Dataset Information

A mass-tolerant database search identifies a large proportion of unassigned spectra in shotgun proteomics as modified peptides

ABSTRACT: Less than half of all MS/MS spectra acquired in shotgun proteomics typically result in a confident peptide match. Here we present an ultra-tolerant Sequest database search that allowed peptide matching even with modifications of unknown masses up to Â±500 Da. From an HEK293 cell proteome-wide dataset (9,513 proteins and 396,736 peptides), a Â±500-Da search matched an additional 184,000 modified peptides. These were linked to both biological and chemical modifications representing 523 distinct ï??mass bins including phosphorylation, glycosylation, and methylation. We attempted to localize all unknown modification masses to specific regions within a peptide, and known modifications were accurately assigned to the correct amino acids with frequencies often >90%. These data demonstrate that a large fraction of previously unassignable spectra are assignable to peptide sequences with modifications.

INSTRUMENT(S): Q Exactive

ORGANISM(S): Homo Sapiens (ncbitaxon:9606)

SUBMITTER: Dr Steven P Gygi

PROVIDER: MSV000079841 | MassIVE | Sun Jun 19 19:14:00 BST 2016

SECONDARY ACCESSION(S): PXD001468

REPOSITORIES: MassIVE

ACCESS DATA

Dataset's files

Source:

			Action	DRS
		Other

Items per page:

1 - 1 of 1

Publications

A mass-tolerant database search identifies a large proportion of unassigned spectra in shotgun proteomics as modified peptides.

Chick Joel M JM Kolippakkam Deepak D Nusinow David P DP Zhai Bo B Rad Ramin R Huttlin Edward L EL Gygi Steven P SP

Nature biotechnology 20150615 7

Fewer than half of all tandem mass spectrometry (MS/MS) spectra acquired in shotgun proteomics experiments are typically matched to a peptide with high confidence. Here we determine the identity of unassigned peptides using an ultra-tolerant Sequest database search that allows peptide matching even with modifications of unknown masses up to ± 500 Da. In a proteome-wide data set on HEK293 cells (9,513 proteins and 396,736 peptides), this approach matched an additional 184,000 modified peptides, w ...[more]

PMID: 26076430

Similar Datasets

Project description:Protein identification was carried out by searching against a joint Human + Yeast Swissprot database (Uniprot release 57.3 May 2009, 26885 entries), in order to increase statistical power of peptide identification, which was supplemented with porcine trypsin, and using SEQUEST algorithm (Bioworks 3.2 package, Thermo Finnigan). For SILAC, we used variable (methionine oxidation, lysine and arginine modification of +6 Da) and fixed modifications (cysteine carboxamidomethylation). Precursor mass tolerance was set to 2Da, fragment mass tolerance to 1.2Da and up to two missed cleavage sites for trypsin were allowed. The same collections of MS/MS spectra were also searched against inverted databases constructed from the same target databases. SEQUEST results were analyzed using the probability ratio method 22b, taking into account isoelectric points of peptides to improve peptide identification 9b. False discovery rates (FDR) of peptide identifications were calculated from the search results against the inverted databases using the refined method 9b. The combination of stable isotope labeling (SIL) with mass spectrometry (MS) allows comparison of the abundance of thousands of proteins in complex mixtures. However, interpretation of the large datasets generated by these techniques remains a challenge since appropriate statistical standards are lacking. Here we present the first generally applicable model that accurately explains the behavior of data obtained using current SIL approaches, including 18O, iTRAQ and SILAC labeling, and different MS instruments. The model decomposes the total technical variance into the spectral, peptide and protein variance components, and its general validity was demonstrated by confronting 48 experimental distributions against 18 different null hypotheses. In addition to its general applicability, the performance of the algorithm was similar or better than that of other existing methods. The model also provides a general framework to integrate quantitative information, allowing a comparative analysis of the results obtained from different SIL experiments. The model was applied to the global analysis of protein alterations induced by low H2O2 concentrations in yeast, demonstrating the increased statistical power that may be achieved by rigorous data integration. Our results highlight the importance of establishing an adequate and validated statistical framework for the analysis of high-throughput data. Except for TOF/TOF MS/MS files, protein identification was carried out by searching against a joint Human+YeastSwissprot database (Uniprot release 57.3 May 2009, 26885 entries), in order to increase statistical power of peptide identification, which was supplemented with porcine trypsin, and using SEQUEST algorithm (Bioworks 3.2 package, Thermo Finnigan). For 18O-labeled samples variable (methionine oxidation, lysine and arginine modification of +4 Da) and fixed modifications (cysteine carboxamidomethylation) were used. For iTRAQ, we allowed variable (methionine oxidation) and fixed modifications (cysteine carboxamidomethylation, lysine and N- terminal modification of +144.1020 Da). For SILAC, we used variable (methionine oxidation, lysine and arginine modification of +6 Da) and fixed modifications (cysteine carboxamidomethylation). Precursor mass tolerance was set to 2Da, fragment mass tolerance to 1.2Da and up to two missed cleavage sites for trypsin were allowed. The same collections of MS/MS spectra were also searched against inverted databases constructed from the same target databases. SEQUEST results were analyzed using the probability ratio method 22b, taking into account isoelectric points of peptides to improve peptide identification 9b. False discovery rates (FDR) of peptide identifications were calculated from the search results against the inverted databases using the refined method 9b. TOF/TOF MS/MS spectra were converted to MGF files by using the 4000 Series Explorer V.3.5.1 internal processor and the following parameters: mass range from precursor -20 Da to 60 Da, peak density of 50 peaks per 200 Da with S/N >1 and minimum area >1 with a maximum number of 100 peaks per precursor. These files were analyzed with the Mascot V2.2.06 search engine using the database and search conditions indicated above.

Project description:The continuing ‘desire’ in obtaining more quantitative and detailed information from cellular proteomics experiments in regards to proteome coverage and protein modifications asks for a systematic investigation of the protein extraction and digestion protocols, in particular when working with unicellular organisms with a strong cell wall such as the model yeast S. cerevisiae. The selection of a suitable sample preparation workflow is crucial to obtain in-depth proteome coverage, as the use of certain sample preparation protocols may bias the protein identification or induce (unexpected) peptide modifications. Here, we made an extensive comparison of preparation workflows commonly applied to S. cerevisiae. Workflows were compared on the basis of identified MS/MS spectra, peptide sequences and number and type of modifications using both restricted (Peaks) and unrestricted (TagGraph) database search approaches. The proteome coverage was mainly affected by the sample collection approach, while it was maximized using a FASP. Extensive reagent specific peptide modifications were detected when using formic acid, but also when using acetone. Such artefacts split the analyte mass signals and generate additional chemical noise that may also elute differently compared to the native peptide. The use of both an unrestricted and restricted database search increased identification rates significantly and resulted in the identification of approximately 70% of the MS2 spectra for the best protocol. The unidentified spectra were assessed by their de novo sequencing score. This confirmed that those spectra consisted by a majority of very low quality spectra, insufficient to match to database sequences. However, a small fraction of the unidentified spectra showed high quality, which presumably derive from unknown sequence variants not present in the database. This study demonstrates the high importance of the sample preparation workflow and the obtained results will guide researchers in the field to optimize sample preparation procedures.