Abstract
The evolution of data exchange in Mass Spectrometry spans decades and has ranged from human‐readable text files representing individual scans or collections thereof (McDonald et al., 2004) through the official standard XML‐based (Harold, Means, & Udemadu, 2005) data interchange standard (Deutsch, 2012), to increasingly compressed (Teleman et al., 2014) variants of this standard sometimes requiring purely binary adjunct files (Römpp et al., 2011). While the desire to maintain even partial human readability is understandable, the inherent mismatch between XML's textual and irregular format relative to the numeric and highly regular nature of actual spectral data, along with the explosive growth in dataset scales and the resulting need for efficient (binary and indexed) access has led to a phenomenon referred to as “technical drift” (Davis, 2013). While the drift is being continuously corrected using adjunct formats, compression schemes, and programs (Röst et al., 2015), we propose that the future of Mass Spectrometry Exchange Formats lies in the continued reliance and development of the PSI‐MS (Mayer et al., 2014) controlled vocabulary, along with an expedited shift to an alternative, thriving and well‐supported ecosystem for scientific data‐exchange, storage, and access in binary form, namely that of HDF5 (Koranne, 2011). Indeed, pioneering efforts to leverage this universal, binary, and hierarchical data‐format have already been published (Wilhelm et al., 2012; Rübel et al., 2013) though they have under‐utilized self‐description, a key property shared by HDF5 and XML. We demonstrate that a straightforward usage of plain (“vanilla”) HDF5 yields immediate returns including, but not limited to, highly efficient data access, platform independent data viewers, a variety of libraries (Collette, 2014) for data retrieval and manipulation in many programming languages and remote data access through comprehensive RESTful data‐servers. © 2016 The Authors. Mass Spectrometry Reviews published by Wiley Periodicals, Inc. Mass Spec Rev 36:668–673, 2017
Highlights
As a general rule, mass spectrometers produce output stored in manufacturer-specific proprietary file formats
Beyond the inherently reduced accessibility of proprietary binary files the formats and tools used are often burdened with requirements for backward compatibility bridging decades, as well as software dependencies precluding the use of vendorprovided tools on UNIX-like systems central to many highperformance data analytic pipelines. In response to this situation, the proteomics community set about to define a controlled vocabulary for mass spectrometry data as well as a standard format for data interchange. It settled on the PSI-MS for the controlled vocabulary: expressed as a 17,000 Line OBO v1.2 file (Open Biomedical Ontologies [Smith et al, 2007]) this extremely detailed and increasingly comprehensive formal standard defines and inter-relates most key concepts in the field of mass-spectrometry and mass-spectrometry-based proteomics (e.g., MS:1000628 is the formal accession for a “basepeak chromatogram” which is a kind of MS:1000810, i.e., “mass chromatogram”)
Following the adoption of mzML by the mass spectrometry community, a gulf of ever increasing size has emerged between the performance characteristics of XML and the requirements of sharing and efficiently accessing data sets: software systems built around the XML-based format suffer from a mismatch between the data being stored and the storage format being used, with a resulting penalty in terms of file size and performance characteristics
Summary
Mass spectrometers produce output stored in manufacturer-specific proprietary file formats. A combination of the need to efficiently store ever larger data sets, regulatory In response to this situation, the proteomics community set about to define a controlled vocabulary for mass spectrometry data as well as a standard format for data interchange. The official data interchange format that was selected by the community, namely mzML (Deutsch, 2008), does provide this mapping, and does so in every instance (i.e., in every individual mzML file). This is because the PSI community desired the standard to be self-describing, as well as universally supported and human readable. These requirements led to the choice of XML as the underlying representation technology for the mzML format
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
