psidev-pi-dev — General Development List for Proteomics Informatics

Re: [Psidev-pi-dev] CV term "scoring model"

[Pierre-Alain B. <pie...@is...>]

I'm fine with only one term.
In Phenyx, the GUI description in two levels is pointing to one file for 
each combination. The important information is the scoring model xml 
file itself (esquire3000-gbaa-is-hmm.scoring.xml for instance). The 
items in the GUI are "only" descriptors to help choosing the appropriate 
scoring file.
As the GUI information is always interpreted, we should just define if 
the instrument details corresponds to the GUI selection or to the 
effective "scoring model used".


FYI, the two levels "instrument type" and "scoring model" are 
represented in Phenyx in this "hyerarchical" manner for different reasons.
- One is to show the possibility to generate multiple specific scoring 
models for an instrument type (LTQ and Qtrap do not fragment in the same 
way) or to differentiate fragmentation pattern interpretation between 
say unmodified and systematically modified peptides (such as iTRAQ for 
instance)
- An other one is to avoid the one-long-list drop-dowm menu
- A third one is to explicitely describe different instrumental setups 
but using the same scoring model (ESI-Linear trap LTQ is the same 
scoring model as ESI-LTQ-Orbitrap CID_LTQ_scan_LTQ). Some people have 
difficulty to choose an ESI-trap scoring if they use an Orbitrap. Some 
users do not understand the difference between a fragmentation in the 
LTQ or the C-trap and need to be "guided" through the interface.
- And a fourth one is to allow each one to define its own "relationship"



Martin Eisenacher wrote:
> Hi Pierre-Alain!
>
> In the AnalysisXML CV we have a branch 
> "instrument details" with a sub-branch "instrument type" (e.g. ESI-QUAD, MALDI-TOF) and a 
> sub-branch "scoring model" which was probably created with phenyx in mind.
>
> What do you think: 
> Leave it as ONE term and specify a (string) value which is the name of the
> scoring model xml file in phenyx?
>
> Or specify (additionally) child terms for "default" scorings:?
> For example:
> esquire3000+.Da_.1.scoring.xml
> esquire3000+.Da_.2.scoring.xml
> esquire3000-gbaa-is-hmm.scoring.xml
> ...
> Some more?
>
>   bye from Bochum
>    Martin
>
>
> ------------------------------------------------------------------------------
> SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
> The future of the web can't happen without you.  Join us at MIX09 to help
> pave the way to the Next Web now. Learn more and register at
> http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
> _______________________________________________
> Psidev-pi-dev mailing list
> Psi...@li...
> https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev
>
>   

Re: [Psidev-pi-dev] Minutes of Telecon. 11th December

[Eric D. <ede...@sy...>]

Hi everyone, I have just regenerated the docs based on the latest xsd, cv,
mapping, examples in SVN:

http://www.peptideatlas.org/tmp/AnalysisXML_working.html

There is still a known issue that some definitions are not quite being
pulled out of the xsd correctly, i.e. if a element has several definitions
depending on its context, one of the definitions is presented, but not
necessarily the correct one in that context. This will be fixed sometime.

Note also that the embedded figures have not been updated in a while and
might not be optimal. If someone sends me updates figures, I'll add.

The version suitable for pasting into a Word doc is:

http://www.peptideatlas.org/tmp/AnalysisXML_working_simple.html

Regards,
Eric





> -----Original Message-----
> From: Jones, Andy [mailto:And...@li...]
> Sent: Thursday, December 11, 2008 9:32 AM
> To: psi...@li...
> Subject: Re: [Psidev-pi-dev] Minutes of Telecon. 11th December
> 
> New schema uploaded by SVN:
> 
> Made rank attribute mandatory.
> 
> Added attribute passThreshold to SpectrumIdentificationItem and
> ProteinDetectionHypothesis.
> 
> 
> 
> > -----Original Message-----
> > From: Jennifer Siepen [mailto:jen...@ma...]
> > Sent: 11 December 2008 17:22
> > To: psi...@li...
> > Subject: [Psidev-pi-dev] Minutes of Telecon. 11th December
> >
> > Hi,
> >
> > Please find the minutes of todays teleconference here:
> >
> > http://www.psidev.info/index.php?q=node/400
> >
> > Thanks,
> >
> > Jenny
> >
> > --
> >
> > Dr. Jennifer Siepen
> > Research Associate
> > University of Manchester
> > Faculty of Life Sciences
> > Michael Smith Building
> > Manchester
> > M13 9PT
> >
> > Tel: 0161 275 1680
> >
> >
> > ------------------------------------------------------------------------
> ------
> > SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas,
> Nevada.
> > The future of the web can't happen without you.  Join us at MIX09 to
> help
> > pave the way to the Next Web now. Learn more and register at
> >
> http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.co
> m/
> > _______________________________________________
> > Psidev-pi-dev mailing list
> > Psi...@li...
> > https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev
> 
> --------------------------------------------------------------------------
> ----
> SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas,
> Nevada.
> The future of the web can't happen without you.  Join us at MIX09 to help
> pave the way to the Next Web now. Learn more and register at
> http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.co
> m/
> _______________________________________________
> Psidev-pi-dev mailing list
> Psi...@li...
> https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev

[Psidev-pi-dev] CV term "scoring model"

[Martin E. <mar...@ru...>]

Hi Pierre-Alain!

In the AnalysisXML CV we have a branch 
"instrument details" with a sub-branch "instrument type" (e.g. ESI-QUAD, MALDI-TOF) and a 
sub-branch "scoring model" which was probably created with phenyx in mind.

What do you think: 
Leave it as ONE term and specify a (string) value which is the name of the
scoring model xml file in phenyx?

Or specify (additionally) child terms for "default" scorings:?
For example:
esquire3000+.Da_.1.scoring.xml
esquire3000+.Da_.2.scoring.xml
esquire3000-gbaa-is-hmm.scoring.xml
...
Some more?

  bye from Bochum
   Martin

[Psidev-pi-dev] Issue 42 in psi-pi: Issues with the CV

[<cod...@go...>]

Comment #61 on issue 42 by eisenachM: Issues with the CV
http://code.google.com/p/psi-pi/issues/detail?id=42

Discussed/Solved in TeleCon 11th December:

Comment 38:
nativeID: change back to just using mzML 'id'? Andy to change doc

Comment 40:
- no Newt.obo: ignored in validator
- 2nd parent for Paragon scores: done during DocProc (Sean)

Comment 41:
- OBOEdit cannot handle UO import and relationship rows
=> use OBOEdit 2 beta
=> try relationship rows without colon

Comment 44:
- <SourceFile> and <pf:fileFormat> elements are not renamed
(URIs may point to a database "location")
- Martin to add CV term "data stored in database"

Comment 57:
- "use child terms only" is different to "use leaf terms only"
(because child can be internal, if it has children itself)
=> no mechanism to prevent
=> leave CV structure very simple
=> name internal nodes so that they are not being used "intuitively"

Comment 60 (decoy terms):
- Martin to do it like he suggested (instead of real+virtual: forward+decoy)


-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

Re: [Psidev-pi-dev] Minutes of Telecon. 11th December

[Jones, A. <And...@li...>]

New schema uploaded by SVN:

Made rank attribute mandatory.

Added attribute passThreshold to SpectrumIdentificationItem and
ProteinDetectionHypothesis.



> -----Original Message-----
> From: Jennifer Siepen [mailto:jen...@ma...]
> Sent: 11 December 2008 17:22
> To: psi...@li...
> Subject: [Psidev-pi-dev] Minutes of Telecon. 11th December
> 
> Hi,
> 
> Please find the minutes of todays teleconference here:
> 
> http://www.psidev.info/index.php?q=node/400
> 
> Thanks,
> 
> Jenny
> 
> --
> 
> Dr. Jennifer Siepen
> Research Associate
> University of Manchester
> Faculty of Life Sciences
> Michael Smith Building
> Manchester
> M13 9PT
> 
> Tel: 0161 275 1680
> 
> 
> ------------------------------------------------------------------------------
> SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
> The future of the web can't happen without you.  Join us at MIX09 to help
> pave the way to the Next Web now. Learn more and register at
> http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
> _______________________________________________
> Psidev-pi-dev mailing list
> Psi...@li...
> https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev

[Psidev-pi-dev] Minutes of Telecon. 11th December

[Jennifer S. <jen...@ma...>]

Hi,

Please find the minutes of todays teleconference here:

http://www.psidev.info/index.php?q=node/400

Thanks,

Jenny

-- 

Dr. Jennifer Siepen
Research Associate
University of Manchester
Faculty of Life Sciences
Michael Smith Building
Manchester
M13 9PT

Tel: 0161 275 1680

Re: [Psidev-pi-dev] PSI-PI Working group conference call on Thursday 11th December at 4:00pm UK time

[Jones, A. <And...@li...>]

Hi all,

Here are a couple of additional minor issues to discuss tomorrow:

- Encoding NCBI translational tables, I'm not sure about the current encoding in
the instance doc.
- Comment from Martin on the spec doc as to whether OBOEdit can handle the
has_units addition to the CV

Cheers
Andy

> -----Original Message-----
> From: David Creasy [mailto:dc...@ma...]
> Sent: 10 December 2008 10:57
> To: psi...@li...
> Subject: [Psidev-pi-dev] PSI-PI Working group conference call on Thursday 11th
> December at 4:00pm UK time
> 
> Hi everyone,
> 
> There will be a short AnalysisXML working group conference call tomorrow
> at 4:00pm UK time:
> http://www.timeanddate.com/worldclock/fixedtime.html?day=11&month=12&year=2
> 008&hour=16&min=00&sec=0&p1=136
> 
> Minutes from last meeting at:
> http://psidev.info/index.php?q=node/398
> 
> Agenda:
> - Review minutes of last meeting
> - Native ID. Looks as though we change back to just using mzML 'id'?
> - Do we want only CV leaves in an instance document or are internal
>    nodes possible too?
>      http://code.google.com/p/psi-pi/issues/detail?id=42#c57
>    Andreas to follow up and present a solution.
> - Number of peptides / proteins to be reported:
>      http://code.google.com/p/psi-pi/issues/detail?id=45
> - Any more CV issues? Review list of unused CV items if any?
> - Any more issues with the examples?
> - Outstanding actions before submitting to the document process.
> - Use of issue tracker.
> 
> Latest documentation:
> http://psi-
> pi.googlecode.com/svn/trunk/specification_document/analysisXML_SpecificationDr
> aft.doc
> 
> 
> Dial in details:
> 
> + Germany: 08001012079
> + Switzerland: 0800000860
> + UK: 08081095644
> + USA: 1-866-314-3683
> + Generic international: +44 2083222500 (UK number)
> 
> access code: 297427
> 
> --
> David Creasy
> Matrix Science
> 64 Baker Street
> London W1U 7GB, UK
> Tel: +44 (0)20 7486 1050
> Fax: +44 (0)20 7224 1344
> 
> dc...@ma...
> http://www.matrixscience.com
> 
> Matrix Science Ltd. is registered in England and Wales
> Company number 3533898
> 
> 
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
> The future of the web can't happen without you.  Join us at MIX09 to help
> pave the way to the Next Web now. Learn more and register at
> http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
> _______________________________________________
> Psidev-pi-dev mailing list
> Psi...@li...
> https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev

[Psidev-pi-dev] Issue 42 in psi-pi: Issues with the CV

[<cod...@go...>]

Comment #60 on issue 42 by eisenachM: Issues with the CV
http://code.google.com/p/psi-pi/issues/detail?id=42

Suggestion:
At the moment we have some CV terms below "quality estimation with decoy  
database",
which actually describe the pre-generated search database (like "decoy DB  
accession
regexp" or "decoy DB derived from ...").

We should move them below "search database details" as branch "decoy DB  
details"
together with "decoy DB generation algorithm", "decoy DB type"  
(e.g. "reverse",
"shuffle", "random") and "decoy DB composition" (e.g. "real+virtual"  
[formerly known
as "forward+reverse"] and "only virtual").

All this is for decoy searches using a pre-generated database.

Decoy approaches with no explicite decoy database (like Mascot Decoy?) need  
a new
term "quality estimation with implicite decoy" next to "quality estimation  
with decoy
DB" under "search input details".



-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

[Psidev-pi-dev] PSI-PI Working group conference call on Thursday 11th December at 4:00pm UK time

[David C. <dc...@ma...>]

Hi everyone,

There will be a short AnalysisXML working group conference call tomorrow 
at 4:00pm UK time:
http://www.timeanddate.com/worldclock/fixedtime.html?day=11&month=12&year=2008&hour=16&min=00&sec=0&p1=136

Minutes from last meeting at:
http://psidev.info/index.php?q=node/398

Agenda:
- Review minutes of last meeting
- Native ID. Looks as though we change back to just using mzML 'id'?
- Do we want only CV leaves in an instance document or are internal
   nodes possible too?
     http://code.google.com/p/psi-pi/issues/detail?id=42#c57
   Andreas to follow up and present a solution.
- Number of peptides / proteins to be reported:
     http://code.google.com/p/psi-pi/issues/detail?id=45
- Any more CV issues? Review list of unused CV items if any?
- Any more issues with the examples?
- Outstanding actions before submitting to the document process.
- Use of issue tracker.

Latest documentation:
http://psi-pi.googlecode.com/svn/trunk/specification_document/analysisXML_SpecificationDraft.doc


Dial in details:

+ Germany: 08001012079
+ Switzerland: 0800000860
+ UK: 08081095644
+ USA: 1-866-314-3683
+ Generic international: +44 2083222500 (UK number)

access code: 297427

-- 
David Creasy
Matrix Science
64 Baker Street
London W1U 7GB, UK
Tel: +44 (0)20 7486 1050
Fax: +44 (0)20 7224 1344

dc...@ma...
http://www.matrixscience.com

Matrix Science Ltd. is registered in England and Wales
Company number 3533898

Re: [Psidev-pi-dev] Issue 42 in psi-pi: Issues with the CV

[Martin E. <mar...@ru...>]

> Do we allow internal nodes of the PSI:PI ontology show up in an axml
> instance?
> 
> E.g. at
> /AnalysisXML/AnalysisProtocolCollection/ProteinDetectionProtocol/AnalysisParams
> the
> mapping file currently specifies that only the child terms of
> "PI:00194 - quality estimation with decoy database" are allowed. However,
> the MPC
> example uses exactly this term, to say that a quality estimation was done
> using a
> decoy database. What I like to do is moving all the children of PI:00194 up
> into "search input details -> quality estimation options" and keep the term as an
> indicator that quality estimation was done by decoy database.
At this position in the CV I can imagine to allow only child terms.
The MPC example is redundant as
it uses "PI:00194 - quality estimation with decoy database" and
"PI:00197 - forward+reverse decoy DB" and
"PI:00283 - decoy DB accession regexp".

The first (non-child) term could be dropped, so no changes are necessary.

What we need for the example is another term "decoy DB generation method" or
"decoy DB generation algorithm". 
I will propose that separately in issue 42.
 
  Bye
    Martin

[Psidev-pi-dev] Issue 42 in psi-pi: Issues with the CV

[<cod...@go...>]

Comment #59 on issue 42 by jensiepen: Issues with the CV
http://code.google.com/p/psi-pi/issues/detail?id=42

Terms needed for the SpectraST example

output file type is pepXML (this file type is not specific to SpectraST)

and associated peptide/spectra scores
dot : the dot product of two spectra, measuring spectral similarity
dot_bias : a measure of how much of the dot product is dominated by a few  
peaks
discriminant score F : spectraST spectrum score
delta: normalised difference between dot product of top hit and runner-up

-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

[Psidev-pi-dev] Issue 42 in psi-pi: Issues with the CV

[<cod...@go...>]

Comment #58 on issue 42 by jensiepen: Issues with the CV
http://code.google.com/p/psi-pi/issues/detail?id=42

Terms needed for the SpectraST example

output file type is pepXML (this file type is not specific to SpectraST)

and two peptide/spectra scores
dotvalue : the dot product of two spectra, measuring spectral similarity
dot_biasvalue : a measure of how much of the dot product is dominated by a  
few peaks

-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

[Psidev-pi-dev] Issue 45 in psi-pi: Number of peptides / proteins to be reported

[<cod...@go...>]

Comment #1 on issue 45 by andrewrobertjones: Number of peptides / proteins  
to be reported
http://code.google.com/p/psi-pi/issues/detail?id=45

I think we can support this for SpectrumIdentifications with no schema  
change. We
can encourage data producers to produce large number of spectrum  
identifications.
The identifications deemed "correct" are the ones referenced from
ProteinDetectionHypotheses. This then also supports the use case of peptide
identification scores being boosted once a protein detection has been made.  
PRIDE
can make a call on how many of the SpectrumIdentifications to load.

This does not solve the FDR problem for proteins though. However, I think  
the
solution of creating two different instance documents would be okay. If  
sending to
PRIDE, just send the list of proteins found. If using analysisXML as an  
intermediate
format, send a long list of proteins (and peptides) with no set criteria  
for what
is "correct".

-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

[Psidev-pi-dev] Issue 45 in psi-pi: Number of peptides / proteins to be reported

[<cod...@go...>]

Status: Accepted
Owner: dcreasy
Labels: Type-Defect Priority-Medium

New issue 45 by dcreasy: Number of peptides / proteins to be reported
http://code.google.com/p/psi-pi/issues/detail?id=45

Writing documentation always brings a few problems to the surface...

Consider 2 separate use cases:
  - Import into a repository such as Pride
  - Import into an application such as Scaffold or Peptide Prophet

With the first, you probably only want to report and store proteins above a
certain threshold that for example gives an FDR described in the instance
document.
With the second, we need to give scores for all spectra. (See:
http://code.google.com/p/psi-pi/wiki/NotesForDocumentation#How_many_should__be_saved
for details).

One option is to insist that these two cases require two different instance
documents.

Another option is to 'encourage' people to save a larger number of results
and add (say) a boolean attribute to <SpectrumIdentificationItem> and
<ProteinDetectionHypothesis> that indicates that the result is above the
specified thresholds.
Without the boolean flag, if all results are saved (as in the current
Mascot examples), a repository such as Pride would need to understand the
relevant CV and only store results above a given score.

Any thoughts / preferences?



-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

[Psidev-pi-dev] Issue 30 in psi-pi: Specifying the enzyme rules used in the search

[<cod...@go...>]

Updates:
	Status: Fixed

Comment #40 on issue 30 by dcreasy: Specifying the enzyme rules used in the  
search
http://code.google.com/p/psi-pi/issues/detail?id=30

(No comment was entered for this change.)

-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

[Psidev-pi-dev] New spec doc

[Jones, A. <And...@li...>]

Hi all,
New spec doc committed by SVN. Please review this and return comments to me by
Tues evening. 

Any problems with the instance doc examples in Section 6 are coming from various
examples:

  --examplesFile
../svn/examples/Mascot_MSMS_example.axml,../svn/examples/MPC_example.axml,../svn
/examples/PMF_example.axml,../svn/examples/Mascot_N15_example.axml,../svn/exampl
es/Mascot_NA_example.axml

So, please fix these rather than telling me :-)

At the moment, the hyperlinks in the document go back to the HTML version online
rather than within document. Anyone know how to fix these throughout the Word
doc?

If anyone has flagged a problem to me that I haven't updated in this version,
feel free to tell me again, I may have missed things...

Cheers
Andy

 

[Psidev-pi-dev] Minutes of Telecon 5 Dec. 2008

[Phil J. @ E. <pj...@eb...>]

Hi,

Please find the minutes here:

http://psidev.info/index.php?q=node/398

best regards,

Phil.

-- 
Phil Jones
Senior Software Engineer
InterPro Team
PANDA Group, EMBL-EBI
Wellcome Trust Genome Campus
Hinxton, Cambridge, CB10 1SD
UK.

Work phone: +44 1223 494406 (NEW NUMBER)
Skype: philip-jones

[Psidev-pi-dev] Guide to adding minutes for PSI-PI Meetings.

[Phil J. @ E. <pj...@eb...>]

Hi all,

Please find a guide to adding minutes and maintaining the PSI-PI pages here:

http://psidev.info/index.php?q=node/397

And reproduced here:



http://psidev.info Web Site
===============

Guide to adding minutes for the PSI-PI Working Group

(NB - written as if for someone who has never seen the site before, so
plenty of obvious stuff in here...)

Getting a new Account
---------------------
Click on 'Request new account' if you currently have no account and
subscribe to the 'Proteomic Informatics' group

Adding An Agenda / Meeting Time
-------------------------------

1. Log in.
2. New menu items will appear on the left hand side - click on the
'event' item under 'create content'
3. Add the start / end date and time and indicate the time zone.  Give
the meeting a title and some content in the 'Body' section (this is
mandatory).
4. It is a good idea to include international time for the meeting in
the body, that can be set up on the 'World Clock' website
http://www.timeanddate.com/worldclock/ by following the 'Fixed, Past,
Future Time' link.
5. Under the 'Categories' section on the Meeting page, select 'Phone
Conferences' under the 'Proteomics Informatics' heading.
6. Check the 'Public' checkbox.  (Ignore the 'Proteomic Informatic' checkbox.)
7. Click Submit.

The meeting day will now be highlighted on the 'Events' calendar on the left.

Adding Minutes
--------------

1. Log in
2. Click on the date of the meeting on the Events Calendar.
3. The meeting agenda will appear inside a pink outlined box.
4. Click on the Meeting title in the pink box.
5. At the foot of the resulting page, click on 'post minutes for this event'.
6. Add Attendees and minutes as required.
5. Under the 'Categories' section on the Meeting page, select 'Phone
Conferences' under the 'Proteomics Informatics' heading.
6. Check the 'Public' checkbox.  (Ignore the 'Proteomic Informatic' checkbox.)
7. Click Submit.

Updating the PSI-PI Home Page
-----------------------------

1. Log in
2. Navigate to http://psidev.info/index.php?q=node/319
3. Click on the 'edit' tab at the top.
4. In the 'Body' editor pane, scroll down to the 'Next Meeting /
Teleconference' and 'Meeting Minutes' section.
5. Update as necessary with links to the new agenda / minutes pages.
Note that the rich text editor includes a 'link' and 'unlink' icon for
adding and removing hyperlinks.






Best regards,

Phil.

-- 
Phil Jones
Senior Software Engineer
InterPro Team
PANDA Group, EMBL-EBI
Wellcome Trust Genome Campus
Hinxton, Cambridge, CB10 1SD
UK.

Work phone: +44 1223 494406 (NEW NUMBER)
Skype: philip-jones

[Psidev-pi-dev] PSI-PI Working group conference call on Friday 5th December at 1:30pm UK time

[David C. <dc...@ma...>]

Hi everyone,

There will be a short AnalysisXML working group conference call today at 
1:30pm UK time:
http://www.timeanddate.com/worldclock/fixedtime.html?day=5&month=12&year=2008&hour=13&min=30&sec=0&p1=136

Sorry for the late notice. It's been hard to find a time when people are 
available and apologies that it's too early in the day for folks from the US

Minutes from last meeting at:
http://psidev.info/index.php?q=node/395

Agenda:

- Hand over to the new secretary. Any issues?
- Future conference calls
- Outstanding actions before submitting to the document process.
Latest documentation:

http://psi-pi.googlecode.com/svn/trunk/specification_document/analysisXML_SpecificationDraft.doc


Dial in details:

+ Germany: 08001012079
+ Switzerland: 0800000860
+ UK: 08081095644
+ USA: 1-866-314-3683
+ Generic international: +44 2083222500 (UK number)

access code: 297427

-- 
David Creasy
Matrix Science
64 Baker Street
London W1U 7GB, UK
Tel: +44 (0)20 7486 1050
Fax: +44 (0)20 7224 1344

dc...@ma...
http://www.matrixscience.com

Matrix Science Ltd. is registered in England and Wales
Company number 3533898

Re: [Psidev-pi-dev] Vacancy: secretary required for the Proteomics Informatics group

[David C. <dc...@ma...>]

Hi everyone,

I'm pleased to say that Jenny Siepen from Manchester has agreed to take 
on this role. And a big thanks to Phil from all of us.

David

David Creasy wrote:
> Hi all,
> 
> Phil Jones from the EBI has been the very capable secretary for PSI-PI 
> group for quite some time. Sadly, he is no longer able to be involved 
> with analysisXML and needs to step down as the secretary. There is 
> therefore a need a replacement. Obvious duties involve taking minutes at 
> the weekly telephone conference calls and updating the PSI-PI web pages 
> etc.
> 
> If you would like to be considered for the role, please contact
> me off list by the end of Wednesday 3rd December.
> 
> General PSI Org chart is here:
> http://psidev.info/index.php?q=node/202
> 
> Thanks,
> 
> David Creasy
> (dc...@ma...)
> 

-- 
David Creasy
Matrix Science
64 Baker Street
London W1U 7GB, UK
Tel: +44 (0)20 7486 1050
Fax: +44 (0)20 7224 1344

dc...@ma...
http://www.matrixscience.com

Matrix Science Ltd. is registered in England and Wales
Company number 3533898

[Psidev-pi-dev] Issue 42 in psi-pi: Issues with the CV

[<cod...@go...>]

Comment #57 on issue 42 by a.bertsch0815: Issues with the CV
http://code.google.com/p/psi-pi/issues/detail?id=42

Do we allow internal nodes of the PSI:PI ontology show up in an axml  
instance?

E.g. at
/AnalysisXML/AnalysisProtocolCollection/ProteinDetectionProtocol/AnalysisParams  
the
mapping file currently specifies that only the child terms of
"PI:00194 - quality estimation with decoy database" are allowed. However,  
the MPC
example uses exactly this term, to say that a quality estimation was done  
using a
decoy database. What I like to do is moving all the children of PI:00194 up  
into
"search input details -> quality estimation options" and keep the term as an
indicator that quality estimation was done by decoy database.

To generalize this, I would enforce through semantic validation hat only  
leaves of
the ontology can be used. This makes the ontology kind of flat, however, it  
reduces
the complexity while interpreting the used terms and is much clearer I  
think.

I haven't tried how many changes are needed, but IMO only smaller changes  
to the CV
would be necessary. Any comments? Maybe we could discuss it during the  
telcon today.

-- 
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

Re: [Psidev-pi-dev] FW: Spec doc

[Eric D. <ede...@sy...>]

I have refreshed the on-line docs at:

http://www.peptideatlas.org/tmp/AnalysisXML_working.html

I notice that several of the xxxx_ref attributes do not have definitions.

Regards,
Eric


> -----Original Message-----
> From: David Creasy [mailto:dc...@ma...]
> Sent: Thursday, December 04, 2008 9:50 AM
> To: Andreas Bertsch
> Cc: psi...@li...
> Subject: Re: [Psidev-pi-dev] FW: Spec doc
> 
> Hi Andreas,
> 
> Andreas Bertsch wrote:
> > Hi David,
> >
> >> Andy suggested:
> >> Andreas can you add: "search tolerance plus value" and "search
> tolerance
> >> minus value" to the CV and mapping file and I'll change the schema,
> does
> >> this seem reasonable to everyone.?
> > Sorry, that is my fault, I have overlooked this structure. I have added
> the cv
> > two terms, 412 and 413, plus and minus value. The Mascot MSMS example
> would
> > look like this using the new terms.
> Thanks. I've updated the Mascot examples, and all seem to validate
> properly now.
> 
> >
> > I have also updated the mapping file accordingly. Now two values
> (exactly two
> > terms) are required, one plus and one minus value. The units are
> directly
> > associated with the two terms. So the "mass units" tree becomes obsolete
> > (absolut(e)ly), and I would suggest to delete it, if there is no good
> reason
> > to keep it.
> Yes, I would think that we can delete these.
> 
>  > I also think that a distinction into absolute and relative is not
>  > necessary as it is clear by specifying the units.
> Agree.
> 
> I've just one remaining validation error:
> 
> Error: CV term used in invalid element: 'PI:00410 - translation start
> codons' at element
> '/AnalysisXML/AnalysisProtocolCollection/SpectrumIdentificationProtocol/Da
> tabaseTranslation/TranslationTable'
> 
> I think that the mapping is wrong - or maybe you need a new parent term
> for this and PI:00025?
> 
> David
> 
> 
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the
> world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Psidev-pi-dev mailing list
> Psi...@li...
> https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev

Re: [Psidev-pi-dev] FW: Spec doc

[David C. <dc...@ma...>]

Hi Andreas,

Andreas Bertsch wrote:
> Hi David,
> 
>> Andy suggested:
>> Andreas can you add: “search tolerance plus value” and “search tolerance
>> minus value” to the CV and mapping file and I’ll change the schema, does
>> this seem reasonable to everyone…?
> Sorry, that is my fault, I have overlooked this structure. I have added the cv 
> two terms, 412 and 413, plus and minus value. The Mascot MSMS example would 
> look like this using the new terms.
Thanks. I've updated the Mascot examples, and all seem to validate 
properly now.

> 
> I have also updated the mapping file accordingly. Now two values (exactly two 
> terms) are required, one plus and one minus value. The units are directly 
> associated with the two terms. So the "mass units" tree becomes obsolete 
> (absolut(e)ly), and I would suggest to delete it, if there is no good reason 
> to keep it. 
Yes, I would think that we can delete these.

 > I also think that a distinction into absolute and relative is not
 > necessary as it is clear by specifying the units.
Agree.

I've just one remaining validation error:

Error: CV term used in invalid element: 'PI:00410 - translation start 
codons' at element 
'/AnalysisXML/AnalysisProtocolCollection/SpectrumIdentificationProtocol/DatabaseTranslation/TranslationTable'

I think that the mapping is wrong - or maybe you need a new parent term 
for this and PI:00025?

David

Re: [Psidev-pi-dev] FW: Spec doc

[Pierre-Alain B. <pie...@is...>]

Yes, there are tools using more complex rules than only one global metrics:

Phenyx uses this for the fragments in some scoring models (although not 
specified by the user),
Aldente has the option for PMF
Profound proposes for PMF two separate tolerances for data submitted 
with monoisotopic or average precision (a combination is possible)
MSTag (from ProteinProspector) has a "systematic error" value with the 
same unit as the parent tolerance one
MS-Fit has a similar option
Mascot allows "local" tolerance values if the input file is mgf (at 
least the format allows to specify specific tolerances for a given spectrum)

I believe that now we have to live with a first version that considers 
only one simple metrics, particularly if including the complexity would 
require a schema change.

Pierre-Alain


Andreas Bertsch wrote:
> Hi Pierre-Alain,
>
>   
>> Just a question about these tolerances:
>> if one need to specify a union of 2 sets of tolerances, such as 200ppm
>> and 0.1Da (which means 200ppm with a minimum of 0.1Da), can one just
>> have twice the pair of tolerances? for instance:
>>       <FragmentTolerance>
>>         <pf:cvParam accession="PI:00412" name="search tolerance plus value"
>> value="0.5" cvRef="PSI-PI" unitAccession="UO:0000221" unitName="dalton"
>> unitCvRef="UO" />
>>         <pf:cvParam accession="PI:00412" name="search tolerance plus value"
>> value="500" cvRef="PSI-PI" unitAccession="UO:0000221" unitName="ppm"
>> unitCvRef="UO" />
>>         <pf:cvParam accession="PI:00413" name="search tolerance minus
>> value" value="0.5" cvRef="PSI-PI" unitAccession="UO:0000221"
>> unitName="dalton" unitCvRef="UO" />
>>         <pf:cvParam accession="PI:00412" name="search tolerance plus value"
>> value="500" cvRef="PSI-PI" unitAccession="UO:0000221" unitName="ppm"
>> unitCvRef="UO" />
>>       </FragmentTolerance>
>>     
> At the moment this cannot be done. And just repeating the terms would be 
> confusing in my opinion. The meaning that one is a minimum mass tolerance and 
> the other one is the one used if it is greater than the threshold would not be 
> directly obvious while reading the file. If we want to support such constructs 
> we need schema changes here, I think. Are there any search engines which allow 
> for such a specification of the tolerance?
>
> Cheers,
> A.
>
>
>
>   

Re: [Psidev-pi-dev] FW: Spec doc

[Andreas B. <be...@in...>]

Hi Pierre-Alain,

> Just a question about these tolerances:
> if one need to specify a union of 2 sets of tolerances, such as 200ppm
> and 0.1Da (which means 200ppm with a minimum of 0.1Da), can one just
> have twice the pair of tolerances? for instance:
>       <FragmentTolerance>
>         <pf:cvParam accession="PI:00412" name="search tolerance plus value"
> value="0.5" cvRef="PSI-PI" unitAccession="UO:0000221" unitName="dalton"
> unitCvRef="UO" />
>         <pf:cvParam accession="PI:00412" name="search tolerance plus value"
> value="500" cvRef="PSI-PI" unitAccession="UO:0000221" unitName="ppm"
> unitCvRef="UO" />
>         <pf:cvParam accession="PI:00413" name="search tolerance minus
> value" value="0.5" cvRef="PSI-PI" unitAccession="UO:0000221"
> unitName="dalton" unitCvRef="UO" />
>         <pf:cvParam accession="PI:00412" name="search tolerance plus value"
> value="500" cvRef="PSI-PI" unitAccession="UO:0000221" unitName="ppm"
> unitCvRef="UO" />
>       </FragmentTolerance>
At the moment this cannot be done. And just repeating the terms would be 
confusing in my opinion. The meaning that one is a minimum mass tolerance and 
the other one is the one used if it is greater than the threshold would not be 
directly obvious while reading the file. If we want to support such constructs 
we need schema changes here, I think. Are there any search engines which allow 
for such a specification of the tolerance?

Cheers,
A.



-- 
Div. for Simulation of Biological Systems, WSI, University of Tuebingen
Room C322, Sand 14, 72076 Tuebingen, Germany
phone: +49-7071-29-70461 fax: +49-7071-29-5152
http://www-bs.informatik.uni-tuebingen.de