Identifying neural signatures mediating behavioral symptoms and psychosis onset: High-dimensional whole brain functional mediation analysis.

Along the pathway from behavioral symptoms to the development of psychotic disorders sits the multivariate mediating brain. The functional organization and structural topography of large-scale multivariate neural mediators among patients with brain disorders, however, are not well understood. Here, we design a high-dimensional brain-wide functional mediation framework to investigate brain regions that intermediate between baseline behavioral symptoms and future conversion to full psychosis among individuals at clinical high risk (CHR). Using resting-state functional magnetic resonance imaging (fMRI) data from 263 CHR subjects, we extract an α brain atlas and a ß brain atlas: the former underlines brain areas associated with prodromal symptoms and the latter highlights brain areas associated with disease onset. In parallel, we identify and separate mediators that potentially positively and negatively mediate symptoms and psychosis, respectively, and quantify the effect of each neural mediator on disease development. Taken together, these results paint a brain-wide picture of neural markers that are potentially mediating behavioral symptoms and the development of psychotic disorders; additionally, they underscore a statistical framework that is useful to uncover large-scale intermediating variables in a regulatory biological system.

The case of the disappearing drug target.

In this issue of Molecular Cell, Hoshi et al. (2010) report two examples in which small molecule inhibitors are rendered ineffective when their kinase targets are involved in protein-protein interactions, highlighting differences between in vivo and in vitro inhibition kinetics.

Structures of the Gß-CCT and PhLP1-Gß-CCT complexes reveal a mechanism for G-protein ß-subunit folding and Gßγ dimer assembly.

G-protein signaling depends on the ability of the individual subunits of the G-protein heterotrimer to assemble into a functional complex. Formation of the G-protein ßγ (Gßγ) dimer is particularly challenging because it is an obligate dimer in which the individual subunits are unstable on their own. Recent studies have revealed an intricate chaperone system that brings Gß and Gγ together. This system includes cytosolic chaperonin containing TCP-1 (CCT; also called TRiC) and its cochaperone phosducin-like protein 1 (PhLP1). Two key intermediates in the Gßγ assembly process, the Gß-CCT and the PhLP1-Gß-CCT complexes, were isolated and analyzed by a hybrid structural approach using cryo-electron microscopy, chemical cross-linking coupled with mass spectrometry, and unnatural amino acid cross-linking. The structures show that Gß interacts with CCT in a near-native state through interactions of the Gγ-binding region of Gß with the CCTγ subunit. PhLP1 binding stabilizes the Gß fold, disrupting interactions with CCT and releasing a PhLP1-Gß dimer for assembly with Gγ. This view provides unique insight into the interplay between CCT and a cochaperone to orchestrate the folding of a protein substrate.

Probabilistic Generation of Mass Spectrometry Molecular Abundance Variance for Case and Control Replicates.

Shotgun differential mass spectrometry, the untargeted discovery of statistically significant differences between two or more samples, is a popular application with potential to advance biomarker detection, disease diagnostics, and other health objectives. Although many methods have been proposed, few have been quantitatively evaluated. The lack of ground truth data for shotgun difference detection limits quantitative evaluation and algorithmic advancement. While public mass-spectrometry data sets of single samples abound, data sets with more than one sample are rare, and data sets with the thousands of samples necessary to capture the complexity of real world populations are nonexistent due to technological and cost limitations. We present MSabundanceSIM, novel software for simulating any number of molecular samples based on one or a few real world data sets. The software uses a probabilistic model to generate case and control populations, with intuitive user parameters for tuning. We demonstrate variability by comparing to a real world data set over a range of abundances with differing biological and experimental variation coefficients. MSabundanceSIM is implemented in Ruby, is freely available, requires no external dependencies, and is suitable for a range of applications.

LC-MS alignment in theory and practice: a comprehensive algorithmic review.

Liquid chromatography-mass spectrometry is widely used for comparative replicate sample analysis in proteomics, lipidomics and metabolomics. Before statistical comparison, registration must be established to match corresponding analytes from run to run. Alignment, the most popular correspondence approach, consists of constructing a function that warps the content of runs to most closely match a given reference sample. To date, dozens of correspondence algorithms have been proposed, creating a daunting challenge for practitioners in algorithm selection. Yet, existing reviews have highlighted only a few approaches. In this review, we describe 50 correspondence algorithms to facilitate practical algorithm selection. We elucidate the motivation for correspondence and analyze the limitations of current approaches, which include prohibitive runtimes, numerous user parameters, model limitations and the need for reference samples. We suggest and describe a paradigm shift for overcoming current correspondence limitations by building on known liquid chromatography-mass spectrometry behavior.

Whole blood and urine bioactive Hepcidin-25 determination using liquid chromatography mass spectrometry.

Hepcidin is a small cysteine-rich signaling peptide that regulates blood serum iron concentrations [1-4]. Patients with chronic inflammation are known to have elevated levels of hepcidin in their blood and urine and often suffer from anemia as a result [5-10]. Measuring and quantifying the amount of active hepcidin in blood and urine can help to determine the cause and severity of the anemia thereby helping physicians determine the correct course of treatment [11-16]. We have developed a simple technique to isolate, chemically modify, and concentrate hepcidin from blood and urine coupled to high-pressure liquid chromatography mass spectrometry that can accurately and reproducibly measure and quantify the active hormone.

Controlling for confounding variables in MS-omics protocol: why modularity matters.

As the field of bioinformatics research continues to grow, more and more novel techniques are proposed to meet new challenges and improvements upon solutions to long-standing problems. These include data processing techniques and wet lab protocol techniques. Although the literature is consistently thorough in experimental detail and variable-controlling rigor for wet lab protocol techniques, bioinformatics techniques tend to be less described and less controlled. As the validation or rejection of hypotheses rests on the experiment's ability to isolate and measure a variable of interest, we urge the importance of reducing confounding variables in bioinformatics techniques during mass spectrometry experimentation.

JAMSS: proteomics mass spectrometry simulation in Java.

UNLABELLED: Countless proteomics data processing algorithms have been proposed, yet few have been critically evaluated due to lack of labeled data (data with known identities and quantities). Although labeling techniques exist, they are limited in terms of confidence and accuracy. In silico simulators have recently been used to create complex data with known identities and quantities. We propose Java Mass Spectrometry Simulator (JAMSS): a fast, self-contained in silico simulator capable of generating simulated MS and LC-MS runs while providing meta information on the provenance of each generated signal. JAMSS improves upon previous in silico simulators in terms of its ease to install, minimal parameters, graphical user interface, multithreading capability, retention time shift model and reproducibility. AVAILABILITY AND IMPLEMENTATION: The simulator creates mzML 1.1.0. It is open source software licensed under the GPLv3. The software and source are available at https://github.com/optimusmoose/JAMSS.

Automated structural classification of lipids by machine learning.

MOTIVATION: Modern lipidomics is largely dependent upon structural ontologies because of the great diversity exhibited in the lipidome, but no automated lipid classification exists to facilitate this partitioning. The size of the putative lipidome far exceeds the number currently classified, despite a decade of work. Automated classification would benefit ongoing classification efforts by decreasing the time needed and increasing the accuracy of classification while providing classifications for mass spectral identification algorithms. RESULTS: We introduce a tool that automates classification into the LIPID MAPS ontology of known lipids with >95% accuracy and novel lipids with 63% accuracy. The classification is based upon simple chemical characteristics and modern machine learning algorithms. The decision trees produced are intelligible and can be used to clarify implicit assumptions about the current LIPID MAPS classification scheme. These characteristics and decision trees are made available to facilitate alternative implementations. We also discovered many hundreds of lipids that are currently misclassified in the LIPID MAPS database, strongly underscoring the need for automated classification. AVAILABILITY AND IMPLEMENTATION: Source code and chemical characteristic lists as SMARTS search strings are available under an open-source license at https://www.github.com/princelab/lipid_classifier.

A coherent mathematical characterization of isotope trace extraction, isotopic envelope extraction, and LC-MS correspondence.

BACKGROUND: Liquid chromatography-mass spectrometry is a popular technique for high-throughput protein, lipid, and metabolite comparative analysis. Such statistical comparison of millions of data points requires the generation of an inter-run correspondence. Though many techniques for generating this correspondence exist, few if any, address certain well-known run-to-run LC-MS behaviors such as elution order swaps, unbounded retention time swaps, missing data, and significant differences in abundance. Moreover, not all extant correspondence methods leverage the rich discriminating information offered by isotope envelope extraction informed by isotope trace extraction. To date, no attempt has been made to create a formal generalization of extant algorithms for these problems. RESULTS: By enumerating extant objective functions for these problems, we elucidate discrepancies between known LC-MS data behavior and extant approaches. We propose novel objective functions that more closely model known LC-MS behavior. CONCLUSIONS: Through instantiating the proposed objective functions in the form of novel algorithms, practitioners can more accurately capture the known behavior of isotope traces, isotopic envelopes, and replicate LC-MS data, ultimately providing for improved quantitative accuracy.

Current controlled vocabularies are insufficient to uniquely map molecular entities to mass spectrometry signal.

BACKGROUND: The comparison of analyte mass spectrometry precursor (MS1) signal is central to many proteomic (and other -omic) workflows. Standard vocabularies for mass spectrometry exist and provide good coverage for most experimental applications yet are insufficient for concise and unambiguous description of data concepts spanning the range of signal provenance from a molecular perspective (e.g. from charged peptides down to fine isotopes). Without a standard unambiguous nomenclature, literature searches, algorithm reproducibility and algorithm evaluation for MS-omics data processing are nearly impossible. RESULTS: We show how terms from current official ontologies are too vague or ambiguous to explicitly map molecular entities to MS signals and we illustrate the inconsistency and ambiguity of current colloquially used terms. We also propose a set of terms for MS1 signal that uniquely, succinctly and intuitively describe data concepts spanning the range of signal provenance from full molecule downs to fine isotopes. We suggest that additional community discussion of these terms should precede any further standardization efforts. We propose a novel nomenclature that spans the range of the required granularity to describe MS data processing from the perspective of the molecular provenance of the MS signal. CONCLUSIONS: The proposed nomenclature provides a chain of succinct and unique terms spanning the signal created by a charged molecule down through each of its constituent subsignals. We suggest that additional community discussion of these terms should precede any further standardization efforts.

Programmed cell death protein 5 interacts with the cytosolic chaperonin containing tailless complex polypeptide 1 (CCT) to regulate ß-tubulin folding.

Programmed cell death protein 5 (PDCD5) has been proposed to act as a pro-apoptotic factor and tumor suppressor. However, the mechanisms underlying its apoptotic function are largely unknown. A proteomics search for binding partners of phosducin-like protein, a co-chaperone for the cytosolic chaperonin containing tailless complex polypeptide 1 (CCT), revealed a robust interaction between PDCD5 and CCT. PDCD5 formed a complex with CCT and ß-tubulin, a key CCT-folding substrate, and specifically inhibited ß-tubulin folding. Cryo-electron microscopy studies of the PDCD5·CCT complex suggested a possible mechanism of inhibition of ß-tubulin folding. PDCD5 bound the apical domain of the CCTß subunit, projecting above the folding cavity without entering it. Like PDCD5, ß-tubulin also interacts with the CCTß apical domain, but a second site is found at the sensor loop deep within the folding cavity. These orientations of PDCD5 and ß-tubulin suggest that PDCD5 sterically interferes with ß-tubulin binding to the CCTß apical domain and inhibits ß-tubulin folding. Given the importance of tubulins in cell division and proliferation, PDCD5 might exert its apoptotic function at least in part through inhibition of ß-tubulin folding.

The genomes, proteomes, and structures of three novel phages that infect the Bacillus cereus group and carry putative virulence factors.

This article reports the results of studying three novel bacteriophages, JL, Shanette, and Basilisk, which infect the pathogen Bacillus cereus and carry genes that may contribute to its pathogenesis. We analyzed host range and superinfection ability, mapped their genomes, and characterized phage structure by mass spectrometry and transmission electron microscopy (TEM). The JL and Shanette genomes were 96% similar and contained 217 open reading frames (ORFs) and 220 ORFs, respectively, while Basilisk has an unrelated genome containing 138 ORFs. Mass spectrometry revealed 23 phage particle proteins for JL and 15 for Basilisk, while only 11 and 4, respectively, were predicted to be present by sequence analysis. Structural protein homology to well-characterized phages suggested that JL and Shanette were members of the family Myoviridae, which was confirmed by TEM. The third phage, Basilisk, was similar only to uncharacterized phages and is an unrelated siphovirus. Cryogenic electron microscopy of this novel phage revealed a T=9 icosahedral capsid structure with the major capsid protein (MCP) likely having the same fold as bacteriophage HK97 MCP despite the lack of sequence similarity. Several putative virulence factors were encoded by these phage genomes, including TerC and TerD involved in tellurium resistance. Host range analysis of all three phages supports genetic transfer of such factors within the B. cereus group, including B. cereus, B. anthracis, and B. thuringiensis. This study provides a basis for understanding these three phages and other related phages as well as their contributions to the pathogenicity of B. cereus group bacteria. Importance: The Bacillus cereus group of bacteria contains several human and plant pathogens, including B. cereus, B. anthracis, and B. thuringiensis. Phages are intimately linked to the evolution of their bacterial hosts and often provide virulence factors, making the study of B. cereus phages important to understanding the evolution of pathogenic strains. Herein we provide the results of detailed study of three novel B. cereus phages, two highly related myoviruses (JL and Shanette) and an unrelated siphovirus (Basilisk). The detailed characterization of host range and superinfection, together with results of genomic, proteomic, and structural analyses, reveal several putative virulence factors as well as the ability of these phages to infect different pathogenic species.

Massifquant: open-source Kalman filter-based XC-MS isotope trace feature detection.

MOTIVATION: Isotope trace (IT) detection is a fundamental step for liquid or gas chromatography mass spectrometry (XC-MS) data analysis that faces a multitude of technical challenges on complex samples. The Kalman filter (KF) application to IT detection addresses some of these challenges; it discriminates closely eluting ITs in the m/z dimension, flexibly handles heteroscedastic m/z variances and does not bin the m/z axis. Yet, the behavior of this KF application has not been fully characterized, as no cost-free open-source implementation exists and incomplete evaluation standards for IT detection persist. RESULTS: Massifquant is an open-source solution for KF IT detection that has been subjected to novel and rigorous methods of performance evaluation. The presented evaluation with accompanying annotations and optimization guide sets a new standard for comparative IT detection. Compared with centWave, matchedFilter and MZMine2-alternative IT detection engines-Massifquant detected more true ITs in a real LC-MS complex sample, especially low-intensity ITs. It also offers competitive specificity and equally effective quantitation accuracy. AVAILABILITY AND IMPLEMENTATION: Massifquant is integrated into XCMS with GPL license ≥ 2.0 and hosted by Bioconductor: http://bioconductor.org. Annotation data are archived at http://hdl.lib.byu.edu/1877/3232. Parameter optimization code and documentation is hosted at https://github.com/topherconley/optimize-it.

Proteomics, lipidomics, metabolomics: a mass spectrometry tutorial from a computer scientist's point of view.

BACKGROUND: For decades, mass spectrometry data has been analyzed to investigate a wide array of research interests, including disease diagnostics, biological and chemical theory, genomics, and drug development. Progress towards solving any of these disparate problems depends upon overcoming the common challenge of interpreting the large data sets generated. Despite interim successes, many data interpretation problems in mass spectrometry are still challenging. Further, though these challenges are inherently interdisciplinary in nature, the significant domain-specific knowledge gap between disciplines makes interdisciplinary contributions difficult. RESULTS: This paper provides an introduction to the burgeoning field of computational mass spectrometry. We illustrate key concepts, vocabulary, and open problems in MS-omics, as well as provide invaluable resources such as open data sets and key search terms and references. CONCLUSIONS: This paper will facilitate contributions from mathematicians, computer scientists, and statisticians to MS-omics that will fundamentally improve results over existing approaches and inform novel algorithmic solutions to open problems.

Metriculator: quality assessment for mass spectrometry-based proteomics.

SUMMARY: Quality control in mass spectrometry-based proteomics remains subjective, labor-intensive and inconsistent between laboratories. We introduce Metriculator, a software designed to facilitate long-term storage of extensive performance metrics as introduced by NIST in 2010. Metriculator features a web interface that generates interactive comparison plots for contextual understanding of metric values and an automated metric generation toolkit. The comparison plots are designed for at-a-glance determination of outliers and trends in the datasets, together with relevant statistical comparisons. Easy-to-use quantitative comparisons and a framework for integration plugins will encourage a culture of quality assurance within the proteomics community. AVAILABILITY AND IMPLEMENTATION: Available under the MIT license at http://github.com/princelab/metriculator.

Statistical agglomeration: peak summarization for direct infusion lipidomics.

MOTIVATION: Quantification of lipids is a primary goal in lipidomics. In direct infusion/injection (or shotgun) lipidomics, accurate downstream identification and quantitation requires accurate summarization of repetitive peak measurements. Imprecise peak summarization multiplies downstream error by propagating into species identification and intensity estimation. To our knowledge, this is the first analysis of direct infusion peak summarization in the literature. RESULTS: We present two novel peak summarization algorithms for direct infusion samples and compare them with an off-machine ad hoc summarization algorithm as well as with the propriety Xcalibur algorithm. Our statistical agglomeration algorithm reduces peakwise error by 38% mass/charge (m/z) and 44% (intensity) compared with the ad hoc method over three datasets. Pointwise error is reduced by 23% (m/z). Compared with Xcalibur, our statistical agglomeration algorithm produces 68% less m/z error and 51% less intensity error on average on two comparable datasets. AVAILABILITY: The source code for Statistical Agglomeration and the datasets used are freely available for non-commercial purposes at https://github.com/optimusmoose/statistical_agglomeration. Modified Bin Aggolmeration is freely available in MSpire, an open source mass spectrometry package at https://github.com/princelab/mspire/.

Mitochondrial fission mediates ceramide-induced metabolic disruption in skeletal muscle.

Ceramide is a sphingolipid that serves as an important second messenger in an increasing number of stress-induced pathways. Ceramide has long been known to affect the mitochondria, altering both morphology and physiology. We sought to assess the impact of ceramide on skeletal muscle mitochondrial structure and function. A primary observation was the rapid and dramatic division of mitochondria in ceramide-treated cells. This effect is likely to be a result of increased Drp1 (dynamin-related protein 1) action, as ceramide increased Drp1 expression and Drp1 inhibition prevented ceramide-induced mitochondrial fission. Further, we found that ceramide treatment reduced mitochondrial O2 consumption (i.e. respiration) in cultured myotubes and permeabilized red gastrocnemius muscle fibre bundles. Ceramide treatment also increased H2O2 levels and reduced Akt/PKB (protein kinase B) phosphorylation in myotubes. However, inhibition of mitochondrial fission via Drp1 knockdown completely protected the myotubes and fibre bundles from ceramide-induced metabolic disruption, including maintained mitochondrial respiration, reduced H2O2 levels and unaffected insulin signalling. These data suggest that the forced and sustained mitochondrial fission that results from ceramide accrual may alter metabolic function in skeletal muscle, which is a prominent site not only of energy demand (via the mitochondria), but also of ceramide accrual with weight gain.

Mspire-Simulator: LC-MS shotgun proteomic simulator for creating realistic gold standard data.

The most important step in any quantitative proteomic pipeline is feature detection (aka peak picking). However, generating quality hand-annotated data sets to validate the algorithms, especially for lower abundance peaks, is nearly impossible. An alternative for creating gold standard data is to simulate it with features closely mimicking real data. We present Mspire-Simulator, a free, open-source shotgun proteomic simulator that goes beyond previous simulation attempts by generating LC-MS features with realistic m/z and intensity variance along with other noise components. It also includes machine-learned models for retention time and peak intensity prediction and a genetic algorithm to custom fit model parameters for experimental data sets. We show that these methods are applicable to data from three different mass spectrometers, including two fundamentally different types, and show visually and analytically that simulated peaks are nearly indistinguishable from actual data. Researchers can use simulated data to rigorously test quantitation software, and proteomic researchers may benefit from overlaying simulated data on actual data sets.

Sequence and structural characterization of great salt lake bacteriophage CW02, a member of the T7-like supergroup.

Halophage CW02 infects a Salinivibrio costicola-like bacterium, SA50, isolated from the Great Salt Lake. Following isolation, cultivation, and purification, CW02 was characterized by DNA sequencing, mass spectrometry, and electron microscopy. A conserved module of structural genes places CW02 in the T7 supergroup, members of which are found in diverse aquatic environments, including marine and freshwater ecosystems. CW02 has morphological similarities to viruses of the Podoviridae family. The structure of CW02, solved by cryogenic electron microscopy and three-dimensional reconstruction, enabled the fitting of a portion of the bacteriophage HK97 capsid protein into CW02 capsid density, thereby providing additional evidence that capsid proteins of tailed double-stranded DNA phages have a conserved fold. The CW02 capsid consists of bacteriophage lambda gpD-like densities that likely contribute to particle stability. Turret-like densities were found on icosahedral vertices and may represent a unique adaptation similar to what has been seen in other extremophilic viruses that infect archaea, such as Sulfolobus turreted icosahedral virus and halophage SH1.