Integration of next-generation sequencing and proteomics data: application to the discovery of new coding events in the human spermatozoa

Proteomics and transcriptomics data were integrated in order to identify hard-to-reach uncharacterized proteins which are known to be particularly numerous in human spermatozoa. MS-identifications were launched both against UniprotKB and translated predicted ORF databases generated from four different RNAseq processing workflows. For each validated peptide-spectrum match (PSM), not identified in UniprotKB, but at least in one of the four customized databases, spectrum was carefully checked manually to assess the novel identification reliability.