The mzTab data exchange format: communicating mass-spectrometry-based proteomics and metabolomics experimental results to a wider audience.

The HUPO Proteomics Standards Initiative has developed several standardized data formats to facilitate data sharing in mass spectrometry (MS)-based proteomics. These allow researchers to report their complete results in a unified way. However, at present, there is no format to describe the final qualitative and quantitative results for proteomics and metabolomics experiments in a simple tabular format. Many downstream analysis use cases are only concerned with the final results of an experiment and require an easily accessible format, compatible with tools such as Microsoft Excel or R. We developed the mzTab file format for MS-based proteomics and metabolomics results to meet this need. mzTab is intended as a lightweight supplement to the existing standard XML-based file formats (mzML, mzIdentML, mzQuantML), providing a comprehensive summary, similar in concept to the supplemental material of a scientific publication. mzTab files can contain protein, peptide, and small molecule identifications together with experimental metadata and basic quantitative information. The format is not intended to store the complete experimental evidence but provides mechanisms to report results at different levels of detail. These range from a simple summary of the final results to a representation of the results including the experimental design. This format is ideally suited to make MS-based proteomics and metabolomics results available to a wider biological community outside the field of MS. Several software tools for proteomics and metabolomics have already adapted the format as an output format. The comprehensive mzTab specification document and extensive additional documentation can be found online.

High performance computational analysis of large-scale proteome data sets to assess incremental contribution to coverage of the human genome.

Computational analysis of shotgun proteomics data can now be performed in a completely automated and statistically rigorous way, as exemplified by the freely available MaxQuant environment. The sophisticated algorithms involved and the sheer amount of data translate into very high computational demands. Here we describe parallelization and memory optimization of the MaxQuant software with the aim of executing it on a large computer cluster. We analyze and mitigate bottlenecks in overall performance and find that the most time-consuming algorithms are those detecting peptide features in the MS(1) data as well as the fragment spectrum search. These tasks scale with the number of raw files and can readily be distributed over many CPUs as long as memory access is properly managed. Here we compared the performance of a parallelized version of MaxQuant running on a standard desktop, an I/O performance optimized desktop computer ("game computer"), and a cluster environment. The modified gaming computer and the cluster vastly outperformed a standard desktop computer when analyzing more than 1000 raw files. We apply our high performance platform to investigate incremental coverage of the human proteome by high resolution MS data originating from in-depth cell line and cancer tissue proteome measurements.

A systematic investigation into the nature of tryptic HCD spectra.

Modern mass spectrometry-based proteomics can produce millions of peptide fragmentation spectra, which are automatically identified in databases using sequence-specific b- or y-ions. Proteomics projects have mainly been performed with low resolution collision-induced dissociation (CID) in ion traps and beam-type fragmentation on triple quadrupole and QTOF instruments. Recently, the latter has also become available with Orbitrap instrumentation as higher energy collisional dissociation (HCD), routinely providing full mass range fragmentation with high mass accuracy. To systematically study the nature of HCD spectra, we made use of a large scale data set of tryptic peptides identified with an FDR of 0.0001, from which we extract a subset of more than 16,000 that have little or no contribution from cofragmented precursors. We employed a newly developed computer-assisted "Expert System", which distills our experience and literature knowledge about fragmentation pathways. It aims to automatically annotate the peaks in high mass accuracy fragment spectra while strictly controlling the false discovery rate. Using this Expert System we determined that sequence specific regular ions covering the entire sequence were present for almost all peptides with up to 10 amino acids (median 100%). Peptides up to 20 amino acid length contained sufficient fragmentation to cover 80% of the sequence. Internal fragments are common in HCD spectra but not in high resolution CID spectra (10% vs 1%). The low mass region contains abundant immonium ions (6% of fragment ion intensity), the characteristic a(2), b(2) ion pair (72% of spectra), side chain fragments and reporter ions for peptide modifications such as tyrosine phosphorylation. B- and y-ions account for only 20% of fragment ions by number but 53% by ion intensity. Overall, 84% of the fragment ion intensity was unambiguously explainable. Thus high mass accuracy HCD and CID data are near comprehensively and automatically interpretable.

Expert system for computer-assisted annotation of MS/MS spectra.

An important step in mass spectrometry (MS)-based proteomics is the identification of peptides by their fragment spectra. Regardless of the identification score achieved, almost all tandem-MS (MS/MS) spectra contain remaining peaks that are not assigned by the search engine. These peaks may be explainable by human experts but the scale of modern proteomics experiments makes this impractical. In computer science, Expert Systems are a mature technology to implement a list of rules generated by interviews with practitioners. We here develop such an Expert System, making use of literature knowledge as well as a large body of high mass accuracy and pure fragmentation spectra. Interestingly, we find that even with high mass accuracy data, rule sets can quickly become too complex, leading to over-annotation. Therefore we establish a rigorous false discovery rate, calculated by random insertion of peaks from a large collection of other MS/MS spectra, and use it to develop an optimized knowledge base. This rule set correctly annotates almost all peaks of medium or high abundance. For high resolution HCD data, median intensity coverage of fragment peaks in MS/MS spectra increases from 58% by search engine annotation alone to 86%. The resulting annotation performance surpasses a human expert, especially on complex spectra such as those of larger phosphorylated peptides. Our system is also applicable to high resolution collision-induced dissociation data. It is available both as a part of MaxQuant and via a webserver that only requires an MS/MS spectrum and the corresponding peptides sequence, and which outputs publication quality, annotated MS/MS spectra (www.biochem.mpg.de/mann/tools/). It provides expert knowledge to beginners in the field of MS-based proteomics and helps advanced users to focus on unusual and possibly novel types of fragment ions.

System-wide perturbation analysis with nearly complete coverage of the yeast proteome by single-shot ultra HPLC runs on a bench top Orbitrap.

Yeast remains an important model for systems biology and for evaluating proteomics strategies. In-depth shotgun proteomics studies have reached nearly comprehensive coverage, and rapid, targeted approaches have been developed for this organism. Recently, we demonstrated that single LC-MS/MS analysis using long columns and gradients coupled to a linear ion trap Orbitrap instrument had an unexpectedly large dynamic range of protein identification (Thakur, S. S., Geiger, T., Chatterjee, B., Bandilla, P., Frohlich, F., Cox, J., and Mann, M. (2011) Deep and highly sensitive proteome coverage by LC-MS/MS without prefractionation. Mol. Cell Proteomics 10, 10.1074/mcp.M110.003699). Here we couple an ultra high pressure liquid chromatography system to a novel bench top Orbitrap mass spectrometer (Q Exactive) with the goal of nearly complete, rapid, and robust analysis of the yeast proteome. Single runs of filter-aided sample preparation (FASP)-prepared and LysC-digested yeast cell lysates identified an average of 3923 proteins. Combined analysis of six single runs improved these values to more than 4000 identified proteins/run, close to the total number of proteins expressed under standard conditions, with median sequence coverage of 23%. Because of the absence of fractionation steps, only minuscule amounts of sample are required. Thus the yeast model proteome can now largely be covered within a few hours of measurement time and at high sensitivity. Median coverage of proteins in Kyoto Encyclopedia of Genes and Genomes pathways with at least 10 members was 88%, and pathways not covered were not expected to be active under the conditions used. To study perturbations of the yeast proteome, we developed an external, heavy lysine-labeled SILAC yeast standard representing different proteome states. This spike-in standard was employed to measure the heat shock response of the yeast proteome. Bioinformatic analysis of the heat shock response revealed that translation-related functions were down-regulated prominently, including nucleolar processes. Conversely, stress-related pathways were up-regulated. The proteomic technology described here is straightforward, rapid, and robust, potentially enabling widespread use in the yeast and other biological research communities.

Andromeda: a peptide search engine integrated into the MaxQuant environment.

A key step in mass spectrometry (MS)-based proteomics is the identification of peptides in sequence databases by their fragmentation spectra. Here we describe Andromeda, a novel peptide search engine using a probabilistic scoring model. On proteome data, Andromeda performs as well as Mascot, a widely used commercial search engine, as judged by sensitivity and specificity analysis based on target decoy searches. Furthermore, it can handle data with arbitrarily high fragment mass accuracy, is able to assign and score complex patterns of post-translational modifications, such as highly phosphorylated peptides, and accommodates extremely large databases. The algorithms of Andromeda are provided. Andromeda can function independently or as an integrated search engine of the widely used MaxQuant computational proteomics platform and both are freely available at www.maxquant.org. The combination enables analysis of large data sets in a simple analysis workflow on a desktop computer. For searching individual spectra Andromeda is also accessible via a web server. We demonstrate the flexibility of the system by implementing the capability to identify cofragmented peptides, significantly improving the total number of identified peptides.

High-accuracy identification and bioinformatic analysis of in vivo protein phosphorylation sites in yeast.

Protein phosphorylation is a fundamental regulatory mechanism that affects many cell signaling processes. Using high-accuracy MS and stable isotope labeling in cell culture-labeling, we provide a global view of the Saccharomyces cerevisiae phosphoproteome, containing 3620 phosphorylation sites mapped to 1118 proteins, representatively covering the yeast kinome and a multitude of transcription factors. We show that a single false discovery rate for all peptide identifications significantly overestimates occurrence of rare modifications, such as tyrosine phosphorylation in yeast. The identified phosphorylation sites are predominantly located on irregularly structured and accessible protein regions. We found high evolutionary conservation of phosphorylated proteins and a large overlap of significantly over-represented motifs with the human phosphoproteome. Nevertheless, phosphorylation events at the site level were not highly conserved between yeast and higher eukaryotes, which points to metazoan-specific kinase and substrate families. We constructed a yeast-specific phosphorylation sites predictor on the basis of a support vector machine, which - together with the yeast phosphorylation data - is integrated into the PHOSIDA database (www.phosida.com).