In molecular biology, it is a general assumption that the ensemble of expressed molecules, their activities and interactions determine biological function, cellular states and phenotypes. Stable protein complexes-or macromolecular machines-are, in turn, the key functional entities mediating and modulating most biological processes. Although identifying protein complexes and their subunit composition can now be done inexpensively and at scale, determining their function remains challenging and labor intensive. This study describes Protein Complex Function predictor (PCfun), the first computational framework for the systematic annotation of protein complex functions using Gene Ontology (GO) terms. PCfun is built upon a word embedding using natural language processing techniques based on 1 million open access PubMed Central articles. Specifically, PCfun leverages two approaches for accurately identifying protein complex function, including: (i) an unsupervised approach that obtains the nearest neighbor (NN) GO term word vectors for a protein complex query vector and (ii) a supervised approach using Random Forest (RF) models trained specifically for recovering the GO terms of protein complex queries described in the CORUM protein complex database. PCfun consolidates both approaches by performing a hypergeometric statistical test to enrich the top NN GO terms within the child terms of the GO terms predicted by the RF models. The documentation and implementation of the PCfun package are available at https://github.com/sharmavaruns/PCfun. We anticipate that PCfun will serve as a useful tool and novel paradigm for the large-scale characterization of protein complex function.

SUMMARY: In recent years, SWATH-MS has become the proteomic method of choice for data-independent-acquisition, as it enables high proteome coverage, accuracy and reproducibility. However, data analysis is convoluted and requires prior information and expert curation. Furthermore, as quantification is limited to a small set of peptides, potentially important biological information may be discarded. Here we demonstrate that deep learning can be used to learn discriminative features directly from raw MS data, eliminating hence the need of elaborate data processing pipelines. Using transfer learning to overcome sample sparsity, we exploit a collection of publicly available deep learning models already trained for the task of natural image classification. These models are used to produce feature vectors from each mass spectrometry (MS) raw image, which are later used as input for a classifier trained to distinguish tumor from normal prostate biopsies. Although the deep learning models were originally trained for a completely different classification task and no additional fine-tuning is performed on them, we achieve a highly remarkable classification performance of 0.876 AUC. We investigate different types of image preprocessing and encoding. We also investigate whether the inclusion of the secondary MS2 spectra improves the classification performance. Throughout all tested models, we use standard protein expression vectors as gold standards. Even with our naïve implementation, our results suggest that the application of deep learning and transfer learning techniques might pave the way to the broader usage of raw mass spectrometry data in real-time diagnosis. AVAILABILITY AND IMPLEMENTATION: The open source code used to generate the results from MS images is available on GitHub: https://ibm.biz/mstransc. The raw MS data underlying this article cannot be shared publicly for the privacy of individuals that participated in the study. Processed data including the MS images, their encodings, classification labels and results can be accessed at the following link: https://ibm.box.com/v/mstc-supplementary. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Apolipoprotein E (apoE) is a forefront actor in the transport of lipids and the maintenance of cholesterol homeostasis, and is also strongly implicated in Alzheimer's disease. Upon lipid-binding apoE adopts a conformational state that mediates the receptor-induced internalization of lipoproteins. Due to its inherent structural dynamics and the presence of lipids, the structure of the biologically active apoE remains so far poorly described. To address this issue, we developed an innovative hybrid method combining experimental data with molecular modeling and dynamics to generate comprehensive models of the lipidated apoE4 isoform. Chemical cross-linking combined with mass spectrometry provided distance restraints, characterizing the three-dimensional organization of apoE4 molecules at the surface of lipidic nanoparticles. The ensemble of spatial restraints was then rationalized in an original molecular modeling approach to generate monomeric models of apoE4 that advocated the existence of two alternative conformations. These two models point towards an activation mechanism of apoE4 relying on a regulation of the accessibility of its receptor binding region. Further, molecular dynamics simulations of the dimerized and lipidated apoE4 monomeric conformations revealed an elongation of the apoE N-terminal domain, whereby helix 4 is rearranged, together with Arg172, into a proper orientation essential for lipoprotein receptor association. Overall, our results show how apoE4 adapts its conformation for the recognition of the low density lipoprotein receptor and we propose a novel mechanism of activation for apoE4 that is based on accessibility and remodeling of the receptor binding region.

MOTIVATION: Targeted mass spectrometry comprises a set of powerful methods to obtain accurate and consistent protein quantification in complex samples. To fully exploit these techniques, a cross-platform and open-source software stack based on standardized data exchange formats is required. RESULTS: We present TAPIR, a fast and efficient Python visualization software for chromatograms and peaks identified in targeted proteomics experiments. The input formats are open, community-driven standardized data formats (mzML for raw data storage and TraML encoding the hierarchical relationships between transitions, peptides and proteins). TAPIR is scalable to proteome-wide targeted proteomics studies (as enabled by SWATH-MS), allowing researchers to visualize high-throughput datasets. The framework integrates well with existing automated analysis pipelines and can be extended beyond targeted proteomics to other types of analyses. AVAILABILITY AND IMPLEMENTATION: TAPIR is available for all computing platforms under the 3-clause BSD license at https://github.com/msproteomicstools/msproteomicstools.

Many proteomic investigations summarize the quantitative information across multiple spectral features into protein-level conclusions. Data-independent spectral acquisition (DIA) now generates a lot of interest, as it allows us to quantify many spectral features in a single run. However, the disadvantage of DIA experiments as compared, e.g., to Selected Reaction Monitoring (SRM) is that the features are subject to interferences and noise. We argue that between-run variation provides an additional insight for distinguishing good-quality and noisy DIA features. To appropriately use the quantitative between-run variation, it is important to account for the properties experimental design, and distinguish random artifacts from the biological changes. We have previously proposed a method (Chang et al., ASMS 2013) that accounts for the experimental design to eliminate features with low information content. In this project we furthermore emphasized that conducting regularization helps us avoid exploring every subset of features exhaustively, and allows us to conduct hypothesis tests later on so that we would be able to control the false discovery rate of the feature selection process. We evaluated our proposed approach by using three datasets that have some notion of ground truth: an extensive simulation study, a controlled mixture where proteins were spiked into a complex background in known concentrations, and a study of 232 plasma samples, where 18 proteins were quantified in both SWAH and SRM mode in presence of heavy labeled reference peptides. We worked on [ 1 ] protein-level estimates of fold changes between conditions, [ 2 ] sensitivity and specificity of detecting changes in protein abundance, and [ 3 ] accuracy of relative quantification of protein abundance in individual biological samples. A family of linear mixed models similar to that in MSstats http://www.msstats.org were fit to all the datasets. Then we conducted the regularization and hypothesis test to control the selection false discovery rate. The results demonstrated that our proposed feature selection approach enhanced sensitivity and specificity of the conclusions, was robust to the amount of noisy fragments, and increased the correlation of subject quantification between SRM and DIA workflows. Importantly, the performance exceeded that of the frequently used 'top 3' approach, which consists of using three spectral features with the highest average intensity between runs. Furthermore, we showed that our proposed approach outperforms using correlation to select the information features.

MOTIVATION: The determination of absolute quantities of proteins in biological samples is necessary for multiple types of scientific inquiry. While relative quantification has been commonly used in proteomics, few proteomic datasets measuring absolute protein quantities have been reported to date. Various technologies have been applied using different types of input data, e.g. ion intensities or spectral counts, as well as different absolute normalization strategies. To date, a user-friendly and transparent software supporting large-scale absolute protein quantification has been lacking. RESULTS: We present a bioinformatics tool, termed aLFQ, which supports the commonly used absolute label-free protein abundance estimation methods (TopN, iBAQ, APEX, NSAF and SCAMPI) for LC-MS/MS proteomics data, together with validation algorithms enabling automated data analysis and error estimation. AVAILABILITY AND IMPLEMENTATION: aLFQ is written in R and freely available under the GPLv3 from CRAN (http://www.cran.r-project.org). Instructions and example data are provided in the R-package. The raw data can be obtained from the PeptideAtlas raw data repository (PASS00321). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) is widely used for quantitative proteomic investigations. The typical output of such studies is a list of identified and quantified peptides. The biological and clinical interest is, however, usually focused on quantitative conclusions at the protein level. Furthermore, many investigations ask complex biological questions by studying multiple interrelated experimental conditions. Therefore, there is a need in the field for generic statistical models to quantify protein levels even in complex study designs. RESULTS: We propose a general statistical modeling approach for protein quantification in arbitrary complex experimental designs, such as time course studies, or those involving multiple experimental factors. The approach summarizes the quantitative experimental information from all the features and all the conditions that pertain to a protein. It enables both protein significance analysis between conditions, and protein quantification in individual samples or conditions. We implement the approach in an open-source R-based software package MSstats suitable for researchers with a limited statistics and programming background. CONCLUSIONS: We demonstrate, using as examples two experimental investigations with complex designs, that a simultaneous statistical modeling of all the relevant features and conditions yields a higher sensitivity of protein significance analysis and a higher accuracy of protein quantification as compared to commonly employed alternatives. The software is available at http://www.stat.purdue.edu/~ovitek/Software.html.

MOTIVATION: Chemical cross-linking of proteins or protein complexes and the mass spectrometry-based localization of the cross-linked amino acids in peptide sequences is a powerful method for generating distance restraints on the substrate's topology. RESULTS: Here, we introduce the algorithm Xwalk for predicting and validating these cross-links on existing protein structures. Xwalk calculates and displays non-linear distances between chemically cross-linked amino acids on protein surfaces, while mimicking the flexibility and non-linearity of cross-linker molecules. It returns a 'solvent accessible surface distance', which corresponds to the length of the shortest path between two amino acids, where the path leads through solvent occupied space without penetrating the protein surface. AVAILABILITY: Xwalk is freely available as a web server or stand-alone JAVA application at http://www.xwalk.org.

BACKGROUND: Modern data generation techniques used in distributed systems biology research projects often create datasets of enormous size and diversity. We argue that in order to overcome the challenge of managing those large quantitative datasets and maximise the biological information extracted from them, a sound information system is required. Ease of integration with data analysis pipelines and other computational tools is a key requirement for it. RESULTS: We have developed openBIS, an open source software framework for constructing user-friendly, scalable and powerful information systems for data and metadata acquired in biological experiments. openBIS enables users to collect, integrate, share, publish data and to connect to data processing pipelines. This framework can be extended and has been customized for different data types acquired by a range of technologies. CONCLUSIONS: openBIS is currently being used by several SystemsX.ch and EU projects applying mass spectrometric measurements of metabolites and proteins, High Content Screening, or Next Generation Sequencing technologies. The attributes that make it interesting to a large research community involved in systems biology projects include versatility, simplicity in deployment, scalability to very large data, flexibility to handle any biological data type and extensibility to the needs of any research domain.

BACKGROUND: Since its inception, proteomics has essentially operated in a discovery mode with the goal of identifying and quantifying the maximal number of proteins in a sample. Increasingly, proteomic measurements are also supporting hypothesis-driven studies, in which a predetermined set of proteins is consistently detected and quantified in multiple samples. Selected reaction monitoring (SRM) is a targeted mass spectrometric technique that supports the detection and quantification of specific proteins in complex samples at high sensitivity and reproducibility. Here, we describe ATAQS, an integrated software platform that supports all stages of targeted, SRM-based proteomics experiments including target selection, transition optimization and post acquisition data analysis. This software will significantly facilitate the use of targeted proteomic techniques and contribute to the generation of highly sensitive, reproducible and complete datasets that are particularly critical for the discovery and validation of targets in hypothesis-driven studies in systems biology. RESULT: We introduce a new open source software pipeline, ATAQS (Automated and Targeted Analysis with Quantitative SRM), which consists of a number of modules that collectively support the SRM assay development workflow for targeted proteomic experiments (project management and generation of protein, peptide and transitions and the validation of peptide detection by SRM). ATAQS provides a flexible pipeline for end-users by allowing the workflow to start or end at any point of the pipeline, and for computational biologists, by enabling the easy extension of java algorithm classes for their own algorithm plug-in or connection via an external web site.This integrated system supports all steps in a SRM-based experiment and provides a user-friendly GUI that can be run by any operating system that allows the installation of the Mozilla Firefox web browser. CONCLUSIONS: Targeted proteomics via SRM is a powerful new technique that enables the reproducible and accurate identification and quantification of sets of proteins of interest. ATAQS is the first open-source software that supports all steps of the targeted proteomics workflow. ATAQS also provides software API (Application Program Interface) documentation that enables the addition of new algorithms to each of the workflow steps. The software, installation guide and sample dataset can be found in http://tools.proteomecenter.org/ATAQS/ATAQS.html.

MOTIVATION: Liquid chromatography tandem mass spectrometry (LC-MS/MS) is the predominant method to comprehensively characterize complex protein mixtures such as samples from prefractionated or complete proteomes. In order to maximize proteome coverage for the studied sample, i.e. identify as many traceable proteins as possible, LC-MS/MS experiments are typically repeated extensively and the results combined. Proteome coverage prediction is the task of estimating the number of peptide discoveries of future LC-MS/MS experiments. Proteome coverage prediction is important to enhance the design of efficient proteomics studies. To date, there does not exist any method to reliably estimate the increase of proteome coverage at an early stage. RESULTS: We propose an extended infinite Markov model DiriSim to extrapolate the progression of proteome coverage based on a small number of already performed LC-MS/MS experiments. The method explicitly accounts for the uncertainty of peptide identifications. We tested DiriSim on a set of 37 LC-MS/MS experiments of a complete proteome sample and demonstrated that DiriSim correctly predicts the coverage progression already from a small subset of experiments. The predicted progression enabled us to specify maximal coverage for the test sample. We demonstrated that quality requirements on the final proteome map impose an upper bound on the number of useful experiment repetitions and limit the achievable proteome coverage.

SUMMARY: We describe an integrative software platform, Prequips, for comparative proteomics-based systems biology analysis that: (i) integrates all information generated from mass spectrometry (MS)-based proteomics as well as from basic proteomics data analysis tools, (ii) visualizes such information for various proteomic analyses via graphical interfaces and (iii) links peptide and protein abundances to external tools often used in systems biology studies. AVAILABILITY: http://prequips.sourceforge.net

BACKGROUND: Crucial foundations of any quantitative systems biology experiment are correct genome and proteome annotations. Protein databases compiled from high quality empirical protein identifications that are in turn based on correct gene models increase the correctness, sensitivity, and quantitative accuracy of systems biology genome-scale experiments. RESULTS: In this manuscript, we present the Drosophila melanogaster PeptideAtlas, a fly proteomics and genomics resource of unsurpassed depth. Based on peptide mass spectrometry data collected in our laboratory the portal http://www.drosophila-peptideatlas.org allows querying fly protein data observed with respect to gene model confirmation and splice site verification as well as for the identification of proteotypic peptides suited for targeted proteomics studies. Additionally, the database provides consensus mass spectra for observed peptides along with qualitative and quantitative information about the number of observations of a particular peptide and the sample(s) in which it was observed. CONCLUSION: PeptideAtlas is an open access database for the Drosophila community that has several features and applications that support (1) reduction of the complexity inherently associated with performing targeted proteomic studies, (2) designing and accelerating shotgun proteomics experiments, (3) confirming or questioning gene models, and (4) adjusting gene models such that they are in line with observed Drosophila peptides. While the database consists of proteomic data it is not required that the user is a proteomics expert.

UNLABELLED: MS-BID (MS Biomarker Discovery Platform) is an integrative computational pipeline for biomarker discovery using LC-MS-based comparative proteomic analysis. This platform consists of several computational tools for: (i) detecting peptides in the collected patterns; (ii) matching detected peptides across a number of LC-MS datasets and (iii) selecting discriminatory peptides between classes of samples. AVAILABILITY: MS-BID source codes, binaries and documentations are freely available under LGPL from http://tools.proteomecenter.org/msBID.php.

MOTIVATION: Mass spectrometry experiments in the field of proteomics produce lists containing tens to thousands of identified proteins. With the protein information and property explorer (PIPE), the biologist can acquire functional annotations for these proteins and explore the enrichment of the list, or fraction thereof, with respect to functional classes. These protein lists may be saved for access at a later time or different location. The PIPE is interoperable with the Firegoose and the Gaggle, permitting wide-ranging data exploration and analysis. The PIPE is a rich-client web application which uses AJAX capabilities provided by the Google Web Toolkit, and server-side data storage using Hibernate. AVAILABILITY: http://pipe.systemsbiology.net.

BACKGROUND: Quantitative proteomics holds great promise for identifying proteins that are differentially abundant between populations representing different physiological or disease states. A range of computational tools is now available for both isotopically labeled and label-free liquid chromatography mass spectrometry (LC-MS) based quantitative proteomics. However, they are generally not comparable to each other in terms of functionality, user interfaces, information input/output, and do not readily facilitate appropriate statistical data analysis. These limitations, along with the array of choices, present a daunting prospect for biologists, and other researchers not trained in bioinformatics, who wish to use LC-MS-based quantitative proteomics. RESULTS: We have developed Corra, a computational framework and tools for discovery-based LC-MS proteomics. Corra extends and adapts existing algorithms used for LC-MS-based proteomics, and statistical algorithms, originally developed for microarray data analyses, appropriate for LC-MS data analysis. Corra also adapts software engineering technologies (e.g. Google Web Toolkit, distributed processing) so that computationally intense data processing and statistical analyses can run on a remote server, while the user controls and manages the process from their own computer via a simple web interface. Corra also allows the user to output significantly differentially abundant LC-MS-detected peptide features in a form compatible with subsequent sequence identification via tandem mass spectrometry (MS/MS). We present two case studies to illustrate the application of Corra to commonly performed LC-MS-based biological workflows: a pilot biomarker discovery study of glycoproteins isolated from human plasma samples relevant to type 2 diabetes, and a study in yeast to identify in vivo targets of the protein kinase Ark1 via phosphopeptide profiling. CONCLUSION: The Corra computational framework leverages computational innovation to enable biologists or other researchers to process, analyze and visualize LC-MS data with what would otherwise be a complex and not user-friendly suite of tools. Corra enables appropriate statistical analyses, with controlled false-discovery rates, ultimately to inform subsequent targeted identification of differentially abundant peptides by MS/MS. For the user not trained in bioinformatics, Corra represents a complete, customizable, free and open source computational platform enabling LC-MS-based proteomic workflows, and as such, addresses an unmet need in the LC-MS proteomics field.