Computational protocol: Hair-bundle proteomes of avian and mammalian inner-ear utricles

Similar protocols

Protocol publication

[…] For ion-trap data, MS2 scan peak lists were created from instrument RAW files using the MSConvert program (v 3.0.6705) from the Proteowizard toolkit,. Compressed text files were created using the command string ‘>msconvert --filter ‘msLevel 2’ --text –gzip’. The text files were parsed with in-house software and MS2-format files were created. Scans had to have minimum peak counts of 15 or greater, and absolute intensities of 100 or greater. Scans having 90% or greater total intensity associated with m/z values less than the precursor m/z value were assigned 1+ charge states. Scans failing this test were duplicated as 2+ and 3+ charge states.Protein databases in FASTA format were downloaded from Ensembl ( for each species: mouse (release 76 with 52,998 sequences), chick (release 77 with 16,366 sequences), and rat (release 75 having 25,724 sequences). Database entries for known major hair-bundle proteins were checked and, in a few cases, replaced with correct and full length versions. Because Ensembl releases lack proper protein description strings, which are available from the Ensembl BioMart tool, the FASTA header lines were modified to include protein description strings. The protein databases used in the Comet searches included 179 common contaminant sequences and sequence-reversed decoy entries. Database preparation used utilities available at The respective FASTA databases are included in the ProteomeXchange datasets.Database searching used version 2014.02 rev. 2 of Comet. The searches were configured similarly for all datasets aside from the proper protein database path. Theoretical peptides were fully tryptic (cleavage at K, R except after P) with a minimum of 2 missed cleavages and within a mass range of 600–5,000 Da. Parent ion mass tolerances of 2.5 Da and average masses were specified. Static modifications of +57.0215 on cysteine residues were applied. Variable oxidations of methionine residues (M+15.9949) were also specified with a maximum of 3 modifications per peptide. Monoisotopic masses and mass tolerances of 1.0005 (0.4 fragment ion bin tolerance) were used for fragment ions. The ion series used in Comet scoring were b- and y-ions along with neutral loss peaks. [...] Chicken hair-bundle data acquired on an Orbitrap mass spectrometer, previously reported as ProteomeXchange dataset PXD000104 (Data Citation 1), was reanalyzed using the the chicken Ensembl database described above (PXD002445; Data Citation 2). Searches were carried out using the search engine Andromeda, built into MaxQuant version software. The default contaminants file associated with the MaxQuant download was edited to remove entries known to be present in hair bundles (e.g., actin) and to add additional impurities that entered the bundle-purification workflow (e.g., keratins, haemoglobins). Protein identifications were reported with an false discovery rate (FDR) of 5%. […]

Pipeline specifications

Software tools ProteoWizard, Andromeda, MaxQuant
Databases ProteomeXchange
Application MS-based untargeted proteomics
Organisms Mus musculus, Gallus gallus, Rattus norvegicus, Homo sapiens