Menu
▾
▴
psidev-pi-dev
|
From: Angel P. <an...@ma...> - 2006-07-31 19:50:39
|
Folks, We have a conference call scheduled tomorrow. Sorry for the late notice, but been a little busy. Agenda: AnalysisUML Model --------------------- Last conference call, I had created a straw man UML model directly from the XML now (in)famous XML schema that myself, Randy Julian, Sean Seymour, David Shteynberg, and a few others (sorry I forget who all was there) for analysisXML. We used the ISB's pepXML and protXML as a starting point. There were a few questions regarding the model over the email list (archives found here: http://sourceforge.net/mailarchive/forum.php?thread_id=22147996&forum_id=48874 ) We will be going over these comments. Requirements spreadsheet for search engines ------------------------------------------------------------------------ David Creasy had started collating a list of search engine requirments. Status and overview of the spreadsheet, available here: http://psidev.sourceforge.net/docstore/view.php?&action=xls_show&id=84 email thread discussing the spreadsheet available here: http://sourceforge.net/mailarchive/forum.php?thread_id=22351418&forum_id=48874 Other agenda items as needed --------------------------------------------------- Phil Jones will not be present to set up the conference call, but his colleague Lennart Martins has kindly agreed to step in and set up the conference call logistics. Please contact him for teleconf details, but they should be the same as last time. len...@eb... Cheers! -- C Angel Pizarro Director, Bioinformatics Facility Institute for Translational Medicine and Therapeutics University of Pennsylvania 806 BRB II/III 421 Curie Blvd. Philadelphia, PA 19104-6160 P: 215-573-3736 F: 215-573-9004 E: an...@ma... |
|
From: Jones, A. <And...@li...> - 2015-01-14 17:02:13
|
Hi all, Conference call to progress mzIdentML at 4pm UK time tomorrow, http://www.timeanddate.com/worldclock/fixedtime.html?iso=20150115T16&p1=%3A&ah=1 Dial in details: + Germany: 08001012079 + Switzerland: 0800000860 + UK: 08081095644 + Generic international: +44 2083222500 (UK number) + US: 877-420-0272 access code: 297427 Agenda: 1. Brief summary of current status of mzid 1.2 2. Review of specification document: https://code.google.com/p/psi-pi/source/browse/trunk/specification_document/specdoc1_2/mzIdentML1.2-draft.doc 3. Discussion of open items: a. Various issues with the mzid 1.2 encoding identified by Harald (see email forwarded separately) 4. Plan for completing mzid 1.2 and writing manuscript 5. AOB Please send me additional items for discussion, Best wishes Andy |
|
From: Jones, A. <And...@li...> - 2015-01-15 16:59:52
|
Minutes from today ******************* Present: Yasset (EBI) Andy (Liverpool) Eric (ISB) Julian and Gerhard (Bochum) Lutz (Edinburgh) - Reviewed section 5.3 of spec doc o Multiple search engines text looks rather daunting – ACTION: Andy to add a figure and simplify text if possible - Pre-fractionation: o Various guidelines but these are mainly notes for implementers, should be simple for readers given that all reading software should load all PSMs from all lists flagged as “final PSM list” o ACTION: Gerhard, can you check whether the validator has any logic for checking whether a given spectrum identification (combination of spectrumID and spectrumRef) is unique within the “final PSM lists” – a file should be flagged as invalid if the same spectrum appears more than once - Peptide-level scores: o Various discussion about whether we should continue with the hacked method. o Yasset strongly expressed the view that Peaks, Thermo etc are only just getting mzid 1.1 export going; if we make schema changes, it will be disastrous for relations with these companies o General agreement to support peptide-level scores through the CV param mechanism proposed, rather than making more radical changes - Mod-localization for peptide scores: o Discussed that this cannot currently be supported o Proposal to try and see if this will work under current peptide-level CVparam mechanism e.g. by having new CV terms for “peptide:A-score” etc, as well as “a-score” for PSM-based mod localization scoring - Validation software – Gerhard said that most mzid 1.2 features are in there now o ACTION: Gerhard to add this to the SVN and send round a link - Crosslinking: o We are going to add the CV terms needed to PSI-MS, since it is proving difficult to get terms in Unimod or PSI-MOD without a lot more discussion - Completing mzid 1.2, Need example files: o ACTION: Lutz to update XL files o ACTION: Andy’s group to make peptide-level stats example file; and multiple search example file o ACTION: Marc/Harald to make examples for peptide-level stats and mod localization (and both together if possible) o ACTION: All – send necessary CV updates to the PSI-MS-vocab list so Gerhard can get these added - Manuscript: o ACTION: Andy to make a start on this; send to Juan and Yasset for help pulling together Please let me know if I missed anything, Best wishes Andy From: Jones, Andy [mailto:And...@li...] Sent: 14 January 2015 17:02 To: ''psi...@li...' (psi...@li...)' Subject: [Psidev-pi-dev] Conference call tomorrow Hi all, Conference call to progress mzIdentML at 4pm UK time tomorrow, http://www.timeanddate.com/worldclock/fixedtime.html?iso=20150115T16&p1=%3A&ah=1 Dial in details: + Germany: 08001012079 + Switzerland: 0800000860 + UK: 08081095644 + Generic international: +44 2083222500 (UK number) + US: 877-420-0272 access code: 297427 Agenda: 1. Brief summary of current status of mzid 1.2 2. Review of specification document: https://code.google.com/p/psi-pi/source/browse/trunk/specification_document/specdoc1_2/mzIdentML1.2-draft.doc 3. Discussion of open items: a. Various issues with the mzid 1.2 encoding identified by Harald (see email forwarded separately) 4. Plan for completing mzid 1.2 and writing manuscript 5. AOB Please send me additional items for discussion, Best wishes Andy |
|
From: Juan A. V. <jav...@gm...> - 2015-02-05 19:29:24
|
Hi all,
Please, let me know if I missed anything.
Present: Andy, Gerhard, Julian, Eric, Harald, Juan Antonio
The whole conference was devoted to checking some of the example files of mzIdentML 1.2 (which can be donwloaded from the SVN under examples 1.2):
1- Multi-search example (generated by Liverpool):
<additionalsearchparams> needs to contain an extra CVparam to indicate that the file contains results coming from different search engines in different PSM lists
The final PSM list CV could be put at the beginning, to facilitate the work of some readers. However, this would imply a schema change, so keep it in mind in the long list of possible improvements.
<cvParam cvRef="PSI-MS" accession="MS:1002439" name="final PSM list UNDER DISCUSSION"></cvParam>
TO DO: Julian to generate a new example file containing intermediate PSM lists.
Andy -> Update spec documentation (section 5.3.1 for listing all the search engines in the AnalysisProtocol)
2- Example containing peptide scores (generated by Liverpool):
<AdditionalSearchParams>
<cvParam cvRef="PSI-MS" accession="MS:1001211" name="parent mass type mono"></cvParam>
<cvParam cvRef="PSI-MS" accession="MS:1001256" name="fragment mass type mono"></cvParam>
<cvParam cvRef="PSI-MS" accession="MS:1002490" name="peptide-level scoring"></cvParam>
<cvParam cvRef="PSI-MS" accession="MS:1002496" name="group PSMs by sequence"></cvParam>
</AdditionalSearchParams>
Does the validator do anything with this information? It seems to be the case but it needs to be checked.
TO DO:
The validator needs to be corrected since it gives an error related to “de novo”.
Andy to post in the list his detailed solution to the current validation errors related to the peptide scores.
3- PeptideShaker example: it contains results from several search engines, peptide scores and mod localization scores in the same file.
- The validator needs to work out which CVterm to look for the regex (mod positition and related scores).
- Parent terms needs to be added in the CV for those that are peptide level and mod localization scores at the same time.
- Solution found to report peptide objects containing the final location of the mod, without taking into account the ones from the individual PSMs.
TO DO: Andy and Harald to try validator with the current examples to test it properly.
<additionalsearchparams> needs to contain an extra CVparam to indicate that the file contains results coming from different search engines in different PSM lists
4- Cross-linking related example.
Cross-linking example. Work ongoing by Lutz.
TO DO: Andy to get x-linkers added to the PSI-MS list.
Next call next Thursday at the same time.
Cheers,
Juan Antonio
|
|
From: Lutz F. <lfi...@st...> - 2015-02-12 18:55:18
|
The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. |
|
From: Jones, A. <And...@li...> - 2015-02-13 10:14:02
|
I certainly believe that we agreed for simplicity that all SIILists should have "MS:1002439:final PSM list UNDER DISCUSSION" or the “intermediate PSM list” term, even those most files are not using combined search engine approaches. I think this was agreed in case other cases arise where intermediate results for some other pre-processing are kept in the file.
I don’t know how to interpret the error message, possibly a bug?
From: Lutz Fischer [mailto:lfi...@st...]
Sent: 12 February 2015 18:55
To: psi...@li...; Jones, Andy
Cc: mayerg97
Subject: Re: [Psidev-pi-dev] Minutes call today (February 6th)
Hi all,
sorry missed last weeks meeting. I had an tooth-op and somehow for the rest of the day I was not really very usable - meaning I plainly forgot about it.
I have a question about "MS:1002439:final PSM list UNDER DISCUSSION".
With the 1.4.8 version of the validator I got an error message, that I have to put this in at least one list. So I added the code to do so. Only now I get a complain with version 1.4.10:
Message 1:
Rule ID: SpectrumIdentificationList_rule
Level: INFO
Context(/cvParam/@accession ) in 2 locations
--> Not all of the 1 values SpectrumIdentificationList's CV terms ['MS:1002439'] found using the Xpath '/cvParam/@accession' matched any of the 1 CvTerm(s):
- Any children term of MS:1001184 (search statistics). The term can be repeated. The matching value has to be the identifier of the term, not its name.
So I wonder what is the current state of that? Should I put the term in - even so I have only one list or leave it out?
Lutz
On 05/02/15 19:29, Juan Antonio Vizcaino wrote:
Hi all,
Please, let me know if I missed anything.
Present: Andy, Gerhard, Julian, Eric, Harald, Juan Antonio
The whole conference was devoted to checking some of the example files of mzIdentML 1.2 (which can be donwloaded from the SVN under examples 1.2):
1- Multi-search example (generated by Liverpool):
<additionalsearchparams> needs to contain an extra CVparam to indicate that the file contains results coming from different search engines in different PSM lists
The final PSM list CV could be put at the beginning, to facilitate the work of some readers. However, this would imply a schema change, so keep it in mind in the long list of possible improvements.
<cvParam cvRef="PSI-MS" accession="MS:1002439" name="final PSM list UNDER DISCUSSION"></cvParam>
TO DO: Julian to generate a new example file containing intermediate PSM lists.
Andy -> Update spec documentation (section 5.3.1 for listing all the search engines in the AnalysisProtocol)
2- Example containing peptide scores (generated by Liverpool):
<AdditionalSearchParams>
<cvParam cvRef="PSI-MS" accession="MS:1001211" name="parent mass type mono"></cvParam>
<cvParam cvRef="PSI-MS" accession="MS:1001256" name="fragment mass type mono"></cvParam>
<cvParam cvRef="PSI-MS" accession="MS:1002490" name="peptide-level scoring"></cvParam>
<cvParam cvRef="PSI-MS" accession="MS:1002496" name="group PSMs by sequence"></cvParam>
</AdditionalSearchParams>
Does the validator do anything with this information? It seems to be the case but it needs to be checked.
TO DO:
The validator needs to be corrected since it gives an error related to “de novo”.
Andy to post in the list his detailed solution to the current validation errors related to the peptide scores.
3- PeptideShaker example: it contains results from several search engines, peptide scores and mod localization scores in the same file.
- The validator needs to work out which CVterm to look for the regex (mod positition and related scores).
- Parent terms needs to be added in the CV for those that are peptide level and mod localization scores at the same time.
- Solution found to report peptide objects containing the final location of the mod, without taking into account the ones from the individual PSMs.
TO DO: Andy and Harald to try validator with the current examples to test it properly.
<additionalsearchparams> needs to contain an extra CVparam to indicate that the file contains results coming from different search engines in different PSM lists
4- Cross-linking related example.
Cross-linking example. Work ongoing by Lutz.
TO DO: Andy to get x-linkers added to the PSI-MS list.
Next call next Thursday at the same time.
Cheers,
Juan Antonio
------------------------------------------------------------------------------
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Psidev-pi-dev mailing list
Psi...@li...<mailto:Psi...@li...>
https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev
--
Lutz Fischer
lfi...@st...<mailto:lfi...@st...>
+44 131 6517057
|
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X