A High-Confidence Human Plasma Proteome Reference Set with Estimated Concentrations in PeptideAtlas

Human blood plasma can be obtained relatively noninvasively and contains proteins from most, if not all, tissues of the body. Therefore, an extensive, quantitative catalog of plasma proteins is an important starting point for the discovery of disease biomarkers. In 2005, we showed that different proteomics measurements using different sample preparation and analysis techniques identify significantly different sets of proteins, and that a comprehensive plasma proteome can be compiled only by combining data from many different experiments. Applying advanced computational methods developed for the analysis and integration of very large and diverse data sets generated by tandem MS measurements of tryptic peptides, we have now compiled a high-confidence human plasma proteome reference set with well over twice the identified proteins of previous high-confidence sets. It includes a hierarchy of protein identifications at different levels of redundancy following a clearly defined scheme, which we propose as a standard that can be applied to any proteomics data set to facilitate cross-proteome analyses. Further, to aid in development of blood-based diagnostics using techniques such as selected reaction monitoring, we provide a rough estimate of protein concentrations using spectral counting. We identified 20,433 distinct peptides, from which we inferred a highly nonredundant set of 1929 protein sequences at a false discovery rate of 1%. We have made this resource available via PeptideAtlas, a large, multiorganism, publicly accessible compendium of peptides identified in tandem MS experiments conducted by laboratories around the world. Human blood plasma can be obtained relatively noninvasively and contains proteins from most, if not all, tissues of the body. Therefore, an extensive, quantitative catalog of plasma proteins is an important starting point for the discovery of disease biomarkers. In 2005, we showed that different proteomics measurements using different sample preparation and analysis techniques identify significantly different sets of proteins, and that a comprehensive plasma proteome can be compiled only by combining data from many different experiments. Applying advanced computational methods developed for the analysis and integration of very large and diverse data sets generated by tandem MS measurements of tryptic peptides, we have now compiled a high-confidence human plasma proteome reference set with well over twice the identified proteins of previous high-confidence sets. It includes a hierarchy of protein identifications at different levels of redundancy following a clearly defined scheme, which we propose as a standard that can be applied to any proteomics data set to facilitate cross-proteome analyses. Further, to aid in development of blood-based diagnostics using techniques such as selected reaction monitoring, we provide a rough estimate of protein concentrations using spectral counting. We identified 20,433 distinct peptides, from which we inferred a highly nonredundant set of 1929 protein sequences at a false discovery rate of 1%. We have made this resource available via PeptideAtlas, a large, multiorganism, publicly accessible compendium of peptides identified in tandem MS experiments conducted by laboratories around the world. Blood plasma contains a combination of subproteomes derived from different tissues, and thus, it potentially provides a window into an individual's state of health. Therefore, a detailed analysis of the plasma proteome holds promise as a source of biomarkers that can be used for the diagnosis and staging of diseases, as well as for monitoring progression and response to therapy. For many years, before the era of proteomics, the classic multivolume reference, The Plasma Proteins by Frank Putnam (1975–1989) (1Putnam F.W. The Plasma Proteins. 2nd Ed. Academic Press, New York1975–1989Google Scholar), provided a foundation for studies of plasma proteins. In 2002, Anderson and Anderson (2Anderson N.L. Anderson N.G. The human plasma proteome: history, character, and diagnostic prospects.Mol. Cell Proteomics. 2002; 1: 845-867Abstract Full Text Full Text PDF PubMed Scopus (3528) Google Scholar) a of plasma proteins by a of and with an plasma proteome reference the of MS used discovery reaction techniques in a in plasma data sets that to be to a comprehensive human plasma proteome reference In 2002, the Human of Human Plasma and provided reference of and and plasma to laboratories tandem MS and protein which by a into a data set of proteins from the The an for proteomics PubMed Scopus Google Scholar) identified peptides, for of the Plasma from the with laboratories and a data set of proteins and a PubMed Scopus Google the Human Plasma New Scopus Google of the for proteins in high-confidence protein identifications from data by a plasma proteome PubMed Scopus Google MS Human Plasma false discovery rate selected reaction in Anderson N.L. The human plasma proteome: a nonredundant developed by combination of Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar) a of nonredundant plasma proteins in the and in data sets Anderson N.L. The human proteome: of protein and of distinct PubMed Scopus Google a human blood proteome: analysis by with Cell Proteomics. 2002; 1: Full Text Full Text PDF PubMed Scopus Google of the human Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google in in the proteins by different false of and methods for protein for of the human plasma PubMed Scopus Google Scholar) to proteins from human the used and the applied for of the Plasma from the with laboratories and a data set of proteins and a PubMed Scopus Google Scholar) with and only to the data proteins in using a Proteomics. 1: Scopus Google Scholar), of which to and to the data previous the with the of proteins inferred from large data and with the proteins identified in different data sets. In 2005, we used a the proteomics analysis PubMed Scopus Google Scholar) to the Human Plasma N.L. Human Plasma PubMed Scopus Google Scholar), data sets and over a to estimate the of identifications made by and 2002; PubMed Scopus Google Scholar) of peptides identified at a false discovery rate of as by data to distinct proteins. of protein with from studies showed the Human Plasma PeptideAtlas, as well as the we that different proteomics experiments using different sample and analysis techniques identify significantly different sets of proteins. We that a comprehensive plasma proteome be compiled only by combining data from many and to as such data as The Human Plasma data identified distinct the in the proteins. In human blood plasma protein reference 1: PubMed Google Scholar) a high-confidence set of human plasma proteins a sample MS and with highly nonredundant set of proteins contains any previous plasma proteome reference The of the to a human plasma proteome reference set of by a of the Human Plasma data in and the data using We data sets to and using a compiled sets of protein identifications at different clearly defined levels of redundancy using a spectral provided a rough estimate of concentrations for a highly nonredundant set of protein sequences to blood-based diagnostic such as using reference peptides for selected reaction monitoring experiments The is a plasma proteome reference set and 1929 highly nonredundant protein sequences at an of which sets of protein identifications at levels of redundancy the for the Human Plasma at contains any protein in the protein that includes at identified set with a of the set which protein sequences the set of identified set with protein sequences for which the identified peptides a of the identified peptides for protein a of the not set which protein includes of the peptides of any of the sequences that not not a set that is not to that of a protein a set of protein sequences that can of the identified of protein sequences in a protein in of protein peptides in and of is with the of the protein to which it is from it contains the set of to in the is protein is is by peptides a of it is by and sequences and the set for is different from that of The protein sequences with the of sequences to the peptides in the and to the protein concentrations using and analysis and analysis of the Plasma reference sample and of PubMed Scopus Google Scholar) spectral from the Human Plasma PeptideAtlas, a a protein in proteins that the The and the to the standard of the at of The the proteins with concentrations in and analysis of the Plasma reference sample and of PubMed Scopus Google Scholar) The at the an estimate of the of the Human Plasma as a of as the of by the of and for identified by of the as in of protein sequences identified of for a of distinct We from data sets of the Plasma from the with laboratories and a data set of proteins and a PubMed Scopus Google Proteomics. 1: Scopus Google Anderson Anderson proteome analysis of human plasma following in using and the and Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Anderson of PubMed Scopus Google Human plasma analysis by and PubMed Scopus Google of the plasma Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google in and in Google Scholar) and from from the data and from from plasma and a of sample preparation techniques of of Proteins and Proteomics. Full Text Full Text PDF PubMed Scopus Google and analysis a of with data set of and for a of includes large experiments For we the data sets into and and the The data sets for the selected from experiments we to an spectral to of data and to proteomics PubMed Scopus Google Scholar) and with and of a spectral for from PubMed Scopus Google Scholar) a spectral of the human spectral Scholar) spectral for false discovery rate in spectral in PubMed Scopus Google Scholar) for contains derived from identified of which of for The for using the proteomics analysis PubMed Scopus Google Scholar), as in for to estimate the of identifications made by and 2002; PubMed Scopus Google Scholar) a for for peptides of of identifications in Cell Scholar) applied to the to by of the data by and The of identifications the of not of the that with identifications to be of experiments the that in experiments and to the to be of peptides identified by with different and of peptides identified with different to a protein derived from Google The protein and in PubMed Scopus Google Scholar), PubMed Scopus Google Scholar), and of In many the protein is in the it is in the includes many to protein sequences the in many this in a to protein sequences that For very large data the at the to be that at the at the protein false discovery for very large proteomics data sets generated by tandem Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google in to a protein for the Human Plasma PeptideAtlas, a of to from to applied to for proteins by tandem PubMed Scopus Google Scholar) to distinct the of and using a of peptides which peptides that to proteins with many identified The set of identified peptides for the Proteomics. 1: Scopus Google Scholar) to many peptides that to not we the set for this of peptides that in the this combining the for to to protein identifications and to protein identifications with sets. The for experiments and distinct peptides and protein identifications compiled to a Human Plasma It is to a of identified proteins such a is by and for the of a Further, the set of identified peptides to the of a protein is a the set of peptides to the of it is that proteins have is to this from the this we compiled sets of protein identifications for this at different levels of For the of this the redundancy of a set of protein identifications is the to which the set contains sequences to of the that redundancy is different from and that the different redundancy levels not to different The protein sets the and used this sets in and a detailed is in We have this to the in The set includes any from the protein to which any identified highly set includes of a protein to a in the Human Plasma PeptideAtlas, that is in the the is in in the set that the protein includes a in the The set is a highly nonredundant set of protein sequences of the identified peptides and it as a proteome reference It includes the protein from protein the protein sequences for of comprehensive and and a of is now to for human protein The PubMed Scopus Google Scholar), with a of in the of which have at the protein a protein includes protein sequences for which the set with the we sequences as well for and for The of the set is a estimate of the of distinct proteins It is important to that the is with to a data a protein that is identified in be in and in The set includes protein sequences that have peptides it from protein sequences in the with peptides of the of identified peptides in the for the set is a set to of the identifications for set of of the protein sequences of the protein and is the It is for a protein to identified as is for using false discovery for very large proteomics data sets generated by tandem Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar) the in for to using the to protein sets. We the data sets from using in to proteins, many of which preparation as in and of using and PubMed Scopus Google to a using proteins with the with and peptides via in a with the this that in the protein now as via We not the spectral it not data with proteins with tandem PubMed Scopus Google Scholar) using a the proteomics analysis PubMed Scopus Google Scholar) a of generated by a of tryptic in in The for from to and of standard of a protein for spectral of PubMed Scopus Google Scholar) that be we the for in in the in this to be with the of to the of to as the of to in the for It is important to this computational of peptides the it not the in the We a Human Plasma using the methods as with a of a of We this to of a protein combining with the It not to for the in the for a in in the protein for the applied to estimate the of the for protein in the that the for a with in the sample for and of protein in PubMed Scopus Google We applied a of the by and protein the of and PubMed Scopus Google For protein identified in the Human Plasma PeptideAtlas, we with a of that to that protein the to the of the is by it to the of available tryptic we a by the of tryptic peptides of from an in by which is very the of tryptic peptides protein the and the by by that We the to the concentrations of proteins. In we concentrations via and in and analysis of the Plasma reference sample and of PubMed Scopus Google In many concentrations proteins. the and from this we an for protein with have for this N.G. of methods for human proteins by Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google in the Human Plasma to for in using from The of the standard from the at the of the and and as an for the in that from at concentrations to at for estimate concentrations in the Plasma PeptideAtlas, we the to for the of the distinct peptides in this the a in the we into the of tryptic peptides in protein and the of peptides the The is in is a the concentrations in the and for protein sequences in we the and peptides from to a Human Plasma that includes from plasma and We the set of experiments and protein sets as concentrations from the used for protein sequences with in a for false discovery of protein identifications in data sets false discovery for very large proteomics data sets generated by tandem Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar), applied to and to the to estimate the a of the by into the of the data the of tryptic peptides in and proteome of the protein identifications that combining the and by only a We a for of the identifications to for of the in the at of highly of the and highly to to and to a tandem to the and PubMed Scopus Google Scholar), tryptic the for at of the to for and only for peptides at of We that to to the in a of the can not from the and We the to a with a protein of as We applied a of to the data and a of to the in a with a protein of to a Plasma protein is identifications not and that computational in this For peptides in we an a resource for for proteomics PubMed Scopus Google Scholar), defined as the of the of in which a is by the of in which the protein is For if is in different and protein is in the is The Human Plasma PeptideAtlas, from data contains 1929 protein sequences with an protein of and the set of protein sequences is a highly nonredundant protein set with protein of peptides with any of the protein The of 1929 protein with concentrations and of and distinct peptides, is in protein in the Human Plasma is by and distinct peptides and and the 1929 protein by and by distinct The previous Human Plasma peptides to nonredundant proteins redundancy to that of the for the The Human Plasma contains identified peptides and protein For of we not estimate the protein of the it using a very protein is the of the The for the and the to estimate by the of data derived from in and false identifications and of a spectral for from PubMed Scopus Google Scholar), a of identifications at any of the of identifications in Cell Scholar), and false identifications in by of the data by For the we used a of experiments in the of this a of false for experiments for experiments. Therefore, we used a to a protein of for the to of for the many in we to the developed to estimate the protein protein identifications computational a protein of analysis protein of proteins and with and false discovery using PubMed Scopus Google Scholar) and protein identifications in of we and that we a of false identifications that the as a we state that the protein is a protein protein identifications have protein identifications of for plasma protein is the disease and at of sample concentrations of relatively proteins relatively not an of Plasma and for Blood Scholar), and it is to have a rough estimate of protein for such as the in of reference peptides for MS as a for for and of protein in PubMed Scopus Google N.G. of methods for human proteins by Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar) and protein the of and PubMed Scopus Google protein for of protein in proteomics by the of peptides Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar) of proteins of spectral used for plasma plasma protein with tandem and PubMed Scopus Google following a of the of and protein the of and PubMed Scopus Google Scholar), we by spectral to for the of tryptic peptides protein and by to protein The concentrations rough and not be as quantitative to of proteins not used for spectral in the of plasma proteins Anderson N.L. of the human plasma proteome the of PubMed Scopus Google the concentrations for of proteins a of of the of the by and a of measurements as of the of and Further, a in a sample not a plasma protein of the the that the very large of data to the to by over many diverse it of the of and for can be to the spectral of proteins to the for proteins of spectral quantitative PubMed Scopus Google the of in from experiments with and concentrations to the for is for the for protein as is at be to be protein sequences in the data from analysis of concentrations for proteins proteins that the of an of human and 1: PubMed Scopus Google Plasma concentrations for proteins can be is of blood sample and The of the concentrations for and is to the in in concentrations in to PubMed Scopus Google Scholar), that such The concentrations spectral of the protein sequences in the Human Plasma of from for to for is to be the protein in plasma with a of (2Anderson N.L. Anderson N.G. The human plasma proteome: history, character, and diagnostic prospects.Mol. Cell Proteomics. 2002; 1: 845-867Abstract Full Text Full Text PDF PubMed Scopus (3528) Google of and Ed. Scholar), is in the of proteins of such as and of the protein sequences in the Human Plasma the and we that computational not for any protein and the not a for not for The peptides, to the used of in the of the Human Plasma of identifications in a protein sequences from the Human Plasma PeptideAtlas, with concentrations not in the Human Plasma set of the peptides to proteins an it is highly that peptides be identified the which in of only in the spectral we used to the the proteins, have in the and not have identified by spectral as in of spectral we very proteins be identified we to the Therefore, we that for of proteins, the from the is of in and set to to peptides with peptides that not for and by the of data of and protein tryptic at the of a can be by for Plasma and protein peptides of the not proteins only in in a to catalog the proteins in the Human Plasma includes experiments from with disease state a of of the plasma Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar), and we that of the protein sequences in the only in of experiments and not in any of the the of proteins The with at we that of proteins in proteins, of the advanced in the experiments of the plasma into and peptides, and of into using analysis a which twice the identifications with experiments from the of only plasma proteins in a advanced Anderson proteome analysis using a quantitative PubMed Scopus Google in sample and the a very large of plasma proteins of to of protein sequences from a plasma proteome reference We the of protein sequences that to by identified as in in the protein as such a We that and not We not sequences as and protein sequences not to The human of is to with for is only for which in the and it is only this which we is in in human as we that as distinct protein sequences have only a of from Further, protein sequences from the that includes at that is not to any in in for an not in set of 1929 protein by the at this includes the proteins as well as of the in of the human plasma using and Scholar) in we it is from a catalog of the human plasma reference set and the in proteins that by MS proteins from to be in the of the in is not of proteome of protein in to in protein In protein In Human Plasma set in highly plasma proteins Anderson N.L. of the human plasma proteome the of PubMed Scopus Google plasma proteins by analysis of tryptic as disease Anderson N.L. reaction of proteins in human Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google derived plasma with identified by of the human plasma using and for the identified by computational derived plasma proteins human blood plasma protein reference 1: PubMed Google and Anderson biomarkers Anderson N.L. of for 1: Google for the identified by computational in a we not to of the analysis of the protein plasma that at identifications from the protein Human Plasma by the of proteins as identified that to the Human Plasma PeptideAtlas, that we not The data of the proteins. with the of experiments and and Human plasma analysis by and PubMed Scopus Google of the plasma Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google The the of proteins with the techniques is not that In a plasma proteome reference set human blood plasma protein reference 1: PubMed Google Scholar) of to of with for the protein in human blood plasma protein reference 1: PubMed Google Scholar), in protein not in the Human Plasma that we identified peptides for the data to the Human Plasma PeptideAtlas, of in the that from different laboratories using different sample and preparation to of protein sequences to the human plasma we the data a spectral and not a we only identified peptides that in experiments and in the spectral the is of the human data in the plasma and many with a with standard be identified with spectral if proteins be to the we to of for we set by identified peptides to a protein many sequences with only redundancy from such a set is a in proteomics and standard methods have In this it is to that protein for a data set can be the of redundancy that as in and data analysis in the of the Plasma PubMed Scopus Google Scholar), and is at with the of identifications with the In a highly nonredundant is a of which proteins that as as to in the In we that we not to have for any of the in we for protein in the that protein a in at of the For such as the of distinct proteins by the a highly nonredundant protein set is For such as with a nonredundant for by of peptides for redundancy is we different protein sets that be in for different not and have in many proteomics For the data set for protein used to sets a set of protein sequences using very with an for in high-confidence protein identifications from data by a plasma proteome PubMed Scopus Google Scholar), to and an set of protein sequences only a to the set defined the we in this we a protein the available and to any that can be to a data for PubMed Scopus Google Scholar) of data and to proteomics PubMed Scopus Google identifications generated for different data sets using can be and is not available to for the of identifications clearly defined and and we propose as a standard for the the Human we that the the protein identifications for proteomics data it is to to the and to the highly nonredundant set for the set for the the be For that of identifications the high-confidence We identifications the set for an we from of the data and an of which is to provide rough for protein concentrations in the Human Plasma to concentrations as many as we the by methods in and analysis of the Plasma reference sample and of PubMed Scopus Google of spectral for concentrations in with now available in rough and not be as quantitative concentrations from and analysis of the Plasma reference sample and of PubMed Scopus Google Scholar), Anderson N.L. of for 1: Google Scholar), and provided as We to this spectral to for subproteomes such as human and data sets that we is to a quantitative protein in and in and and Anderson in Anderson N.L. of for 1: Google Scholar) a of biomarkers proteins to be in with only to be in in the Human Plasma set identified in protein for of protein in proteomics by the of peptides Cell Proteomics. Full Text Full Text PDF PubMed Scopus Google Scholar) as for development of the in a protein is in a sample that is with of the peptides many not at all, in the and with MS analysis of peptides for quantitative PubMed Scopus Google with Cell PubMed Scopus Google that to have protein the of and PubMed Scopus Google of peptides for quantitative PubMed Scopus Google with Cell PubMed Scopus Google by the data with which for peptides in the Human Plasma PeptideAtlas, we an that not it is highly MS data as in of spectral quantitative PubMed Scopus Google experiments of many proteins, proteins, a in which the only peptides proteins of be and the can peptides the and such as of of protein and as an for the includes to For peptides to proteins not in PeptideAtlas, for the to a that can be by the can be generated from to For protein the protein concentrations to in reference peptides at concentrations to in the and in a resource for for proteomics PubMed Scopus Google we and in the of reference from using the used for measurements and we publicly accessible as sets of selected for proteins and PubMed Scopus Google reference for the analysis of the is an of the for and data in of the of the The PubMed Scopus Google with the for the Scholar) and the to the proteomics data PubMed Scopus Google Scholar) contains the data the data of studies into proteome reference sets. of is the of a standard for the of data to the It is the of that plasma data be to from which it be in and into the