File Formats Commonly Used in Mass Spectrometry Proteomics

The application of mass spectrometry (MS) to the analysis of proteomes has enabled the high-throughput identification and abundance measurement of hundreds to thousands of proteins per experiment. However, the formidable informatics challenge associated with analyzing MS data has required a wide variety of data file formats to encode the complex data types associated with MS workflows. These formats encompass the encoding of input instruction for instruments, output products of the instruments, and several levels of information and results used by and produced by the informatics analysis tools. A brief overview of the most common file formats in use today is presented here, along with a discussion of related topics. The application of mass spectrometry (MS) to the analysis of proteomes has enabled the high-throughput identification and abundance measurement of hundreds to thousands of proteins per experiment. However, the formidable informatics challenge associated with analyzing MS data has required a wide variety of data file formats to encode the complex data types associated with MS workflows. These formats encompass the encoding of input instruction for instruments, output products of the instruments, and several levels of information and results used by and produced by the informatics analysis tools. A brief overview of the most common file formats in use today is presented here, along with a discussion of related topics. Mass spectrometry (MS) has accelerated the field of proteomics by enabling the high-throughput identification and abundance measurement of hundreds to thousands of proteins per experiment (1Beck M. Claassen M. Aebersold R. Comprehensive proteomics.Curr. Opin. Biotechnol. 2011; 22: 3-8Crossref PubMed Scopus (72) Google Scholar). The most common workflow for tandem mass spectrometry (MS/MS) (2McCormack A.L. Schieltz D.M. Goode B. Yang S. Barnes G. Drubin D. Yates 3rd, J.R. Direct analysis and identification of proteins in mixtures by LC/MS/MS and database searching at the low-femtomole level.Anal. Chem. 1997; 69: 767-776Crossref PubMed Scopus (446) Google Scholar) generally begins with the isolation of proteins from an original sample, digestion of the proteins into peptides with an enzyme such as trypsin, and separation of the peptides into multiple fractions to reduce the complexity in each fraction. Each fraction is then subjected to liquid chromatography (LC), ionized, and injected into the mass spectrometer. Individual species of peptide ions are isolated and fragmented to generate a fragment ion spectrum which may then be identified via software. In nearly all of these high-throughput workflows, extensive analysis with software is required in order to translate the mass spectra into peptide identifications and perform abundance measurements (3Deutsch E.W. Lam H. Aebersold R. Data analysis and bioinformatics tools for tandem mass spectrometry in proteomics.Physiol. Genomics. 2008; 33: 18-25Crossref PubMed Scopus (121) Google Scholar). There are a wide variety of software tools available to assist with this analysis, including open-source software as well as proprietary and commercial products. As a result of efforts to enable the movement of complex data types among analysis tools and the sharing of data and results with others in the community (4Methods Nature Thou shalt share your data.Nat. Methods. 2008; 5: 209Crossref Scopus (26) Google Scholar), a wide variety of data formats have emerged. These formats may be broadly separated into open formats and proprietary formats. Open formats enable improved data sharing by allowing the data to be read by a variety of software tools without licensing restrictions. Open formats can be further separated into three categories: official standards, de facto standards, and other formats. Official standards are approved by a standards body, typically after a formal process of review and refinement, whereas de facto standards lack any official approval but are widely used by a large number of software tools and generally accepted as being a preferred mechanism of data exchange. Formats not falling into either of these two categories are simply referred to as other formats in this review. Each of the major instrument vendors uses its own proprietary formats, continually updating the formats to support new features of their instruments. Open formats are generally created by the developers of analysis software and databases in order to enable the exchange of data between tools. Some formats have been developed by a single lab and are oriented around that lab's software, whereas other formats have emerged after a long process of collaborative development by a diverse group of contributors, often under the organization of a standards development group. The largest and most active standards development group in MS proteomics is the Human Proteome Organization (HUPO) 1The abbreviations used are:APIapplication programming interfaceAPMLAnnotated Peptide Markup LanguageCVcontrolled vocabularyFuGEFunctional Genomics ExperimentHUPOHuman Proteome OrganizationISBInstitute for Systems BiologyLCliquid chromatographyMGFMascot Generic FormatMSmass spectrometryMS/MStandem mass spectrometryPEFFPSI Extended FASTA FormatPSIProteomics Standards InitiativePSMpeptide-spectrum matchRDFresource definition frameworkSRMselected reaction monitoringTPPTrans-Proteomic PipelineTraMLTransitions Markup LanguageTSVtab-separated value. 1The abbreviations used are:APIapplication programming interfaceAPMLAnnotated Peptide Markup LanguageCVcontrolled vocabularyFuGEFunctional Genomics ExperimentHUPOHuman Proteome OrganizationISBInstitute for Systems BiologyLCliquid chromatographyMGFMascot Generic FormatMSmass spectrometryMS/MStandem mass spectrometryPEFFPSI Extended FASTA FormatPSIProteomics Standards InitiativePSMpeptide-spectrum matchRDFresource definition frameworkSRMselected reaction monitoringTPPTrans-Proteomic PipelineTraMLTransitions Markup LanguageTSVtab-separated value. Proteomics Standards Initiative (PSI) (5Orchard S. Montechi-Palazzi L. Deutsch E.W. Binz P.A. Jones A.R. Paton N. Pizarro A. Creasy D.M. Wojcik H. of in the of Proteomics Data of the Standards PubMed Scopus Google Scholar). The to from commercial instrument software and software developers and to common exchange formats and information that are then and approved as application programming Peptide Markup Genomics Human Proteome Organization for Systems liquid chromatography Generic mass spectrometry tandem mass spectrometry Extended FASTA Proteomics Standards Initiative definition reaction Markup value. application programming Peptide Markup Genomics Human Proteome Organization for Systems liquid chromatography Generic mass spectrometry tandem mass spectrometry Extended FASTA Proteomics Standards Initiative definition reaction Markup value. an overview of the formats in common use in MS proteomics by software tools. The formats be in but are and are common formats, formats of be the overview of all the formats here, a brief discussion of related to the application and of these formats is There are a large number of used formats in MS proteomics at all of data an overview of all the formats here, by their in proteomics workflows. the and are formats in use to information to mass mass software analysis and formats for and Each of these formats is in this order in the A of the major software tools and that of these formats is in of of the most software tools and and the formats D. M. R. D. open software for proteomics tools 2008; PubMed Scopus Google M. A. A. R. N. A. open-source software for mass 2008; PubMed Scopus Google A. N. Aebersold R. A proteomics analysis open file PubMed Scopus Google E.W. L. D. Lam H. N. B. B. Aebersold R. A of the PubMed Scopus Google H. M. N. A. L. an open-source for 2011; PubMed Scopus Google H. A to process and multiple and mass spectrometry data PubMed Scopus Google Deutsch L. an open for the for sharing Proteome 2011; PubMed Scopus Google J.R. M. an for proteomics data PubMed Scopus Google H. L. proteomics Biotechnol. PubMed Scopus Google H. Deutsch E.W. N. Aebersold R. and of a searching for peptide identification from PubMed Scopus Google G. B. A. A. B. M. software for mass Biotechnol. PubMed Scopus Google S. reaction software for Proteome PubMed Scopus Google R. an open proteomics PubMed Scopus Google B. D.M. N. M. B. R. an open for and analyzing proteomics PubMed Scopus Google M. D. L. H. Deutsch E.W. Aebersold R. a software for and for reaction mass 2011; PubMed Scopus Google B. D. A. H. S. N. Aebersold R. and tools for and mass 2008; PubMed Scopus Google S. R. Jones A.R. A. The a to and share proteomics 2011; PubMed Scopus Google Scholar) Open in a new There are common formats not to for information information to There are broadly two categories of such that and that formats for mass The has developed a of formats for S. G. A. H. R. Paton Jones A.R. The from the Proteomics Standards PubMed Scopus Google Scholar) and S. A. M. Binz P.A. the data A the The PubMed Scopus Google Scholar). The is to encode information related to and to can encode the is and the are from the The are with and information that can be in the As with most formats, is to with an The that a is referred to by its number for its in the which is by a definition and by any number of liquid The developed to other of separation to such as and There are associated with such of a proteomics and can be with the of that the of such is and be The is with the However, most of the formats tools have emerged. A other formats are not to but to encode at is the Genomics A.R. M. Aebersold R. R. A. N. H. M. H. Paton Pizarro A. The Genomics an for standards in Biotechnol. PubMed Scopus Google Scholar) which to for information common to most as well as a for to a is in The other common of file is for encoding information for the mass spectrometer. such formats are referred to as and information the mass data in a can and for reaction for Each has its own multiple formats, for its but these are to the software, are not generally to and are not further However, among the mass input formats, an open for encoding has been are the to peptide ions an experiment. is a informatics challenge to and is of to share have been and used in an experiment E.W. Lam H. Aebersold R. a for for proteomics 2008; PubMed Scopus Google Scholar). The has the Markup E.W. M. S. Binz P.A. L. D. L. Aebersold R. for exchange of reaction Scopus Google Scholar). can encode and complex of the associated with each has been to be not to but to other that to can encode a of are to is not by instrument software, most vendors have a to support open-source such as S. reaction software for Proteome PubMed Scopus Google Scholar), M. D. L. H. Deutsch E.W. Aebersold R. a software for and for reaction mass 2011; PubMed Scopus Google Scholar), Deutsch L. an open for the for sharing Proteome 2011; PubMed Scopus Google Scholar), M. A. A. R. N. A. open-source software for mass 2008; PubMed Scopus Google Scholar), E.W. Lam H. Aebersold R. a for for proteomics 2008; PubMed Scopus Google Scholar), and G. The software an for and analysis of proteomics Proteome PubMed Scopus Google Scholar) support The software B. D.M. N. M. B. R. an open for and analyzing proteomics PubMed Scopus Google Scholar) has a application for the and analysis of and its for the not an official has a common to share with other The most diverse group of file types is for encoding the mass spectra that are the result of MS group of formats may be broadly separated into three proprietary formats, complex open formats, and formats. The mass spectra can be in two major as spectra and as spectra are as data such that ion features are at a the instrument that each has a that the may as a of as the original may be in the A of this is in which all data the are simply in of and the required to encode an has been to a spectrum to the a spectrum as the of and of the is but after each may of these types of as by the of these types is in Each has developed formats of its own and continually as new features are required by The formats in three single per and several per instruments, all output is in with the These may spectra spectra as by the can be with a as and as the of are as with the These all information in a and be with a file a that the complexity is in that multiple may be to the with each a by the instrument Data from and are in multiple with a with the These are typically as a and these are from the vendors these These can typically be read by software from the instrument vendors that is not the software is not the from and is to the the software is typically to is of and the and most to the data As a result of most of the vendors available software application programming in the of has the that software can be to read these formats software by the However, these can be used with software the has been in to under the software are of often with extensive that these formats are and to read and to software that data from any use as is a that proprietary data as new software is to read formats, formats being and of software to under a software data Biotechnol. 22: PubMed Scopus Google L. H. M. data mass spectrometry data in proteomics data 5: PubMed Scopus Google Scholar). this in not an to open formats that encode most of the information from each in a that then be by any with The such R. M. Deutsch E.W. B. B. R. H. S. G. Paton R. R. Aebersold R. A common open of mass spectrometry data and its application to proteomics Biotechnol. 22: PubMed Scopus Google Scholar), produced at the for Systems a the developed the The two formats generally to encode the but the data to be a data for mass 2008; PubMed Scopus Google for a discussion The of two formats widely as an that for software developers and among the and to a new that the features from and and The new L. M. M. D. A. S. Pizarro L. N. M. H. Binz P.A. Deutsch E.W. community for mass spectrometry 2011; PubMed Scopus Google Scholar), has been available in its but has been to and are However, as and features by from the A common these formats is that the in file and is with formats M. J.R. a new for sharing mass spectrometry data.Nat. Methods. PubMed Scopus Google Scholar). is to be generally by the in of development and of and The M. M. H. and of mass spectrometry data 2011; Scholar) file by into a and all the of the In a mechanism to all formats from to be the of in the of the of common to the mass output into the a that is common of the such formats the each spectrum to a file for the and the mass of the peptide from the and the then by all the that the formats are the and which in of the and and support the of being to spectra into A to all ion in file and all ion in a the and formats, J.R. D. Yates 3rd, J.R. and and file formats for the of spectra and Mass PubMed Scopus Google Scholar). the most common is the Generic file is to the in that multiple spectra in a single file via separated by in the of the can a including for The file developed by the of the most widely used commercial but is widely by proteomics These formats created with the of and MS proteomics to enable the and of spectra to software, a for which most are not However, have the development of proteomics tools are the to these formats. The to these and the data from the to support to the development of the complex open formats such as and as data are typically (3Deutsch E.W. Lam H. Aebersold R. Data analysis and bioinformatics tools for tandem mass spectrometry in proteomics.Physiol. Genomics. 2008; 33: 18-25Crossref PubMed Scopus (121) Google Scholar) a that to spectra a of a that to spectra a of a de that to the peptide the between Some software can these in a is common in the input formats to these nearly a output including and formats. the output formats are with the software used to the the file and has to results and formats. to use its own as The developed as a to the file J.R. D. Yates 3rd, J.R. and and file formats for the of spectra and Mass PubMed Scopus Google Scholar). R. proteins with tandem mass PubMed Scopus Google Scholar) a output L. M. D.M. Yang Open mass spectrometry Proteome PubMed Scopus Google Scholar) uses its own can an a other their own output and are not all the formats to their to a single that support multiple and to the results of multiple information of the results these the of the A. Aebersold R. to the of peptide identifications by and database Chem. PubMed Scopus Google Scholar) and A. Aebersold R. A for proteins by tandem mass Chem. PubMed Scopus Google Scholar) tools as created the and formats A. N. Aebersold R. A proteomics analysis open file PubMed Scopus Google Scholar). The is to nearly all output information from a and support of the associated with the and of as to enable the of data from multiple instrument vendors a common of software the use for to support the analysis of a common of the A. N. Aebersold R. A proteomics analysis open file PubMed Scopus Google Scholar). developed to the output of all into of the by and other tools such as of being and into an The approved as a but to be a de facto that by a variety of tools of its and lack of the is a a A. N. Aebersold R. A proteomics analysis open file PubMed Scopus Google Scholar), and its is a as from the by the Each be a of a complex group of proteins that The proteins of being identified in the and by information the peptide and peptide ions that the identification of each The approved as a but as a common among tools the As with the results that can be in and results to the peptide and can be in the and formats widely their to as a among tools. The and a number of use and the to and a that support all of the features of and and use The but A.R. M. G. S. B. S. R. Binz P.A. Deutsch E.W. H. M. Pizarro A. Creasy D. The data for mass proteomics PubMed Scopus Google Scholar). of the use but not the to encode of the to a in a with the and the of the complex information to a is under development by the Proteomics under the and A. data for in for be to encode information a variety of and at the of a complex most of these formats, as well as formats, are complex and can be to has for the output results of informatics to be into formats for analysis via tools with These output not all the information from their but most of the information for further analysis is In an to this of a common the is the is a that can the most information peptide and identifications and and can be read into analysis is not to the a common for a of the The is review the process L. H. Paton The formal process and its the PubMed Scopus Google Scholar). MS proteomics workflow to a of features an MS The are typically among multiple and the peptide are and that among the in a that into the of the are often for workflow begins with the as for encoding the output of the mass including proprietary formats and These mass are then by any number of software such as H. Aebersold R. A software for the and of peptide from of data by liquid PubMed Scopus Google Scholar), M. M. M. M. D. R. D. A. M. A of for the analysis of complex mixtures 22: PubMed Scopus Google Scholar), A. S. B. Aebersold R. M. for PubMed Scopus Google Scholar), B. D. A. H. S. N. Aebersold R. and tools for and mass 2008; PubMed Scopus Google Scholar), M. peptide identification mass and Biotechnol. 2008; PubMed Scopus Google Scholar), a for 5: PubMed Scopus Google Scholar), and The output of most of these is of but several support an open file developed at in with others Peptide Markup B. D. A. H. S. N. Aebersold R. and tools for and mass 2008; PubMed Scopus Google Scholar). can encode information all of the features and their including information any number of as well as the results of The in are being into the development of and but at the complex open for In a the mass is to peptide ion to the most The the and input are in the However, these are for the results of analysis of the of formats is The B. D.M. N. M. B. R. an open for and analyzing proteomics PubMed Scopus Google Scholar) uses its own open file to encode such as L. R. M. Aebersold R. data and for Methods. 2011; PubMed Scopus Google Scholar) and H. of and in peptide by multiple reaction mass Chem. PubMed Scopus Google Scholar) use their own formats. analysis such as use proprietary formats. The is the and formats with the of their results in to the results of other of proteomics A new data that has emerged in the is that can a of identified In a the identification results from a analysis are into spectra and spectra with their and other related the spectra used to the These may then be used to new data with a such as H. Deutsch E.W. N. Aebersold R. and of a searching for peptide identification from PubMed Scopus Google Scholar), R. D. peptide mass spectrum for Proteome 5: PubMed Scopus Google Scholar), of peptide spectra from proteomics spectrum Chem. PubMed Scopus Google Scholar) in of in to a with a The for Standards and the largest of of peptide ion and their of in their own The of the can and data with in its an a that is to read and and which is nearly the as the and can and read their own formats, and is to any of these formats to There is to a of data are with such as These tools a of which the input most this is a FASTA which of a of of a single by of The of the is and among Some such as and S. H. A. M. P.A. identification of peptides from tandem mass Chem. PubMed Scopus Google Scholar), the FASTA file into a to searching to Some the FASTA file to to In order to the of formats, the has a new the Extended FASTA the FASTA with the of a to for but a in the in which information the including can be and A related file formats are of the proteomics data is the database L. H. Jones M. D. R. the proteomics identifications 5: PubMed Scopus Google Scholar). to are in the from the is that be by the and most common of is with the H. L. proteomics Biotechnol. PubMed Scopus Google Scholar), common from proteomics analysis the is a to the an experiment from proteomics data to related formats are to MS proteomics but are related such that are used in with MS formats have to their The has developed a for the exchange of the H. L. G. Wojcik L. A. S. S. B. S. H. M. R. D. M. H. A. A. B. M. D. G. D. B. R. The community for the of data.Nat. Biotechnol. 22: PubMed Scopus Google Scholar). is used to exchange information among the after of a by of the by the original are in are from MS proteomics However, of the complexity of the and the of to have a of the has developed the S. Jones A. H. a the of the PubMed Scopus Google Scholar), which is a for encoding a of with each of the information that can be in the be in but often a of the information The is not to but to as a for in which the complexity of is and a be used of be but of the most common and such as G. B. A. A. B. M. software for mass Biotechnol. PubMed Scopus Google Scholar), process MS data in order to generate the development of a for M. S. D. G. M. M. S. D. M. Pizarro A. R. Deutsch M. A. D. A. and of Markup PubMed Google Scholar), an to a to a of that as the for any to encode the results from the analysis from any of experiment. The A.R. M. Aebersold R. R. A. N. H. M. H. Paton Pizarro A. The Genomics an for standards in Biotechnol. PubMed Scopus Google Scholar), has not been widely but several and developed for have been in other formats. is a related information that to support data for MS data is but not The formats presented here, for the mass data formats, are used the proteomics the open formats that be to other are to be The can be used by any field mass the community is to of the The use of the FASTA of as However, to be other as support for by the such as R. A. B. S. H. R. M. N. the PubMed Google Scholar) this to open formats is to an that can be used to any of the formats without the for an software to be that can well any of the formats. has been by the D. M. R. D. open software for proteomics tools 2008; PubMed Scopus Google Scholar) for most of the mass and by M. J.R. a new for sharing mass spectrometry data.Nat. Methods. PubMed Scopus Google Scholar), which an for and formats. The is from the to the data in two formats, which and workflow The is that such the software to the to the and the can an for which the software are which is of is to the software are all in and with these from other is whereas for the open formats in a variety of is the data are often in to analysis, and the results to be to In such be by as read the spectra from the original but is that most be to and are for which support is not such can well for However, to for most A such as and L. B. M. D. B. de database for and peptide Scopus Google Scholar), can read of these open formats use the for encoding the has been is for encoding has a large variety of of and and and is read by which a file be read of that is the of the whereas an file be not a software can the in the file the and be to the a to the the of the file to the of the is a of all formats. such as but to be to of the by the is the of definition can use for its encoding of information In its a of typically in such that these are either in common the In the extensive use of in formats is a to the of to In the development of at the as the for the new However, the of to generally with and this the major that a be that is to the encoding of to the encoding of that of the formats are and of the use of for formats to the to the formats are all and these formats have been widely in most common software be that be and all the for for mass output for encoding the results of database searching and and for encoding the results of the The use of such formats that the data and results can be by of software to There are a number of file formats used in MS These from formats to used of the data to formats. The variety of formats is of the of the formats are developed as new and are is a of in the community to under the of the to complex formats that can encode that can in the community and the of and enable a common of software tools to with data from a variety of These open standards the of proteomics in a that the formats and formats