abcpred/.     

Prediction of residues in discontinuous B-cell epitopes using protein 3D structures

Discovery of discontinuous B-cell epitopes is a major challenge in vaccine design. Previous epitope prediction methods have mostly been based on protein sequences and are not very effective. Here, we present DiscoTope, a novel method for discontinuous epitope prediction that uses protein three-dimensional structural data. The method is based on amino acid statistics, spatial information, and surface accessibility in a compiled data set of discontinuous epitopes determined by X-ray crystallography of antibody/antigen protein complexes. DiscoTope is the first method to focus explicitly on discontinuous epitopes. We show that the new structure-based method has a better performance for predicting residues of discontinuous epitopes than methods based solely on sequence information, and that it can successfully predict epitope residues that have been identified by different techniques. DiscoTope detects 15.5% of residues located in discontinuous epitopes with a specificity of 95%. At this level of specificity, the conventional Parker hydrophilicity scale for predicting linear B-cell epitopes identifies only 11.0% of residues located in discontinuous epitopes. Predictions by the DiscoTope method can guide experimental epitope mapping in both rational vaccine design and development of diagnostic tools, and may lead to more efficient epitope identification.     

T and B-cell epitopes prediction of Iranian saffron (Crocus sativus) profilin by bioinformatics tools

Plant profilins form a well-known panallergen family responsible for cross-sensitization between plant foods and pollens. We sought to map T and B-cell epitopes on the Iranian Crocus sativus profilin by bioinformatics tools. The predicted peptides are useful for further vaccine development.     

Computational prediction of atomic structures of helical membrane proteins aided by EM maps

Integral membrane proteins pose a major challenge for protein-structure prediction because only approximately 100 high-resolution structures are available currently, thereby impeding the development of rules or empirical potentials to predict the packing of transmembrane alpha-helices. However, when an intermediate-resolution electron microscopy (EM) map is available, it can be used to provide restraints which, in combination with a suitable computational protocol, make structure prediction feasible. In this work we present such a protocol, which proceeds in three stages: 1), generation of an ensemble of alpha-helices by flexible fitting into each of the density rods in the low-resolution EM map, spanning a range of rotational angles around the main helical axes and translational shifts along the density rods; 2), fast optimization of side chains and scoring of the resulting conformations; and 3), refinement of the lowest-scoring conformations with internal coordinate mechanics, by optimizing the van der Waals, electrostatics, hydrogen bonding, torsional, and solvation energy contributions. In addition, our method implements a penalty term through a so-called tethering map, derived from the EM map, which restrains the positions of the alpha-helices. The protocol was validated on three test cases: GpA, KcsA, and MscL.     

Analysis and prediction of calcium-binding pockets from apo-protein structures exhibiting calcium-induced localized conformational changes

Calcium binding in proteins exhibits a wide range of polygonal geometries that relate directly to an equally diverse set of biological functions. The binding process stabilizes protein structures and typically results in local conformational change and/or global restructuring of the backbone. Previously, we established the MUG program, which utilized multiple geometries in the Ca²⁺‐binding pockets of holoproteins to identify such pockets, ignoring possible Ca²⁺‐induced conformational change. In this article, we first report our progress in the analysis of Ca²⁺‐induced conformational changes followed by improved prediction of Ca²⁺‐binding sites in the large group of Ca²⁺‐binding proteins that exhibit only localized conformational changes. The MUG^SR algorithm was devised to incorporate side chain torsional rotation as a predictor. The output from MUG^SR presents groups of residues where each group, typically containing two to five residues, is a potential binding pocket. MUG^SR was applied to both X‐ray apo structures and NMR holo structures, which did not use calcium distance constraints in structure calculations. Predicted pockets were validated by comparison with homologous holo structures. Defining a “correct hit” as a group of residues containing at least two true ligand residues, the sensitivity was at least 90%; whereas for a “correct hit” defined as a group of residues containing at least three true ligand residues, the sensitivity was at least 78%. These data suggest that Ca²⁺‐binding pockets are at least partially prepositioned to chelate the ion in the apo form of the protein.     

B-cell epitopes of scleroderma-specific autoantigens

Conclusions It can be concluded that the precise localization of the epitopes on autoantigens associated with scleroderma has not been determined yet, and further subcloning experiments will be required to map the epitopes more precisely. However, the fact that the antigenicities of the C-terminal ends of topo I as well as of CENP-B are highly affected by the length of the fusion segments suggest that most, if not all, antigenic determinants on these parts of the autoantigens are conformational epitopes. Studies based upon molecular modelling of antibodies reacting with antigens suggest that over 90% of B-cell epitopes are conformational [50]. This implies that the most successful approach to allocate B-cell epitopes on autoantigens in the near future may be the use of techniques for mapping conformational epitopes. Such techniques are currently being developed [reviewed in 51]. Until now, the limited data available indicate that the B-cell epitopes on the scleroderma-associated autoantigens are distributed over the entire proteins. The C-terminal parts of the antigens seem to be good candidates for harboring the major autoimmune epitopes, but more experimental data will be necessary to confirm this suggestion.     

Prediction of linear B-cell epitopes

The use of antigenicity scales based on physicochemical properties and the sliding window method in combination with an averaging algorithm and subsequent search for the maximum value is the classical method for B-cell epitope prediction. However, recent studies have demonstrated that the best classical methods provide a poor correlation with experimental data. We review both classical and novel algorithms and present our own implementation of the algorithms. The AAPPred software is available at http://www.bioinf.ru/aappred/.