A common open representation of mass spectrometry data and its application to proteomics research

An evaluation, comparison, and accurate benchmarking of several publicly available MS/MS search algorithms: sensitivity and specificity analysis

MS/MS and associated database search algorithms are essential proteomic tools for identifying peptides. Due to their widespread use, it is now time to perform a systematic analysis of the various algorithms currently in use. Using blood specimens used in the HUPO Plasma Proteome Project, we have evaluated five search algorithms with respect to their sensitivity and specificity, and have also accurately benchmarked them based on specified false-positive (FP) rates. Spectrum Mill and SEQUEST performed well in terms of sensitivity, but were inferior to MASCOT, X!Tandem, and Sonar in terms of specificity. Overall, MASCOT, a probabilistic search algorithm, correctly identified most peptides based on a specified FP rate. The rescoring algorithm, PeptideProphet, enhanced the overall performance of the SEQUEST algorithm, as well as provided predictable FP error rates. Ideally, score thresholds should be calculated for each peptide spectrum or minimally, derived from a reversed-sequence search as demonstrated in this study based on a validated data set. The availability of open-source search algorithms, such as X!Tandem, makes it feasible to further improve the validation process (manual or automatic) on the basis of "consensus scoring", i.e., the use of multiple (at least two) search algorithms to reduce the number of FPs. complement.

Recent advances in blood-related proteomics

Blood is divided in two compartments, namely, plasma and cells. The latter contain red blood cells, leukocytes, and platelets. From a descriptive medical discipline, hematology has evolved towards a pioneering discipline where molecular biology has permitted the development of prognostic and diagnostic indicators for disease. The recent advance in MS and protein separation now allows similar progress in the analysis of proteins. Proteomics offers great promise for the study of proteins in plasma/serum, indeed a number of proteomics databases for plasma/serum have been established. This is a very complex body fluid containing lipids, carbohydrates, amino acids, vitamins, nucleic acids, hormones, and proteins. About 1500 different proteins have recently been identified, and a number of potential new markers of diseases have been characterized. Here, examples of the enormous promise of plasma/serum proteomic analysis for diagnostic/prognostic markers and information on disease mechanism are given. Within the blood are also a large number of different blood cell types that potentially hold similar information. Proteomics of red blood cells, until now, has not improved our knowledge of these cells, in contrast to the major progresses achieved while studying platelets and leukocytes. In the future, proteomics will change several aspects of hematology.

Interpretation of Shotgun Proteomic Data

The shotgun proteomic strategy based on digesting proteins into peptides and sequencing them using tandem mass spectrometry and automated database searching has become the method of choice for identifying proteins in most large scale studies. However, the peptide-centric nature of shotgun proteomics complicates the analysis and biological interpretation of the data especially in the case of higher eukaryote organisms. The same peptide sequence can be present in multiple different proteins or protein isoforms. Such shared peptides therefore can lead to ambiguities in determining the identities of sample proteins. In this article we illustrate the difficulties of interpreting shotgun proteomic data and discuss the need for common nomenclature and transparent informatic approaches. We also discuss related issues such as the state of protein sequence databases and their role in shotgun proteomic analysis, interpretation of relative peptide quantification data in the presence of multiple protein isoforms, the integration of proteomic and transcriptional data, and the development of a computational infrastructure for the integration of multiple diverse datasets.

An analysis of extensible modelling for functional genomics data

Background

Several data formats have been developed for large scale biological experiments, using a variety of methodologies. Most data formats contain a mechanism for allowing extensions to encode unanticipated data types. Extensions to data formats are important because the experimental methodologies tend to be fairly diverse and rapidly evolving, which hinders the creation of formats that will be stable over time.

Results

In this paper we review the data formats that exist in functional genomics, some of which have become de facto or de jure standards, with a particular focus on how each domain has been modelled, and how each format allows extensions. We describe the tasks that are frequently performed over data formats and analyse how well each task is supported by a particular modelling structure.

Conclusion

From our analysis, we make recommendations as to the types of modelling structure that are most suitable for particular types of experimental annotation. There are several standards currently under development that we believe could benefit from systematically following a set of guidelines.

From XML to RDF: how semantic web technologies will change the design of 'omic' standards

With the ongoing rapid increase in both volume and diversity of 'omic' data (genomics, transcriptomics, proteomics, and others), the development and adoption of data standards is of paramount importance to realize the promise of systems biology. A recent trend in data standard development has been to use extensible markup language (XML) as the preferred mechanism to define data representations. But as illustrated here with a few examples from proteomics data, the syntactic and document-centric XML cannot achieve the level of interoperability required by the highly dynamic and integrated bioinformatics applications. In the present article, we discuss why semantic web technologies, as recommended by the World Wide Web consortium (W3C), expand current data standard technology for biological data representation and management.

A uniform proteomics MS/MS analysis platform utilizing open XML file formats

The analysis of tandem mass (MS/MS) data to identify and quantify proteins is hampered by the heterogeneity of file formats at the raw spectral data, peptide identification, and protein identification levels. Different mass spectrometers output their raw spectral data in a variety of proprietary formats, and alternative methods that assign peptides to MS/MS spectra and infer protein identifications from those peptide assignments each write their results in different formats. Here we describe an MS/MS analysis platform, the Trans-Proteomic Pipeline, which makes use of open XML file formats for storage of data at the raw spectral data, peptide, and protein levels. This platform enables uniform analysis and exchange of MS/MS data generated from a variety of different instruments, and assigned peptides using a variety of different database search programs. We demonstrate this by applying the pipeline to data sets generated by ThermoFinnigan LCQ, ABI 4700 MALDI-TOF/TOF, and Waters Q-TOF instruments, and searched in turn using SEQUEST, Mascot, and COMET.

Human Plasma PeptideAtlas

Peptide identifications of high probability from 28 LC-MS/MS human serum and plasma experiments from eight different laboratories, carried out in the context of the HUPO Plasma Proteome Project, were combined and mapped to the EnsEMBL human genome. The 6929 distinct observed peptides were mapped to approximately 960 different proteins. The resulting compendium of peptides and their associated samples, proteins, and genes is made publicly available as a reference for future research on human plasma.

Do we want our data raw? Including binary mass spectrometry data in public proteomics data repositories

With the human Plasma Proteome Project (PPP) pilot phase completed, the largest and most ambitious proteomics experiment to date has reached its first milestone. The correspondingly impressive amount of data that came from this pilot project emphasized the need for a centralized dissemination mechanism and led to the development of a detailed, PPP specific data gathering infrastructure at the University of Michigan, Ann Arbor as well as the protein identifications database project at the European Bioinformatics Institute as a general proteomics data repository. One issue that crept up while discussing which data to store for the PPP concerns whether the raw, binary data coming from the mass spectrometers should be stored, or rather the more compact and already significantly processed peak lists. As this debate is not restricted to the PPP but relates to the proteomics community in general, we will attempt to detail the relative merits and caveats associated with centralized storage and dissemination of raw data and/or peak lists, building on the extensive experience gained during the PPP pilot phase. Finally, some suggestions are made for both immediate and future storage of MS data in public repositories.

Multidimensional protein identification technology (MudPIT): technical overview of a profiling method optimized for the comprehensive proteomic investigation of normal and diseased heart tissue

An optimized analytical expression profiling strategy based on gel-free multidimensional protein identification technology (MudPIT) is reported for the systematic investigation of biochemical (mal)-adaptations associated with healthy and diseased heart tissue. Enhanced shotgun proteomic detection coverage and improved biological inference is achieved by pre-fractionation of excised mouse cardiac muscle into subcellular components, with each organellar fraction investigated exhaustively using multiple repeat MudPIT analyses. Functional-enrichment, high-confidence identification, and relative quantification of hundreds of organelle- and tissue-specific proteins are achieved readily, including detection of low abundance transcriptional regulators, signaling factors, and proteins linked to cardiac disease. Important technical issues relating to data validation, including minimization of artifacts stemming from biased under-sampling and spurious false discovery, together with suggestions for further fine-tuning of sample preparation, are discussed. A framework for follow-up bioinformatic examination, pattern recognition, and data mining is also presented in the context of a stringent application of MudPIT for probing fundamental aspects of heart muscle physiology as well as the discovery of perturbations associated with heart failure.

PRIDE: The proteomics identifications database

The advent of high-throughput proteomics has enabled the identification of ever increasing numbers of proteins. Correspondingly, the number of publications centered on these protein identifications has increased dramatically. With the first results of the HUPO Plasma Proteome Project being analyzed and many other large-scale proteomics projects about to disseminate their data, this trend is not likely to flatten out any time soon. However, the publication mechanism of these identified proteins has lagged behind in technical terms. Often very long lists of identifications are either published directly with the article, resulting in both a voluminous and rather tedious read, or are included on the publisher's website as supplementary information. In either case, these lists are typically only provided as portable document format documents with a custom-made layout, making it practically impossible for computer programs to interpret them, let alone efficiently query them. Here we propose the proteomics identifications (PRIDE) database (http://www.ebi.ac.uk/pride) as a means to finally turn publicly available data into publicly accessible data. PRIDE offers a web-based query interface, a user-friendly data upload facility, and a documented application programming interface for direct computational access. The complete PRIDE database, source code, data, and support tools are freely available for web access or download and local installation.

Metabolic footprinting and systems biology: the medium is the message

One element of classical systems analysis treats a system as a black or grey box, the inner structure and behaviour of which can be analysed and modelled by varying an internal or external condition, probing it from outside and studying the effect of the variation on the external observables. The result is an understanding of the inner make-up and workings of the system. The equivalent of this in biology is to observe what a cell or system excretes under controlled conditions - the 'metabolic footprint' or exometabolome - as this is readily and accurately measurable. Here, we review the principles, experimental approaches and scientific outcomes that have been obtained with this useful and convenient strategy.

A software suite for the generation and comparison of peptide arrays from sets of data collected by liquid chromatography-mass spectrometry

There is an increasing interest in the quantitative proteomic measurement of the protein contents of substantially similar biological samples, e.g. for the analysis of cellular response to perturbations over time or for the discovery of protein biomarkers from clinical samples. Technical limitations of current proteomic platforms such as limited reproducibility and low throughput make this a challenging task. A new LC-MS-based platform is able to generate complex peptide patterns from the analysis of proteolyzed protein samples at high throughput and represents a promising approach for quantitative proteomics. A crucial component of the LC-MS approach is the accurate evaluation of the abundance of detected peptides over many samples and the identification of peptide features that can stratify samples with respect to their genetic, physiological, or environmental origins. We present here a new software suite, SpecArray, that generates a peptide versus sample array from a set of LC-MS data. A peptide array stores the relative abundance of thousands of peptide features in many samples and is in a format identical to that of a gene expression microarray. A peptide array can be subjected to an unsupervised clustering analysis to stratify samples or to a discriminant analysis to identify discriminatory peptide features. We applied the SpecArray to analyze two sets of LC-MS data: one was from four repeat LC-MS analyses of the same glycopeptide sample, and another was from LC-MS analysis of serum samples of five male and five female mice. We demonstrate through these two study cases that the SpecArray software suite can serve as an effective software platform in the LC-MS approach for quantitative proteomics.

Processing methods for differential analysis of LC/MS profile data

Background

Liquid chromatography coupled to mass spectrometry (LC/MS) has been widely used in proteomics and metabolomics research. In this context, the technology has been increasingly used for differential profiling, i.e. broad screening of biomolecular components across multiple samples in order to elucidate the observed phenotypes and discover biomarkers. One of the major challenges in this domain remains development of better solutions for processing of LC/MS data.

Results

We present a software package MZmine that enables differential LC/MS analysis of metabolomics data. This software is a toolbox containing methods for all data processing stages preceding differential analysis: spectral filtering, peak detection, alignment and normalization. Specifically, we developed and implemented a new recursive peak search algorithm and a secondary peak picking method for improving already aligned results, as well as a normalization tool that uses multiple internal standards. Visualization tools enable comparative viewing of data across multiple samples. Peak lists can be exported into other data analysis programs. The toolbox has already been utilized in a wide range of applications. We demonstrate its utility on an example of metabolic profiling of Catharanthus roseus cell cultures.

Conclusion

The software is freely available under the GNU General Public License and it can be obtained from the project web page at: .

The continuing evolution of shotgun proteomics

Shotgun proteomics has emerged as a powerful approach for the analysis of complex protein mixtures, including biofluids, tissues, cells, organelles or protein complexes. Having evolved from the integration of chromatography and mass spectrometry, innovations in sample preparation, multidimensional chromatography, mass spectrometry and proteomic informatics continually facilitate, enable and challenge shotgun proteomics. As a result, shotgun proteomics continues to evolve and enable new areas of biological research, and is beginning to impact human disease diagnosis and therapeutic intervention.

MSight: An image analysis software for liquid chromatography-mass spectrometry

Images obtained from high-throughput mass spectrometry (MS) contain information that remains hidden when looking at a single spectrum at a time. Image processing of liquid chromatography-MS datasets can be extremely useful for quality control, experimental monitoring and knowledge extraction. The importance of imaging in differential analysis of proteomic experiments has already been established through two-dimensional gels and can now be foreseen with MS images. We present MSight, a new software designed to construct and manipulate MS images, as well as to facilitate their analysis and comparison.

Bioinformatic methods to exploit mass spectrometric data for proteomic applications

The new technologies in mass spectrometric analysis of peptides and proteins necessary to accommodate proteomics-scale analyses require, in turn, concomitant development of informatics technologies suitable for very large-scale data handling and analysis. This chapter focuses on the data analysis tools available to the community for analysis of mass spectrometric proteomics data. Different database searching strategies are discussed for peptide and protein identification, and approaches and tools available for comparative quantitative analysis of samples are outlined.

An alternative sampling algorithm for use in liquid chromatography/tandem mass spectrometry experiments

In liquid chromatography/tandem mass spectrometry (LC/MS/MS) analyses of complex peptide mixtures, dynamic exclusion functions are used to minimize repeat selections of identical precursors for collision induced dissociation (CID). We describe a new algorithm for the dynamic exclusion of m/z values during LC/MS/MS. Full-scan based peak exclusion (Fulspec) uses a simplified model of chromatographic peak formation to detect and exclude contaminants present throughout the run or that lead to broad peaks. Therefore, instead of excluding peptides from fragment analysis according to a rigidly predefined time window, the chromatographic properties of the detected analytes are used. The algorithm was tested on two datasets derived from previously published experiments. Fulspec achieves a distribution of CID spectra with minimal tailing on the retention time axis, without resorting to rigid exclusion of m/z values. The procedure further excludes intensities with a bias towards low-quality CID spectra. This combination frees up valuable analytical capacity. The underlying intensity vs. quality analyses challenge the assumption that abundant precursors automatically give the best identifications. Further validation of the algorithm will require its incorporation by equipment manufacturers into the instrument control programs.

Integration with the human genome of peptide sequences obtained by high-throughput mass spectrometry

Peptides derived from protein tandem mass spectrometry data have been mapped to the human genome sequence forming an expandable resource for the proteomic data.

A crucial aim upon the completion of the human genome is the verification and functional annotation of all predicted genes and their protein products. Here we describe the mapping of peptides derived from accurate interpretations of protein tandem mass spectrometry (MS) data to eukaryotic genomes and the generation of an expandable resource for integration of data from many diverse proteomics experiments. Furthermore, we demonstrate that peptide identifications obtained from high-throughput proteomics can be integrated on a large scale with the human genome. This resource could serve as an expandable repository for MS-derived proteome information.