PSIPI

1
PSI-PI

• Protein Informatics Workgroup
• Toledo Meeting 23rd April 2008

2
History...

• Nov. 2004 mzIdent functional by the year's
  end
• Small group, summer 2005, based on pepXML/protXML
  
• San Francisco, April 21-23, 2006 expects to
  finalise the analysisXML standard
• Washington, September 25, 2006 expects to
  finalise the analysisXML standard
• Lyon, April 23-25, 2007 The PSI-PI group will be
  working on finalization of the AnalysisXML
  schema, which is quickly reaching consensus
• Hinxton, Jun 25-27, 2007 . . .

3
Aims for the meeting

• Develop a strategy for completing AnalysisML
• Develop a strategy for completing AnalysisXML
• Develop a strategy for completing AnalysisML
• . . .

4
AnalysisML - Cunning plan

• Review deliverables
• Review use cases
• Review status
• Review 2 other possible approaches
• Decide on what approach to take
• Produce a time / resources plan
• Continue with the work.

5
AnalysisML - Deliverables

• Use cases
• XML schema
• CV
• Documentation, fully describing the format
• Examples for ms, ms-ms, sequence tag and Denovo
  searches from multiple tools
• UML documents (possibly)
• A simple viewer
• Validation tools

6
http//code.google.com/p/psi-pi
7
(No Transcript)
8
AnalysisXML - Use Cases

• To allow the use of a third party tool to examine
  results from an MS, MS-MS, MSn, tag or de novo
  search. Any combination of these types of search
  should be supported.
• For a Peptide Mass Fingerprint (PMF) search it
  should be possible to reference the mass spectrum
  and the matched peaks for each peptide, where the
  mass spectrum is available.
• For an MS-MS search it should be possible to
  reference the mass spectrum supporting each
  peptide identification, where the mass spectrum
  is available.
• For a peptide sequence tag search, there should
  be sufficient information to validate that a
  result is correct.
• There should be sufficient information stored in
  the instance document to enable a user to run the
  same search on the same or another search engine.

9
AnalysisXML - Use Cases

• It should be possible to add the results of new
  searches to an existing instance document. For
  example, a PMF search and an MS-MS search of the
  same sample could be saved in the same instance
  document.
• It should be possible to write a viewer
  application that enables you to change an expect
  threshold and re-display the matches that fall
  within the expect value.
• Manual or automated annotation and validation of
  proteins / peptides should be available in the
  file for others to see.
• A third party tool could be used to save
  annotations and validations of identified
  proteins / peptides to an existing instance
  document.
• Quantitation and differential analysis using all
  the popular techniques should be supported.
• A third party tool should be able to generate a
  report detailing all the search parameters
  required by the relevant MCP and MIAPE guidelines.

10
AnalysisXML - Proposed use case

• The current direction of analysisXML (based on
  the use case document) is proteomics-specific,
  but the structure we originally drafted (and FuGE
  itself) is very general. I think analysisXML has
  the potential to become a standard for result
  output including LIMS output and other
  workflows.
• (Randy Julian, 10th Jan 2008)

11
AnalysisML - More use cases?

• Support spectral library searches
• From decoy results in the same or separate
  instance documents, determine the FDR for a
  search.

12
FuGE Based Development

• MDA - Model driven architecture
• Use MagicDraw, (Free community edition is OK) to
  develop UML model
• Generate XSD from model (requires reasonable
  effort to set this up)

13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)