Human Plasma Proteome Project

The HPPP was initiated in 2002 with a series of planning workshops. During 2003-2005, the HPPP prepared and distributed reference specimens of human serum and EDTA-, citrate-, and heparin-plasma to 55 participating laboratories worldwide. The HPPP stimulated access to emerging technologies and generated substantial datasets and integrated databases for proteins detectable and identifiable in human plasma and serum. Experimental protocols used combinations of depletion, fractionation, mass spectrometry, and immunoassay methods linked via search engines and annotation groups to gene and protein databases. We created a new human plasma proteome database. We concluded that EDTA-plasma is the preferred specimen for studies of plasma or serum from human blood.

A special issue of PROTEOMICS in August 2005 presented 28 articles from the HPPP—both collaborative articles and ancillary analyses stimulated by a special small grants program. Collaborating laboratories and working groups of this Pilot Phase of the HPPP addressed (a) specimen stability and protein concentrations; (b) protein identifications from 18 MS/MS datasets, including subproteome analyses; (c) independent analyses from raw MS/MS spectra; (d) search engine performance; (e) biological annotations and insights; (f) antibody arrays; and (g) direct MS/SELDI analyses. A Core Dataset of proteins identified based on two or more peptides had 3020 protein IDs characterized with Gene Ontology, InterPro, Novartis Atlas, Online Mendelian Inheritance in Man, and immunoassay-based concentration determinations. Using different criteria, several subsets of the Core Dataset were described. The work was also published as a book by Wiley “Exploring the Human Plasma Proteome” in 2006.

In Phase II, the HPPP has grown with the contribution of additional large datasets and standardized analysis using Trans-Proteomic Pipeline to create the current version of the Human Plasma PeptideAtlas. Cross-analyses with the urine, kidney, and liver proteome datasets from the HUPO KHUPP and HLPP initiatives were presented at the Sydney HUPO Congress in 2010 and published by Farrah et al.

In 2017, the HPPP reviewed the history of the projects and the advances made in exploring the human plasma proteome since the inception of the HPPP and reviewed several of its recent achievements. Applying the Human Proteome Project Data Interpretation Guidelines (each with ≥ 2 distinct peptides of ≥ 9 amino acids) to historical PeptideAtlas builds going back to 2006, the HPPP described the progress made in the past ten years in plasma proteome coverage. The latest 2017-04 build of Human Plasma PeptideAtlas contained ∼43 million peptide-spectrum matches and 122,730 distinct peptide sequences from 178 individual experiments, with a 1% protein-level FDR globally across all experiments. This yielded 3509 proteins including > 1300 additional proteins with ambiguous evidence. We further compared this set with proteins of historical PeptideAtlas builds, RNA abundances and cellular localization categories. Our recent work also considered targeted mass spectrometry and multiplexed affinity assays. During early 2017, affinity assays targeted ∼2000 proteins of which half were also described in the 2017-04 built. Lastly, the HPPP brings sample handling and study design into focus during the exploration of the plasma proteome, as these may influence the detectability of existing and novel proteins with the chosen technologies.