An integrated view of baseline protein expression in human tissues

Abstract

The availability of proteomics datasets in the public domain, and in the PRIDE database in particular, has increased dramatically in recent years. This unprecedented large-scale availability of data provides an opportunity for combined analyses of datasets to get organism-wide protein expression data in a consistent manner. We have reanalysed 25 public proteomics datasets from healthy human individuals, to assess baseline protein abundance in 32 organs. We defined tissue as a distinct functional or structural region within an organ. Overall, the aggregated dataset contains 68 healthy tissues, corresponding to 3,167 mass spectrometry runs covering 501 samples, coming from 492 individuals. We compared protein expression between the different organs, studied the distribution of proteins across organs, and identified proteins, as well as their isoforms, that are uniquely expressed in certain organs. We also performed gene ontology and pathway enrichment analyses to identify organ-specific enriched biological processes and pathways. As a key point, we have integrated the protein expression results into the resource Expression Atlas, where it can be accessed and visualised either individually or together with gene expression data coming from transcriptomics datasets.

Integrated view and comparative analysis of baseline protein expression in mouse and rat tissues

“…The disadvantage of this approach is that the False Discovery Rate (FDR) statistical thresholds are applied at a dataset level and not to all datasets together as a whole. However, as reported before [15], using a dataset per dataset analysis approach is in our view the only sustainable manner to reanalyse and integrate quantitative proteomics datasets, at least at present. As also reported in previous studies, one of the major bottlenecks was the curation of dataset metadata, consisting of mapping files to samples and biological conditions.…”

“…One of the main aims of this study was to study protein expression across various organs. To enable simplicity of comparison [15] we first grouped 34 different tissues into 14 distinct organs, as discussed in ‘Methods’. We defined ‘tissue’ as a distinct functional or structural region within an ‘organ’.…”

“…The datasets were run separately in multi-threaded mode. The post-processing of results from MaxQuant is explained in detail in [15]. In brief, the protein groups labelled as potential contaminants, decoys and those with less than 2 PSMs were removed.…”

“…Enriched gene ontology (GO) terms analysis was carried out through over-representation test described previously [15], it was combined with "Organ-enriched" and "Group enriched" mapped gene lists for each organ. In addition, Reactome [54] pathway analysis was performed using mapped gene lists and running pathway-topology and over-representation analysis, as reported previously [15].…”

“…As a result, proteomics expression data can be integrated with transcriptomics information, mostly coming from RNA-Seq experiments. So far, we have performed two main meta-analysis studies involving the reanalysis and integration of: (i) 11 public quantitative datasets coming from cell lines and human tumour samples [14]; and (ii) 25 human baseline datasets coming from 32 different organs [15].…”