CNN-PepPred: An open-source tool to create convolutional NN models for the discovery of patterns in peptide sets. Application to peptide-MHC class II binding prediction.
Ontology highlight
ABSTRACT: Summary
The ability to unveil binding patterns in peptide sets has important applications in several biomedical areas, including the development of vaccines. We present an open-source tool, CNN-PepPred, that uses convolutional neural networks to discover such patterns, along with its application to peptide-HLA class II binding prediction. The tool can be used locally on different operating systems, with CPUs or GPUs, to train, evaluate, apply and visualize models.Availability and implementation
CNN-PepPred is freely available as a Python tool with a detailed User's Guide at: https://github.com/ComputBiol-IBB/CNN-PepPred.Supplementary information
Supplementary data are available at Bioinformatics online.
SUBMITTER: Junet V
PROVIDER: S-EPMC8652105 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundThe major histocompatibility complex (MHC) molecule plays a central role in controlling the adaptive immune response to infections. MHC class I molecules present peptides derived from intracellular proteins to cytotoxic T cells, whereas MHC class II molecules stimulate cellular and humoral immunity through presentation of extracellularly derived peptides to helper T cells. Identification of which peptides will bind a given MHC molecule is thus of great importance for the understanding of host-pathogen interactions, and large efforts have been placed in developing algorithms capable of predicting this binding event.ResultsHere, we present a novel artificial neural network-based method, NN-align that allows for simultaneous identification of the MHC class II binding core and binding affinity. NN-align is trained using a novel training algorithm that allows for correction of bias in the training data due to redundant binding core representation. Incorporation of information about the residues flanking the peptide-binding core is shown to significantly improve the prediction accuracy. The method is evaluated on a large-scale benchmark consisting of six independent data sets covering 14 human MHC class II alleles, and is demonstrated to outperform other state-of-the-art MHC class II prediction methods.ConclusionThe NN-align method is competitive with the state-of-the-art MHC class II peptide binding prediction algorithms. The method is publicly available at http://www.cbs.dtu.dk/services/NetMHCII-2.0.
| S-EPMC2753847 | biostudies-literature
Project description:T-cell receptors can recognize foreign peptides bound to major histocompatibility complex (MHC) class-I proteins, and thus trigger the adaptive immune response. Therefore, identifying peptides that can bind to MHC class-I molecules plays a vital role in the design of peptide vaccines. Many computational methods, for example, the state-of-the-art allele-specific method MHCflurry , have been developed to predict the binding affinities between peptides and MHC molecules. In this manuscript, we develop two allele-specific Convolutional Neural Network-based methods named ConvM and SpConvM to tackle the binding prediction problem. Specifically, we formulate the problem as to optimize the rankings of peptide-MHC bindings via ranking-based learning objectives. Such optimization is more robust and tolerant to the measurement inaccuracy of binding affinities, and therefore enables more accurate prioritization of binding peptides. In addition, we develop a new position encoding method in ConvM and SpConvM to better identify the most important amino acids for the binding events. We conduct a comprehensive set of experiments using the latest Immune Epitope Database (IEDB) datasets. Our experimental results demonstrate that our models significantly outperform the state-of-the-art methods including MHCflurry with an average percentage improvement of 6.70% on AUC and 17.10% on ROC5 across 128 alleles.
| S-EPMC8165219 | biostudies-literature
Project description:BackgroundComputational scanning of peptide candidates that bind to a specific major histocompatibility complex (MHC) can speed up the peptide-based vaccine development process and therefore various methods are being actively developed. Recently, machine-learning-based methods have generated successful results by training large amounts of experimental data. However, many machine learning-based methods are generally less sensitive in recognizing locally-clustered interactions, which can synergistically stabilize peptide binding. Deep convolutional neural network (DCNN) is a deep learning method inspired by visual recognition process of animal brain and it is known to be able to capture meaningful local patterns from 2D images. Once the peptide-MHC interactions can be encoded into image-like array(ILA) data, DCNN can be employed to build a predictive model for peptide-MHC binding prediction. In this study, we demonstrated that DCNN is able to not only reliably predict peptide-MHC binding, but also sensitively detect locally-clustered interactions.ResultsNonapeptide-HLA-A and -B binding data were encoded into ILA data. A DCNN, as a pan-specific prediction model, was trained on the ILA data. The DCNN showed higher performance than other prediction tools for the latest benchmark datasets, which consist of 43 datasets for 15 HLA-A alleles and 25 datasets for 10 HLA-B alleles. In particular, the DCNN outperformed other tools for alleles belonging to the HLA-A3 supertype. The F1 scores of the DCNN were 0.86, 0.94, and 0.67 for HLA-A*31:01, HLA-A*03:01, and HLA-A*68:01 alleles, respectively, which were significantly higher than those of other tools. We found that the DCNN was able to recognize locally-clustered interactions that could synergistically stabilize peptide binding. We developed ConvMHC, a web server to provide user-friendly web interfaces for peptide-MHC class I binding predictions using the DCNN. ConvMHC web server can be accessible via http://jumong.kaist.ac.kr:8080/convmhc .ConclusionsWe developed a novel method for peptide-HLA-I binding predictions using DCNN trained on ILA data that encode peptide binding data and demonstrated the reliable performance of the DCNN in nonapeptide binding predictions through the independent evaluation on the latest IEDB benchmark datasets. Our approaches can be applied to characterize locally-clustered patterns in molecular interactions, such as protein/DNA, protein/RNA, and drug/protein interactions.
| S-EPMC5745637 | biostudies-literature
Project description:The real cause of breast cancer is very challenging to determine and therefore early detection of the disease is necessary for reducing the death rate due to risks of breast cancer. Early detection of cancer boosts increasing the survival chance up to 8%. Primarily, breast images emanating from mammograms, X-Rays or MRI are analyzed by radiologists to detect abnormalities. However, even experienced radiologists face problems in identifying features like micro-calcifications, lumps and masses, leading to high false positive and high false negative. Recent advancement in image processing and deep learning create some hopes in devising more enhanced applications that can be used for the early detection of breast cancer. In this work, we have developed a Deep Convolutional Neural Network (CNN) to segment and classify the various types of breast abnormalities, such as calcifications, masses, asymmetry and carcinomas, unlike existing research work, which mainly classified the cancer into benign and malignant, leading to improved disease management. Firstly, a transfer learning was carried out on our dataset using the pre-trained model ResNet50. Along similar lines, we have developed an enhanced deep learning model, in which learning rate is considered as one of the most important attributes while training the neural network. The learning rate is set adaptively in our proposed model based on changes in error curves during the learning process involved. The proposed deep learning model has achieved a performance of 88% in the classification of these four types of breast cancer abnormalities such as, masses, calcifications, carcinomas and asymmetry mammograms.
| S-EPMC8389446 | biostudies-literature
Project description:Major Histocompatibility Complex (MHC) class I molecules enable cytotoxic T lymphocytes to destroy virus-infected or cancerous cells, thereby preventing disease progression. MHC class I molecules provide a snapshot of the contents of a cell by binding to protein fragments arising from intracellular protein turnover and presenting these fragments at the cell surface. Competing fragments (peptides) are selected for cell-surface presentation on the basis of their ability to form a stable complex with MHC class I, by a process known as peptide optimization. A better understanding of the optimization process is important for our understanding of immunodominance, the predominance of some T lymphocyte specificities over others, which can determine the efficacy of an immune response, the danger of immune evasion, and the success of vaccination strategies. In this paper we present a dynamical systems model of peptide optimization by MHC class I. We incorporate the chaperone molecule tapasin, which has been shown to enhance peptide optimization to different extents for different MHC class I alleles. Using a combination of published and novel experimental data to parameterize the model, we arrive at a relation of peptide filtering, which quantifies peptide optimization as a function of peptide supply and peptide unbinding rates. From this relation, we find that tapasin enhances peptide unbinding to improve peptide optimization without significantly delaying the transit of MHC to the cell surface, and differences in peptide optimization across MHC class I alleles can be explained by allele-specific differences in peptide binding. Importantly, our filtering relation may be used to dynamically predict the cell surface abundance of any number of competing peptides by MHC class I alleles, providing a quantitative basis to investigate viral infection or disease at the cellular level. We exemplify this by simulating optimization of the distribution of peptides derived from Human Immunodeficiency Virus Gag-Pol polyprotein.
| S-EPMC3195949 | biostudies-literature
Project description:The MHC class I antigen presentation system enables T cell immunosurveillance of cancers and viruses. Rapidly degraded nascent polypeptides (DRiPs) provide many class I peptide ligands. By knocking down each of the 80 ribosomal proteins, we identified proteins that modulate peptide generation without altering source protein expression. We show that 60S ribosomal proteins L6 (RPL6) and RPL28, which are adjacent on the ribosome, play opposite roles in generating an influenza A virus encoded peptide. RPL6 depletion decreases ubiquitin-dependent peptide presentation, while RPL28 depletion increases ubiquitin-dependent and -independent peptide presentation. 40S ribosomal protein S28 (RPS28) knockdown increases total peptide supply in uninfected cells by increasing DRiP synthesis from non-canonical translation of “untranslated” regions and non-AUG start codons, and sensitizes tumor cells for T cell targeting. RPL23 knocKDut decreases overall peptide supply, impairing T cell recognition of tumor cells. Our findings raise the possibility of modulating immunosurveillance by pharmacologically targeting ribosomes.
2019-01-21 | GSE114484 | GEO
Project description:Purpose of reviewThe molecular and cellular mechanisms that underlie allorecognition of MHC class II molecules have been the subject of much debate and experimentation in recent decades. In this review, we discuss several aspects of MHC class II structure, peptide acquisition and TcR-MHC-peptide interactions that have particular relevance to recognition of cells bearing allogeneic class II molecules.Recent findingsFirst, MHC polymorphism is heavily biased toward those amino acids that influence stable peptide binding by MHC class II. Second, the peptide repertoire presented by class II molecules is highly diverse and can be edited substantially by the molecular catalyst HLA-DM and by tissue-specific expression of HLA-DO, stress and cytokines. Third, T-cell receptor docking onto MHC peptide consistently involves substantial contacts with the bound peptide in the MHC class II molecule. Finally, there is increasing evidence that T-cell recognition of MHC is, in part, germline encoded through T-cell-receptor V region contacts with MHC class II alpha helices.SummaryTogether, these conclusions support the view that allorecognition of MHC class II molecules is likely to parallel key aspects of conventional CD4 T-cell recognition, with allele-dependent variation in peptide representation accounting in large part for the high precursor frequency of alloreactive CD4 T cells.
| S-EPMC3085167 | biostudies-literature
Project description:Assessment of mRNA expression changes in the B-lymphoblastoid cell line Awells after 6 h and 24 h of starvation-induced autophagy Keywords: ordered
2005-04-01 | GSE2435 | GEO
Project description:T cells are vital for adaptive immune responses that protect against pathogens and cancers. The T cell receptor (TCR)-CD3 complex comprises a diverse αβ TCR heterodimer in noncovalent association with three invariant CD3 dimers. The TCR is responsible for recognizing antigenic peptides bound to MHC molecules (pMHC), while the CD3 dimers relay activation signals to the T cell. However, the mechanisms by which TCR engagement by pMHC is transmitted to CD3 remain mysterious, although there is growing evidence that mechanosensing and allostery both play a role. Here, we carried out NMR analysis of a human autoimmune TCR (MS2-3C8) that recognizes a self-peptide from myelin basic protein presented by the MHC class II molecule HLA-DR4. We observed pMHC-induced NMR signal perturbations in MS2-3C8 that indicate long-range effects on TCR β chain conformation and dynamics. Our results demonstrate that, in addition to expected changes in the NMR resonances of pMHC-contacting residues, perturbations extend to the Vβ/Vα, Vβ/Cβ, and Cβ/Cα interfacial regions. Moreover, the pattern of long-range perturbations is similar to that detected previously in the β chains of two MHC class I-restricted TCRs, thereby revealing a common allosteric pathway among three unrelated TCRs. Molecular dynamics (MD) simulations predict similar pMHC-induced effects. Taken together, our results demonstrate that pMHC binding induces long-range allosteric changes in the TCR β chain at conserved sites in both representative MHC class I- and class II-restricted TCRs, and that these sites may play a role in the transmission of signaling information.
| S-EPMC8356565 | biostudies-literature
Project description:Numerous in silico methods predicting peptide binding to major histocompatibility complex (MHC) class I molecules have been developed over the last decades. However, the multitude of available prediction tools makes it non-trivial for the end-user to select which tool to use for a given task. To provide a solid basis on which to compare different prediction tools, we here describe a framework for the automated benchmarking of peptide-MHC class I binding prediction tools. The framework runs weekly benchmarks on data that are newly entered into the Immune Epitope Database (IEDB), giving the public access to frequent, up-to-date performance evaluations of all participating tools. To overcome potential selection bias in the data included in the IEDB, a strategy was implemented that suggests a set of peptides for which different prediction methods give divergent predictions as to their binding capability. Upon experimental binding validation, these peptides entered the benchmark study.The benchmark has run for 15 weeks and includes evaluation of 44 datasets covering 17 MHC alleles and more than 4000 peptide-MHC binding measurements. Inspection of the results allows the end-user to make educated selections between participating tools. Of the four participating servers, NetMHCpan performed the best, followed by ANN, SMM and finally ARB.Up-to-date performance evaluations of each server can be found online at http://tools.iedb.org/auto_bench/mhci/weekly. All prediction tool developers are invited to participate in the benchmark. Sign-up instructions are available at http://tools.iedb.org/auto_bench/mhci/join.mniel@cbs.dtu.dk or bpeters@liai.orgSupplementary data are available at Bioinformatics online.
| S-EPMC4481849 | biostudies-other