Methods

PeptideAtlas collects raw mass spectrometry proteomics datasets from laboratories around the world and reprocesses them in a uniform bioinformatics workflow using the Trans-Proteomic Pipeline. Each PeptideAtlas build comprises a set of reprocessed experiments from a single species or subset of samples (such has human plasma) from a species. Processed results are filtered to a quality level such that there is a 1% false discovery rate at the protein level. All peptide identifications of sufficient quality to enter a build are mapped to the Ensembl genome (v67.37) using the Ensembl toolkit. Genomic coordinates for all identified peptides to all their Ensembl protein, transcript, and gene mappings, including intron spans, as calculated by the Ensembl toolkit are stored in the PeptideAtlas database.