Brett Trost

Computational prediction of eukaryotic phosphorylation sites

MOTIVATION Kinase-mediated phosphorylation is the central mechanism of post-translational modification to regulate cellular responses and phenotypes. Signaling defects associated with protein phosphorylation are linked to many diseases, particularly cancer. Characterizing protein kinases and their substrates enhances our ability to understand and treat such diseases and broadens our knowledge of signaling networks in general. While most or all protein kinases have been identified in well-studied eukaryotes, the sites that they phosphorylate have been only partially elucidated. Experimental methods for identifying phosphorylation sites are resource intensive, so the ability to computationally predict potential sites has considerable value. RESULTS Many computational techniques for phosphorylation site prediction have been proposed, most of which are available on the web. These techniques differ in several ways, including the machine learning technique used; the amount of sequence information used; whether or not structural information is used in addition to sequence information; whether predictions are made for specific kinases or for kinases in general; and sources of training and testing data. This review summarizes, categorizes and compares the available methods for phosphorylation site prediction, and provides an overview of the challenges that are faced when designing predictors and how they have been addressed. It should therefore be useful both for those wishing to choose a phosphorylation site predictor for their particular biological application, and for those attempting to improve upon established techniques in the future. CONTACT brett.trost@usask.ca.

Peptidology: short amino acid modules in cell biology and immunology

Summary.Short amino acid motifs, either linear sequences or discontinuous amino acid groupings, can interact with specific protein domains, so exerting a central role in cell adhesion, signal transduction, hormone activity, regulation of transcript expression, enzyme activity, and antigen-antibody interaction. Here, we analyze the literature for such critical short amino acid motifs to determine the minimal peptide length involved in biologically important interactions. We report the pentapeptide unit as a common minimal amino acid sequence critically involved in peptide-protein interaction and immune recognition. The present survey may have implications in defining the dimensional module for peptide-based therapeutical approaches such as the development of novel antibiotics, enzyme inhibitors/activators, mimetic agonists/antagonists of neuropeptides, thrombolitic agents, specific anti-viral agents, etc. In such a therapeutical context, it is of considerable interest that low molecular weight peptides can easily cross biological barriers, are less susceptible to protease attacks, and can be administered at high concentrations. In addition, small peptides are a rational target for strategies aimed at antigen-specific immunotherapeutical intervention. As an example, specific short peptide fragments might be used to elicit antibodies capable of reacting with the full-length proteins containing the peptide fragment’s amino acid sequence, so abolishing the risk of cross-reactivity.

Strength in numbers: achieving greater accuracy in MHC-I binding prediction by combining the results from multiple prediction tools

BackgroundPeptides derived from endogenous antigens can bind to MHC class I molecules. Those which bind with high affinity can invoke a CD8+ immune response, resulting in the destruction of infected cells. Much work in immunoinformatics has involved the algorithmic prediction of peptide binding affinity to various MHC-I alleles. A number of tools for MHC-I binding prediction have been developed, many of which are available on the web.ResultsWe hypothesize that peptides predicted by a number of tools are more likely to bind than those predicted by just one tool, and that the likelihood of a particular peptide being a binder is related to the number of tools that predict it, as well as the accuracy of those tools. To this end, we have built and tested a heuristic-based method of making MHC-binding predictions by combining the results from multiple tools. The predictive performance of each individual tool is first ascertained. These performance data are used to derive weights such that the predictions of tools with better accuracy are given greater credence. The combined tool was evaluated using ten-fold cross-validation and was found to signicantly outperform the individual tools when a high specificity threshold is used. It performs comparably well to the best-performing individual tools at lower specificity thresholds. Finally, it also outperforms the combination of the tools resulting from linear discriminant analysis.ConclusionA heuristic-based method of combining the results of the individual tools better facilitates the scanning of large proteomes for potential epitopes, yielding more actual high-affinity binders while reporting very few false positives.

A Systematic Approach for Analysis of Peptide Array Kinome Data

A new method of analysis of kinome data takes account of the differences between peptide arrays and DNA microarrays. The central roles of kinases in cellular processes and diseases make them highly attractive as indicators of biological responses and as therapeutic targets. Peptide arrays are emerging as an important means of characterizing kinome activity. Currently, the computational tools used to perform high-throughput kinome analyses are not specifically tailored to the nature of the data, which hinders extraction of biological information and overall progress in the field. We have developed a method for kinome analysis, which is implemented as a software pipeline in the R environment. Components and parameters were chosen to address the technical and biological characteristics of kinome microarrays. We performed comparative analysis of kinome data sets that corresponded to stimulation of immune cells with ligands of well-defined signaling pathways: bovine monocytes treated with interferon-γ (IFN-γ), CpG-containing nucleotides, or lipopolysaccharide (LPS). The data sets for each of the treatments were analyzed with our methodology as well as with three other commonly used approaches. The methods were evaluated on the basis of statistical confidence of calculated values with respect to technical and biological variability, and the statistical confidence (P values) by which the known signaling pathways could be independently identified by the pathway analysis of InnateDB (a Web-based resource for innate immunity interactions and pathways). By considering the particular attributes of kinome data, we found that our approach identified more of the peptides involved in the pathways than did the other compared methods and that it did so at a much higher degree of statistical confidence.

Complete Genome Sequence of the Beer Spoilage Organism Pediococcus claussenii ATCC BAA-344T

Pediococcus claussenii is a common brewery contaminant. We have sequenced the chromosome and plasmids of the type strain P. claussenii ATCC BAA-344. A ropy variant was chosen for sequencing to obtain genetic information related to growth in beer, as well as exopolysaccharide and possibly biofilm formation by this organism.

PIIKA 2: An Expanded, Web-Based Platform for Analysis of Kinome Microarray Data

Kinome microarrays are comprised of peptides that act as phosphorylation targets for protein kinases. This platform is growing in popularity due to its ability to measure phosphorylation-mediated cellular signaling in a high-throughput manner. While software for analyzing data from DNA microarrays has also been used for kinome arrays, differences between the two technologies and associated biologies previously led us to develop Platform for Intelligent, Integrated Kinome Analysis (PIIKA), a software tool customized for the analysis of data from kinome arrays. Here, we report the development of PIIKA 2, a significantly improved version with new features and improvements in the areas of clustering, statistical analysis, and data visualization. Among other additions to the original PIIKA, PIIKA 2 now allows the user to: evaluate statistically how well groups of samples cluster together; identify sets of peptides that have consistent phosphorylation patterns among groups of samples; perform hierarchical clustering analysis with bootstrapping; view false negative probabilities and positive and negative predictive values for t-tests between pairs of samples; easily assess experimental reproducibility; and visualize the data using volcano plots, scatterplots, and interactive three-dimensional principal component analyses. Also new in PIIKA 2 is a web-based interface, which allows users unfamiliar with command-line tools to easily provide input and download the results. Collectively, the additions and improvements described here enhance both the breadth and depth of analyses available, simplify the user interface, and make the software an even more valuable tool for the analysis of kinome microarray data. Both the web-based and stand-alone versions of PIIKA 2 can be accessed via http://saphire.usask.ca.

DAPPLE: a pipeline for the homology-based prediction of phosphorylation sites

SUMMARY While many experimentally characterized phosphorylation sites exist for certain organisms, such as human, rat and mouse, few sites are known for other organisms, hampering related research efforts. We have developed a software pipeline called DAPPLE that automates the process of using known phosphorylation sites from other organisms to identify putative sites in an organism of interest. AVAILABILITY DAPPLE is available as a web server at http://saphire.usask.ca. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

Computational phosphorylation site prediction in plants using random forests and organism-specific instance weights

MOTIVATION Phosphorylation is the most important post-translational modification in eukaryotes. Although many computational phosphorylation site prediction tools exist for mammals, and a few were created specifically for Arabidopsis thaliana, none are currently available for other plants. RESULTS In this article, we propose a novel random forest-based method called PHOSFER (PHOsphorylation Site FindER) for applying phosphorylation data from other organisms to enhance the accuracy of predictions in a target organism. As a test case, PHOSFER is applied to phosphorylation sites in soybean, and we show that it more accurately predicts soybean sites than both the existing Arabidopsis-specific predictors, and a simpler machine-learning scheme that uses only known phosphorylation sites and non-phosphorylation sites from soybean. In addition to soybean, PHOSFER will be extended to other organisms in the near future.

Ebola Virus Modulates Transforming Growth Factor β Signaling and Cellular Markers of Mesenchyme-Like Transition in Hepatocytes

ABSTRACT Ebola virus (EBOV) causes a severe hemorrhagic disease in humans and nonhuman primates, with a median case fatality rate of 78.4%. Although EBOV is considered a public health concern, there is a relative paucity of information regarding the modulation of the functional host response during infection. We employed temporal kinome analysis to investigate the relative early, intermediate, and late host kinome responses to EBOV infection in human hepatocytes. Pathway overrepresentation analysis and functional network analysis of kinome data revealed that transforming growth factor (TGF-β)-mediated signaling responses were temporally modulated in response to EBOV infection. Upregulation of TGF-β signaling in the kinome data sets correlated with the upregulation of TGF-β secretion from EBOV-infected cells. Kinase inhibitors targeting TGF-β signaling, or additional cell receptors and downstream signaling pathway intermediates identified from our kinome analysis, also inhibited EBOV replication. Further, the inhibition of select cell signaling intermediates identified from our kinome analysis provided partial protection in a lethal model of EBOV infection. To gain perspective on the cellular consequence of TGF-β signaling modulation during EBOV infection, we assessed cellular markers associated with upregulation of TGF-β signaling. We observed upregulation of matrix metalloproteinase 9, N-cadherin, and fibronectin expression with concomitant reductions in the expression of E-cadherin and claudin-1, responses that are standard characteristics of an epithelium-to-mesenchyme-like transition. Additionally, we identified phosphorylation events downstream of TGF-β that may contribute to this process. From these observations, we propose a model for a broader role of TGF-β-mediated signaling responses in the pathogenesis of Ebola virus disease. IMPORTANCE Ebola virus (EBOV), formerly Zaire ebolavirus, causes a severe hemorrhagic disease in humans and nonhuman primates and is the most lethal Ebola virus species, with case fatality rates of up to 90%. Although EBOV is considered a worldwide concern, many questions remain regarding EBOV molecular pathogenesis. As it is appreciated that many cellular processes are regulated through kinase-mediated phosphorylation events, we employed temporal kinome analysis to investigate the functional responses of human hepatocytes to EBOV infection. Administration of kinase inhibitors targeting signaling pathway intermediates identified in our kinome analysis inhibited viral replication in vitro and reduced EBOV pathogenesis in vivo. Further analysis of our data also demonstrated that EBOV infection modulated TGF-β-mediated signaling responses and promoted “mesenchyme-like” phenotypic changes. Taken together, these results demonstrated that EBOV infection specifically modulates TGF-β-mediated signaling responses in epithelial cells and may have broader implications in EBOV pathogenesis.

Divergent Immune Responses to Mycobacterium avium subsp. paratuberculosis Infection Correlate with Kinome Responses at the Site of Intestinal Infection

ABSTRACT Mycobacterium avium subsp. paratuberculosis is the causative agent of Johnes disease (JD) in cattle. M. avium subsp. paratuberculosis infects the gastrointestinal tract of calves, localizing and persisting primarily in the distal ileum. A high percentage of cattle exposed to M. avium subsp. paratuberculosis do not develop JD, but the mechanisms by which they resist infection are not understood. Here, we merge an established in vivo bovine intestinal segment model for M. avium subsp. paratuberculosis infection with bovine-specific peptide kinome arrays as a first step to understanding how infection influences host kinomic responses at the site of infection. Application of peptide arrays to in vivo tissue samples represents a critical and ambitious step in using this technology to understand host-pathogen interactions. Kinome analysis was performed on intestinal samples from 4 ileal segments subdivided into 10 separate compartments (6 M. avium subsp. paratuberculosis-infected compartments and 4 intra-animal controls) using bovine-specific peptide arrays. Kinome data sets clustered into two groups, suggesting unique binary responses to M. avium subsp. paratuberculosis. Similarly, two M. avium subsp. paratuberculosis-specific immune responses, characterized by different antibody, T cell proliferation, and gamma interferon (IFN-γ) responses, were also observed. Interestingly, the kinomic groupings segregated with the immune response groupings. Pathway and gene ontology analyses revealed that differences in innate immune and interleukin signaling and particular differences in the Wnt/β-catenin pathway distinguished the kinomic groupings. Collectively, kinome analysis of tissue samples offers insight into the complex cellular responses induced by M. avium subsp. paratuberculosis in the ileum and provides a novel method to understand mechanisms that alter the balance between cell-mediated and antibody responses to M. avium subsp. paratuberculosis infection.