1 - 37 of 37 results


Extrapolates from variants with known binding specificity to those where no experimental data are available. PickPocket uses information derived from similar protein structures to infer which residues in the major histocompatibility complex (MHC) molecule may interact with. It appears to be very useful in often encountered situations where there is not enough data available for conventional data-driven approaches to succeed. The tool achieves higher predictive performance values than NetMHCpan for alleles that were distant to any MHC molecule with characterized binding specificity.


Predicts peptide–major histocompatibility complex (MHC)-I complex stability. NetMHCstabpan is based on neural network pan-specific method. It allows the user to include affinity predictions. The tool can enhance the ability to predict T cell peptide immunogenicity. It was evaluated on more than 25,000 quantitative stability data covering 75 different human leukocyte antigen (HLA) molecules. The results show that the tool significantly outperforms the stability prediction.


Predicts binding of peptides to any known major histocompatibility complex (MHC) class I molecule. NetMHCcons is a server that allows the user in an automatic manner to obtain the most accurate predictions for any given MHC molecule. This method provides affinity predictions for any peptide of length 8–11 amino acids to any given MHC class I molecule of known protein sequence. It also provides a possibility for the user to choose MHC molecule in question from a list of alleles or alternatively upload the MHC protein sequence of interest.


A novel and efficient epitope prediction method that predicts epitopes for any Class-1 HLA allele, by estimating the binding strengths of peptide-HLA complexes which is achieved through learning pair-potentials important for peptide binding. It relies on the strength of the mechanistic understanding of peptide-HLA recognition and provides an estimate of the total ligand space for each allele. HLaffy has four distinct computational modules, a) structural modelling, estimating statistical pair-potentials and constraint derivation, b) implicit modelling and interaction profiling, c) feature representation and binding affinity prediction and d) use of graphical models to extract peptide sequence signatures to predict epitopes for HLA class I alleles. The performance of HLaffy is seen to be superior to the currently available methods.


Decouples global and local conformational searches and generates diverse native-like peptide conformations for peptide binding to major histocompatibility complex class I (pMHC-I). GradDock uses a ranking function specific for pMHC-I which was reconstructed from the Rosetta scoring terms. It provides three steps: (1) the generation of peptide candidates from the sequence; (2) the insertion of peptides into the MHC-I molecule; and (3) the ranking of the peptides. It can create native-like peptides with a certain degree of diversity, which is sufficient for describing the pMHC-I space.

HLA-CNN / Human Leukocyte Antigen-Convolutional Neural Network

Extracts a vector space distributed representation of amino acids from Human Leukocyte Antigen (HLA) peptide data that preserved property critical for covalent bonding. HLA-CNN is a deep convolutional neural network architecture for the task of HLA class I-peptide binding prediction. HLA-CNN architecture achieves state-of-the-art results in the vast majority of the latest two Immune Epitope Database (IEDB) weekly automated benchmark datasets.


In order to see potential off-target recognition when designing new lead targets, until now one needed to search protein databases for approximate hits and than evaluate each hit for its potential to be an epitope. The Expitope web server combines all these searches and evaluation in one place and even reports the expression of the associated transcripts in all vital human tissues to facilitate TCR selection. This framework is a helpful tool to exclude potential cross-reactivity in the early stage of TCR selection for use in design of adoptive T cell immunotherapy.

NIELuter / Natural Immunopeptidome Eluter

Predicts peptides eluted from six human leukocyte antigen (HLA) alleles (A0201, B0702, B3501, B4403, B5301, and B5701) of 8–11 amino acids. NIELuter is based on a combination of five support vector machine (SVM) models trained with position-specific amino acid composition, position-specific dipeptide composition, Hidden Markov Model, binary encoding, and BLOSUM62 feature. This tool is able to produce good results in 7 out of 24 groups of HLA allele-peptide length combinations.


Predicts peptide-major histocompatibility complex class-I (pMHC-I) binding affinities. MHCflurry contains (1) an embedding layer which transforms amino acids to learned vector representations, (2) a single hidden layer with tanh nonlinearity, and (3) a sigmoidal scalar output. It maps input peptides to a 32-dimensional space, which then feeds into a fully connected layer. MHCflurry can augment its training data with peptides randomly generated in silico from a uniform distribution.


Ranks all possible peptides from an input protein using the PSSM coefficients. The predictive power of the method was evaluated by running RANKPEP on proteins known to bear MHCI K(b)- and D(b)-restricted T-cell epitopes. Analysis of the results indicates that > 80% of these epitopes are among the top 2% of scoring peptides. In addition, the RANKPEP server also allows the user to enter additional profiles, making the server a powerful and versatile computational biology benchmark for the prediction of peptide-MHC binding.

IMS / Immunogenetic Management Software

Permits multiplexed analysis of complex immunogenetic traits that are necessary for the accurate planning and execution of experiments involving large animal models, including nonhuman primates. IMS is capable of housing complex pedigree relationships, microsatellite-based major histocompatibility complex (MHC) typing data, as well as MHC pyrosequencing expression analysis of class I alleles. It includes an automated MHC haplotype naming algorithm and has accomplished an innovative visualization protocol that allows users to view multiple familial and MHC haplotype relationships through a single, interactive graphical interface. Detailed DNA and RNA-based data can also be queried and analyzed in a highly accessible fashion, and flexible search capabilities allow experimental choices to be made based on multiple, individualized and expandable immunogenetic factors. This web application is implemented in Java, MySQL, Tomcat, and Apache, with supported browsers including Internet Explorer and Firefox on Windows and Safari on Mac OS.


An accurate MHC binders prediction method for the large number of class I MHC alleles. nHLAPred allows to identify the promiscuous MHC class I binders (peptides that can bind to large number of alleles) having proteasomal cleavage site at C-terminus. This leads to identification of MHC class I restricted T cell epitopes in an antigen sequence. The server is partitioned in two parts ComPred and ANNpred. In ComPred part the prediction is based on the hybrid approach of Quantitative matrices and artificial neural network. In ANNPred the prediction is solely based on artificial neural network.

Epitope Prediction

Predicts HLA (Human Leukocyte Antigen) class I restricted cytotoxic T lymphocytes (CTL) epitopes. Epitope Prediction which is based on logistic regression, is simple to implement. It is solved by finding a single global maximum. In contrast to almost all other work in this area, a single model was trained on epitopes from all HLA alleles and supertypes. Epitope Prediction is therefore able to leverage data across all HLA alleles and/or their supertypes, automatically learning what information should be shared and also how to combine allelespecific, supertype-specific, and global information in a principled way.

HLA Assignment

Provides a statistical model for identifying Human Leukocyte Antigen (HLA)-restricted epitopes from ELISpot data. By looking at patterns across a broad range of donors, HLA Assignment can determine (probabilistically) which of the HLA alleles are likely to be responsible for the observed reactivities. Additionally, it can provide a good estimate of the number of false positives generated by our analysis (i.e., the false discovery rate). This model allows users to learn about new HLA-restricted epitopes from ELISpot data in an efficient, cost-effective, and high-throughput manner. This approach was applied to data from donors infected with Human Immunodeficiency Virus (HIV) and identified many potential new HLA restrictions. Among 134 such predictions, six were confirmed in the lab and the remainder could not be ruled as invalid. These results shed light on the extent of HLA class I promiscuity, which has significant implications for the understanding of HLA class I antigen presentation and vaccine development.