Respect your Results

From SPCTools

(Difference between revisions)
Jump to: navigation, search
Revision as of 20:25, 12 August 2014
Dshteynb (Talk | contribs)
(ReSpect moved to "reSpect")
← Previous diff
Current revision
Dshteynb (Talk | contribs)
(reSpect your Results)
Line 1: Line 1:
-= reSpect your Results =+= reSpect =
''reSpect'' is a tool that allows you to identify '''even more peptides''' from your existing spectra '''without collecting anymore data'''. It can help boost identification rates for low abundance ionic species in datasets containing chimeric spectra. ''reSpect'' is a tool that allows you to identify '''even more peptides''' from your existing spectra '''without collecting anymore data'''. It can help boost identification rates for low abundance ionic species in datasets containing chimeric spectra.

Current revision

reSpect

reSpect is a tool that allows you to identify even more peptides from your existing spectra without collecting anymore data. It can help boost identification rates for low abundance ionic species in datasets containing chimeric spectra.

Running reSpect:

  • Open the "Utilities"->"reSpect" tab under the Petunia Web Interface
  • Select a pepXML file with PeptideProphet and/or iProphet probabilities as input
  • Select an MS/MS mz-tolerance appropriate for the type of instrument, e.g. 0.4 for the Yeast Orbitrap Sample Files
  • When the program completes you will have new mzML or mzXML files along-side your original data, but containing an "_rs" suffix appended to the end of the original base-name of the file.
  • Make a copy of the original parameter file.
  • In the parameter file adjust the precursor mass setting to match the Selection Window, e.g. +- 3.1 Daltons
  • reSpect will remove the original precursor charge from its output. You may also want to set the assumed charge to [+1, +4 or +5] although many search engines will take care of this automatically.
  • Search the reSpect results using the adjusted parameter file.

Processing reSpect Search Results Using TPP

  • Change the "Write output to file:" to another name e.g. interact-rs.pep.xml
  • Change Filter out results below this PeptideProphet probability: to 0.0
  • Open the "Analysis Pipeline"->"Analyze Peptides" tab under the Petunia Web Interface
  • Select your reSpect pepXML files
  • Make sure "use accurate mass binning" is unchecked
  • You may identify more results if you enable "Use decoy hits to pin down the negative distribution" and "Use Non-parametric model", if you do this also enable "Use Non-parametric model (can only be used with decoy option)"
  • It may also help to enable "RUN InterProphet" (a.k.a. iProphet)
  • Make sure to check the FDR plots carefully to verify that the model is giving you a conservative estimate of probability for both PSMs as predicted by PeptideProphet and Distinct Peptides as predicted by iProphet
  • If you don't see a valid result from PeptideProphet you can always try the CLEVEL option [1]
Personal tools