[Psidev-ms-dev] Controlled vocabulary release candidate v. 3.13.0

[David O. <do...@eb...>]

Hi community,

Attached to this email (and conveniently compressed), the new version 
3.13.0.
If everybody thinks is fine, I'll submit this tomorrow evening (Europe 
time, of curse).

=====================================================
Changes in the 3.13.0 version
=====================================================

Added terms:
- MS:1001879 (offset voltage)
- MS:1001880 (in-source collision-induced dissociation)
- MS:1001881 (mz5 file)
- MS:1001882 (transition validation attribute)
- MS:1001883 (coefficient of variation)
- MS:1001884 (signal-to-noise ratio)
- MS:1001885 (command-line parameters)

Changed names:
- MS:1000877 (tube lens voltage)
- MS:1000139 (4000 QTRAP)

Changed definitions:
- MS:1000932 (TripleTOF 5600)
- MS:1000672 (Cliquid)



=====================================================
Pending things
=====================================================


Pending things 1 -> Two terms (MS:1001843 and id: MS:1001844 -peak 
intensity and area- now sons of MS:1000042 ! peak intensity) proposed by 
Eric Deutsch about “peak area” and “peak height” (last email Oct 10th 
"PSI-MS CV: peak intensity area terms").
Related to this, the term MS:1001845 ("peak area" too) has been 
obsoleted because duplicated toMS:1001844.
After this point, several mails re-opened the debate (mails from Oct12th 
- Oct25th).

Pending things 2 -> Some comments in Eric Deutsch mail (Oct 10th) not 
yet implemented in the controlled vocabulary and to be discussed :
2) Is there really a difference between “peak area” and “XIC area”. I 
suspect not. If there really is an intended subtle difference (e.g. 
“peak area” takes into account background removal, which “XIC area” is 
irrespective of background) then we should define this.
3) It seems to me that all these term names are incomplete and 
potentially misleading. For example, the “peak area” term is not really 
for the concept of “peak area”; it is the concept of “quantifying signal 
by measuring peak area”. One can infer this by knowing the parent, but 
if our goal is to create term names that can stand on their own, I think 
these should be clarified. It will be tempting to users to use the “peak 
area” term to provide a measurement of a peak area.
4) For 1859, normalized to what?
5) For 1130, can peptides have an area? Mass specs don’t see peptides, 
they see peptide ions. And they see them as peaks. So it would see that 
“peptide raw area” is a badly named term that probably when decomposed 
means the same as “peak area”. Or, if I’m wrong, can we improve the 
definition?

Pending things 3 -> Some comments by David Ovelleiro (mail Sept 27th, 
"Possible need of changing some things under "identification result 
details" (MS:1001405)") not yet implemented in the controlled vocabulary:
- comment 1: there are two terms, MS:1001362 and MS:1001114, which in 
addition to be children to their respective parents (MS:1001116 and 
MS:1001105 resp), are also direct children to MS:1001405. The problem I 
see here is that the two parents, are also direct descendants of 
MS:1001405. Is this not redundant and unnecessary? My proposal is to 
remove the is_a (direct) relationship to MS:1001405.
Please, check the picture "screen1.jpg" for a more graphical description.
- comment 2: the term "Mascot query number" (MS:1001528) is direct child 
to "spectrum identification result details" (MS:1001405). Don't you 
think that this term would be better placed under "search engine 
specific score" (MS:1001153) (child to the previous MS:1001528)
- comment 3: the terms related to the "False Discovery Rate" are, in my 
opinion, some confusing at this point. I attach a screen-shot called 
screen2.jpg to illustrate what I'm saying. At least two of the terms 
("pep:global FDR" and "prot:global FDR") seem miss located to me. Maybe 
they should work like "local FDR", with a unique term called "global 
FDR" child to both "peptide" and "proteine" / "identification confidence 
metric" (MS:1001198 and MS:1001092). Or maybe two terms could be used 
(the way is now), but children to MS:1001198 and MS:1001092 and changing 
the prefix "pep:" and "prot:" by the proper "peptide" and "protein".

Two replies (Eric Deutsch and David Creasy, mails Sept 27th) seem to 
give support to the first point. Second point rejected. Third point 
extended in Eric Deutsch mail.

Pending things 4 -> the "modification specificity N-term/C-term" related 
terms were NOT modified following the proposal in Martin Eisenacher mail 
(Sept 1st). In mail sent by David Ovelleiro (Sept 16th) changes related 
to modifications specificity were put to a stop until more consensus was 
reached. This should be clarified (pretty sure a final soultion for this 
is needed).




-- 
David Ovelleiro
Bioinformatician
PRIDE Group
Proteomics Services Team, PANDA Group
EMBL European Bioinformatics Institute
Wellcome Trust Genome Campus
Hinxton, Cambridge, UK
CB10 1SD