Dataset Information

Improving peptide-MHC class I binding prediction for unbalanced datasets.

ABSTRACT: Establishment of peptide binding to Major Histocompatibility Complex class I (MHCI) is a crucial step in the development of subunit vaccines and prediction of such binding could greatly reduce costs and accelerate the experimental process of identifying immunogenic peptides. Many methods have been applied to the prediction of peptide-MHCI binding, with some achieving outstanding performance. Because of the experimental methods used to measure binding or affinity between peptides and MHCI molecules, however, available datasets are enriched for nonbinders, and thus highly unbalanced. Although there is no consensus on the ideal class distribution for training sets, extremely unbalanced datasets can be detrimental to the performance of prediction algorithms.We have developed a decision-theoretic framework to construct cost-sensitive trees to predict peptide-MHCI binding and have used them to 1) Assess the impact of the training data's class distribution on classifier accuracy, and 2) Compare resampling and cost-sensitive methods as approaches to compensate for training data imbalance. Our results confirm that highly unbalanced training sets can reduce the accuracy of classifier predictions and show that, in the peptide-MHCI binding context, resampling methods do not improve the classifier performance. In contrast, cost-sensitive methods significantly improve accuracy of decision trees. Finally, we propose the use of a training scheme that, when the training set is enriched for nonbinders, consistently improves the overall classifier accuracy compared to cost-insensitive classifiers and, in particular, increases the sensitivity of the classifiers. This method minimizes the expected classification cost for large datasets.Our method consistently improves the performance of decision trees in predicting peptide-MHC class I binding by using cost-balancing techniques to compensate for the imbalance in the training dataset.

SUBMITTER: Sales AP

PROVIDER: S-EPMC2586639 | biostudies-literature | 2008 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Improving peptide-MHC class I binding prediction for unbalanced datasets.

Sales Ana Paula AP Tomaras Georgia D GD Kepler Thomas B TB

BMC bioinformatics 20080919

<h4>Background</h4>Establishment of peptide binding to Major Histocompatibility Complex class I (MHCI) is a crucial step in the development of subunit vaccines and prediction of such binding could greatly reduce costs and accelerate the experimental process of identifying immunogenic peptides. Many methods have been applied to the prediction of peptide-MHCI binding, with some achieving outstanding performance. Because of the experimental methods used to measure binding or affinity between peptid ...[more]

PMID: 18803836

Similar Datasets

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.

Project description:BACKGROUND:MHC class II proteins bind oligopeptide fragments derived from proteolysis of pathogen antigens, presenting them at the cell surface for recognition by CD4+ T cells. Human MHC class II alleles are grouped into three loci: HLA-DP, HLA-DQ and HLA-DR. In contrast to HLA-DR and HLA-DQ, HLA-DP proteins have not been studied extensively, as they have been viewed as less important in immune responses than DRs and DQs. However, it is now known that HLA-DP alleles are associated with many autoimmune diseases. Quite recently, the X-ray structure of the HLA-DP2 molecule (DPA*0103, DPB1*0201) in complex with a self-peptide derived from the HLA-DR ?-chain has been determined. In the present study, we applied a validated molecular docking protocol to a library of 247 modelled peptide-DP2 complexes, seeking to assess the contribution made by each of the 20 naturally occurred amino acids at each of the nine binding core peptide positions and the four flanking residues (two on both sides). RESULTS:The free binding energies (FBEs) derived from the docking experiments were normalized on a position-dependent (npp) and on an overall basis (nap), and two docking score-based quantitative matrices (DS-QMs) were derived: QMnpp and QMnap. They reveal the amino acid preferences at each of the 13 positions considered in the study. Apart from the leading role of anchor positions p1 and p6, the binding to HLA-DP2 depends on the preferences at p2. No effect of the flanking residues was found on the peptide binding predictions to DP2, although all four of them show strong preferences for particular amino acids. The predictive ability of the DS-QMs was tested using a set of 457 known binders to HLA-DP2, originating from 24 proteins. The sensitivities of the predictions at five different thresholds (5%, 10%, 15%, 20% and 25%) were calculated and compared to the predictions made by the NetMHCII and IEDB servers. Analysis of the DS-QMs indicated an improvement in performance. Additionally, DS-QMs identified the binding cores of several known DP2 binders. CONCLUSIONS:The molecular docking protocol, as applied to a combinatorial library of peptides, models the peptide-HLA-DP2 protein interaction effectively, generating reliable predictions in a quantitative assessment. The method is structure-based and does not require extensive experimental sequence-based data. Thus, it is universal and can be applied to model any peptide - protein interaction.

Dataset Information

Improving peptide-MHC class I binding prediction for unbalanced datasets.

Publications

Improving peptide-MHC class I binding prediction for unbalanced datasets.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets