Menu
▾
▴
Re: [Psidev-ms-dev] [Psidev-ms-vocab] [Psidev-pi-dev] mzid 1.2 reviews
|
From: mayerg97 <ger...@ru...> - 2017-01-26 15:28:53
|
Hallo Andy, in that case I would simply propose to add some sub-terms to [Term] id: MS:1001805 name: quantification datatype def: "The data type of the value reported in a QuantLayer for a feature, peptide, protein, protein group." [PSI:MS] is_a: MS:1001129 ! quantification information e.g. [Term] id: MS:1002735 name: feature-level quantification datatype def: "The data type of the value reported in a QuantLayer for a feature." [PSI:PI] is_a: MS:1001805 ! quantification datatype [Term] id: MS:1002736 name: PSM-level quantification datatype def: "The data type of the value reported in a QuantLayer for a PSM." [PSI:PI] is_a: MS:1001805 ! quantification datatype [Term] id: MS:1002737 name: peptide-level quantification datatype def: "The data type of the value reported in a QuantLayer for a peptide." [PSI:PI] is_a: MS:1001805 ! quantification datatype [Term] id: MS:1002738 name: protein-level quantification datatype def: "The data type of the value reported in a QuantLayer for a protein." [PSI:PI] is_a: MS:1001805 ! quantification datatype [Term] id: MS:1002739 name: protein group-level quantification datatype def: "The data type of the value reported in a QuantLayer for a protein group." [PSI:PI] is_a: MS:1001805 ! quantification datatype and to rearrange the current childs of MS:1001805 under these new sub-terms. Would that make sense to you? Best regards, Gerhard Am 26.01.2017 um 10:47 schrieb Jones, Andy: > > Hi Gerhard, > > The part I’m not understanding yet is that we have some CV terms that > are intended to be for peptides in mzid 1.2, meaning a set of grouped > PSMs. Reading software would have to know that they refer to the > grouped peptide concept and not the PSM in which it located them. As > an example: > > [Term] > > id: MS:1002551 > > name: peptide:Ascore > > def: "A-score for PTM site location at the peptide-level." [PSI:PI] > > xref: value-type:xsd\:string "The allowed value-type for this CV term." > > is_a: MS:1002549 ! PTM localization distinct peptide-level statistic > > relationship: has_regexp MS:1002505 ! regular expression for > modification localization scoring > > Which is different from: > > [Term] > > id: MS:1001985 > > name: Ascore > > def: "A-score for PTM site location at the PSM-level." > [DOI:10.1038/nbt1240, PMID:16964243] > > xref: value-type:xsd\:string "The allowed value-type for this CV term." > > is_a: MS:1001968 ! PTM localization PSM-level statistic > > relationship: has_regexp MS:1002505 ! regular expression for > modification localization scoring > > Somewhere higher up in the hierarchy there should be a branch for PSMs > and a branch for peptide level stats or measures. The new terms > proposed need to be within the peptide branch, that’s all I was > suggesting. > > Best wishes > > Andy > > *From:*mayerg97 [mailto:ger...@ru...] > *Sent:* 24 January 2017 11:55 > *To:* psi...@li... > *Subject:* Re: [Psidev-ms-vocab] [Psidev-pi-dev] mzid 1.2 reviews > > Hi Andy, > > yes I think you're right that the parent term 'spectral count peptide > level quantitation' is wrong here, > > since we have the following hierarchy: > > quantitation analysis summary > > spectral counting quantitation analysis > > spectral count peptide level quantitation > > and quantitation analysis summary is defined as "The overall workflow > of this quantitation report." > > > On the other side I don't really understand, why the parent > 'quantification datatype' is > not enough, since by the name of the following three terms it should > be clear that they > are intended as spectral count terms for peptides and not PSM's: > > [Term] > id: MS:1001842 > name: sequence-level spectral count > def: "The number of MS2 spectra identified for a raw peptide sequence > without PTMs and charge state in spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this CV term." > is_a: MS:1001805 ! quantification datatype > > [Term] > id: MS:1002733 > name: peptide-level spectral count > def: "The number of MS2 spectra identified for a peptide sequence > specified by the amino acid one-letter codes plus optional PTMs in > spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this CV term." > is_a: MS:1001805 ! quantification datatype > > [Term] > id: MS:1002734 > name: peptide ion-level spectral count > def: "The number of MS2 spectra identified for a molecular ion defined > by the peptide sequence represented by the amino acid one-letter > codes, plus optional PTMs plus optional charged aducts plus the charge > state, in spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this CV term." > is_a: MS:1001805 ! quantification datatype > > Best regards, > Gerhard > > Am 20.01.2017 um 12:24 schrieb Jones, Andy: > > Hi Gerhard, > > Thanks for this. Can I check – we want to have these as > peptide-level data types. Can you recall the mapping mechanism > that would tell reading software that these are for peptides > rather than PSMs i.e. they would be repeated for all the PSMs > grouped as the same peptide? The parent terms are not completely > clear to me. > > Best wishes > > Andy > > *From:*mayerg97 [mailto:ger...@ru...] > *Sent:* 20 January 2017 09:41 > *To:* psi...@li... > <mailto:psi...@li...>; > psi...@li... > <mailto:psi...@li...>; > psi...@li... > <mailto:psi...@li...> > *Subject:* Re: [Psidev-ms-vocab] [Psidev-pi-dev] mzid 1.2 reviews > > Hello all, > > so according Pierre-Alains proposal we would have > > [Term] > id: MS:1001842 > name: sequence-level spectral count > def: "The number of MS2 spectra identified for a raw peptide > sequence without PTMs and charge state in spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level quantitation > > [Term] > id: MS:1002733 > name: peptide-level spectral count > def: "The number of MS2 spectra identified for a peptide sequence > specified by the amino acid one-letter codes plus optional PTMs in > spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level quantitation > > [Term] > id: MS:1002734 > name: peptide ion-level spectral count > def: "The number of MS2 spectra identified for a molecular ion > defined by the peptide sequence represented by the amino acid > one-letter codes, plus optional PTMs plus optional charged aducts > plus the charge state, in spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level quantitation > > any futher comments? > > Best regards, > Gerhard > > Am 19.01.2017 um 09:16 schrieb Binz Pierre-Alain: > > Hi all, > > Yep, we are redundantly coming to thosse definitions and > interpretations of what those peptidic things are... > > > A peptide is, according to IUPAC gold book > https://goldbook.iupac.org/P04479.html: > > > Peptides : Amides > <https://goldbook.iupac.org/A00266.html> derived from two or > more amino carboxylic acid molecules (the same or different) > by formation of a covalent bond > <https://goldbook.iupac.org/C01384.html> from the carbonyl > carbon of one to the nitrogen atom of another with formal > loss of water. The term is usually applied to structures > formed from α-amino acids, but it includes those derived > from any amino carboxylic acid. > > Therefore it includes a combination of all 20 amino acids + > all kinds of modifications > > A peptide is not a peptide ion in the sense of a MS ion as it > is usually the charge neutral representation of a combination > of a-aminoacids in the case of protein peptides > > What we represent here are either the molecular ion (peptide > sequence represented by one-letter codes + optional PTMs + > charged aduct (H+) + charge state) which we can call peptide > ion : [ANSpK +2H] 2+ (where Sp is a phosphoserine) which we > can interpret with only the sequence with PTMs ANSpK or > only with the primary sequence with no PTM ANSK > > My suggestion then is to keep the term peptide as the most > generic and does not need to specify whether it carries PTMs > or not: > > name: peptide-level spectral > count # of ANSpK > > name: sequence-level spectral count # of ANSK > > name: peptide ion-level spectral > count # of [ANSpK+2H]2+ > > Now if you feel that the term “peptide-level” will be > ambiguous (people will not know that peptide sequence and > peptide (with or without mods) are different, we can give a > clear definition with example or I could recommend using > “peptide form level” or “peptide species level” as analogy to > protein form or protein species vs protein (top-down inspired > definition), but this adds a new definition and make our > filkes larger for no good reasons. > > Pierre-Alain > > *De :*Juan Antonio Vizcaino [mailto:ju...@eb...] > *Envoyé :* mercredi 18 janvier 2017 18:23 > *À :* Eric Deutsch > *Cc :* Tobias Ternent; psi...@li... > <mailto:psi...@li...>; Robert > Chalkley; le...@im... > <mailto:le...@im...>; > jul...@ru... > <mailto:jul...@ru...>; Ghali, Fawaz; > mar...@ru... <mailto:mar...@ru...>; > mayerg97; Andy Jones; Perkins, Simon; > psi...@li... > <mailto:psi...@li...>; ju...@ed... > <mailto:ju...@ed...>; Eugen Netz; Harald Barsnes > *Objet :* Re: [Psidev-pi-dev] mzid 1.2 reviews > > Fine for me. Sorry, I did not realise about what Eric > mentioned. He is completely right. > > Cheers, > > Juan > > On 18 Jan 2017, at 17:18, Eric Deutsch > <ede...@sy... > <mailto:ede...@sy...>> wrote: > > Hi everyone, I agree that splitting this into 3 terms is > good. I disagree that we should misuse the term “peptide” > and then try to clarify it with a parenthetical > expression. I don’t think this is good controlled > vocabulary practice. I suspect the true ontology devotees > would be horrified by this practice. > > May I suggest that we use the style that I adopted for the > mzIdentML CV reorg? This keeps the term name concise, and > we should rely on the definition to remove any ambiguity. > I suggest these names instead: > > name: distinct peptide sequence-level spectral count > > name: modified peptide sequence-level spectral count > name: peptide ion-level spectral count > > What do you think? Does this capture the desired concepts > clearly? > > Regards, > > Eric > > *From:*Juan Antonio Vizcaino [mailto:ju...@eb... > <mailto:ju...@eb...>] > *Sent:*Wednesday, January 18, 2017 1:42 AM > *To:*Eric Deutsch <ede...@sy... > <mailto:ede...@sy...>> > *Cc:*mayerg97 <ger...@ru... > <mailto:ger...@ru...>>; Andy Jones > <And...@li... > <mailto:And...@li...>>;psi...@li... > <mailto:psi...@li...>; Perkins, > Simon <Sim...@li... > <mailto:Sim...@li...>>; Harald Barsnes > <Har...@ui... <mailto:Har...@ui...>>; > Yasset Perez-Riverol <yp...@eb... > <mailto:yp...@eb...>>; Tobias Ternent > <to...@eb... > <mailto:to...@eb...>>;jul...@ru... > <mailto:jul...@ru...>;mar...@ru... > <mailto:mar...@ru...>;lfi...@st... > <mailto:lfi...@st...>;ju...@ed... > <mailto:ju...@ed...>; Eugen Netz > <eug...@tu... > <mailto:eug...@tu...>>; Mathias Walzer > <wa...@in... > <mailto:wa...@in...>>; Oliver > Kohlbacher <oli...@un... > <mailto:oli...@un...>>;le...@im... > <mailto:le...@im...>; Robert Chalkley > <cha...@cg... <mailto:cha...@cg...>>; > Ghali, Fawaz <F....@li... > <mailto:F....@li...>>; Salvador Martínez de > Bartolomé <sma...@pr... > <mailto:sma...@pr...>>;psi...@li... > <mailto:psi...@li...> > *Subject:*Re: mzid 1.2 reviews > > Hi all, > > I agree with Eric on this. The more clarity the better. > What about? > > [Term] > id: MS:1001842 > name: peptide PSM count (peptide as the raw peptide sequence) > > def: "The number of MS2 spectra identified for a given > peptide (considering the raw peptide sequence, and merging > all charge states, modifications in the same peptide > sequence are not considered separately) in spectral > counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this > CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level quantitation > > [Term] > id: MS:1001xxx > name: peptide PSM count (peptide as the combination > between peptide sequence and modifications) > def: "The number of MS2 spectra identified for a given > peptide (considering the peptide sequence plus > modifications, and merging all charge states) in spectral > counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this > CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level quantitation > > [Term] > id: MS:1001xxx > name: peptide PSM count (peptide as the combination > between peptide sequence, modifications and charge state) > > def: "The number of MS2 spectra identified for a given > peptide (considering the peptide sequence and > modifications, and one single charge state) in spectral > counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for this > CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level quantitation > > Of course more variants would need to be created if we > would want to mix the definitions above with different > charge states, but I think these three definitions would > be ok. > > What happens also in the case of ambiguity in the > modification position? So, it could be even more complex. > But I think this level is probably ok for the current > state of things. > > Cheers, > > Juan > > On 17 Jan 2017, at 15:08, Eric Deutsch > <ede...@sy... > <mailto:ede...@sy...>> wrote: > > This is pretty good, but I think we should make it > completely clear whether it is the number of PSMs per > peptide sequence, per peptide with modifications, or > per peptide ion (including separate charge states). > The plain word peptide is often used as a standin for > any of these 3 concepts.. > > Thanks, > > Eric > > *From:*mayerg97 [mailto:ger...@ru... > <mailto:ger...@ru...>] > *Sent:*Tuesday, January 17, 2017 4:39 AM > *To:*Jones, Andy;psi...@li... > <mailto:psi...@li...> > *Cc:*Juan Antonio Vizcaino (ju...@eb... > <mailto:ju...@eb...>); Perkins, Simon; 'Harald > Barsnes'; 'Yasset Perez-Riverol';to...@eb... > <mailto:to...@eb...>;jul...@ru... > <mailto:jul...@ru...>;mar...@ru... > <mailto:mar...@ru...>;lfi...@st... > <mailto:lfi...@st...>;ju...@ed... > <mailto:ju...@ed...>; Eugen Netz; Mathias Walzer; > Oliver Kohlbacher;le...@im... > <mailto:le...@im...>; 'Robert Chalkley'; > Ghali, Fawaz;sma...@pr... > <mailto:sma...@pr...>; 'Eric > Deutsch;psi...@li... > <mailto:psi...@li...> > *Subject:*Re: mzid 1.2 reviews > > Hi Andy, > > yes we did a CV restructuring, > seehttps://github.com/HUPO-PSI/mzIdentML/issues/25, > but I think this problem is not related to the > restructuring. > > What about > > [Term] > id: MS:1001842 > name: peptide PSM count > def: "The number of MS2 spectra identified for this > peptide in spectral counting." [PSI:PI] > xref: value-type:xsd\:int "The allowed value-type for > this CV term." > is_a: MS:1001805 ! quantification datatype > is_a: MS:1002015 ! spectral count peptide level > quantitation > > Best regards, > Gerhard > > Am 11.01.2017 um 16:56 schrieb Jones, Andy: > > Hi all, > > (Sent to PSI-PI mailing list plus authors of the > mzid paper, in case any of you are not on the > mailing list). > > We have received 5 reviews of the mzid 1.2 > specifications via the PSI process, as well as the > 2 reviews received on the manuscript from MCP, for > which it makes sense to consider together. I think > most of these can be addressed fairly simply. I > have already fixed a few typos in the > specification document that were identified, and > made one or two other sensible clarifications to > wording. These changes have been checked into > GitHub. I have started response documents to both > processes in the relevant zip files. > > *If you would like to contribute, please could you > send me any feedback within one week (by 19^th > Jan) if possible.*After that point, I will submit > the mzid 1.2 specs back to the PSI process (for > which Sylvie the editor recommended minor > corrections), unless anything comes up that needs > extensive further discussion. Once they are > accepted, I will resubmit the paper to MCP, as > reporting the mzid 1.2 final specifications. > > Just flagging a few points for specific people to > address: > > -Robert and Alexander: one of the reviewer asks > whether your cross-linking software will export > mzid 1.2, anything you can say here? > > -Fawaz – one of the reviewers seemed to find some > validation errors with some of our mzid 1.2 files, > please could you take a look > > -Juan – one of the MCP reviewers suggests we > change the figures. Can you give me a view on > whether you think we need to do this? I’m of the > view to make minor improvements, but not radically > change them. > > -Gerhard – see response to MCP reviews. One of the > reviewers would like to know how spectral counts > can be added at the peptide-level. I think this > may need a new parent term for one given CV term > (see my notes in the response to reviews doc), > could you take a look? > > Best wishes > > Andy > > -- > > *--------------------------------------------------------------------* > > *Dipl. Inform. med., Dipl. Wirtsch. Inf. GERHARD MAYER* > > *PhD student* > > *Medizinisches Proteom-Center* > > *DEPARTMENT Medical Bioinformatics* > > *Building*ZKF E.049a|Universitätsstraße 150 | D-44801 > Bochum > > *Fon*+49 (0)234 32-21006|*Fax*+49 (0)234 32-14554 > > *E-mail*ger...@ru... <mailto:ger...@ru...> > > www.medizinisches-proteom-center.de > <http://www.medizinisches-proteom-center.de/> > > > > > > ------------------------------------------------------------------------------ > > Check out the vibrant tech community on one of the world's most > > engaging tech sites, SlashDot.org!http://sdm.link/slashdot > > > > > > _______________________________________________ > > Psidev-ms-vocab mailing list > > Psi...@li... > <mailto:Psi...@li...> > > https://lists.sourceforge.net/lists/listinfo/psidev-ms-vocab > > -- > > *--------------------------------------------------------------------* > > *Dipl. Inform. med., Dipl. Wirtsch. **Inf. GERHARD MAYER* > > *PhD student* > > *Medizinisches Proteom-Center* > > *DEPARTMENT Medical Bioinformatics* > > *Building *ZKF E.049a | Universitätsstraße 150 | D-44801 Bochum > > *Fon *+49 (0)234 32-21006 | *Fax *+49 (0)234 32-14554 > > *E-mail *ger...@ru... <mailto:ger...@ru...> > > www.medizinisches-proteom-center.de > <http://www.medizinisches-proteom-center.de/> > > > > > ------------------------------------------------------------------------------ > > Check out the vibrant tech community on one of the world's most > > engaging tech sites, SlashDot.org!http://sdm.link/slashdot > > > > > _______________________________________________ > > Psidev-ms-vocab mailing list > > Psi...@li... > <mailto:Psi...@li...> > > https://lists.sourceforge.net/lists/listinfo/psidev-ms-vocab > > -- > > *--------------------------------------------------------------------* > > *Dipl. Inform. med., Dipl. Wirtsch. **Inf. GERHARD MAYER* > > *PhD student* > > *Medizinisches Proteom-Center* > > *DEPARTMENT Medical Bioinformatics* > > *Building *ZKF E.049a | Universitätsstraße 150 | D-44801 Bochum > > *Fon *+49 (0)234 32-21006 | *Fax *+49 (0)234 32-14554 > > *E-mail *ger...@ru... <mailto:ger...@ru...> > > www.medizinisches-proteom-center.de > <http://www.medizinisches-proteom-center.de/> > > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > > > _______________________________________________ > Psidev-ms-vocab mailing list > Psi...@li... > https://lists.sourceforge.net/lists/listinfo/psidev-ms-vocab -- *--------------------------------------------------------------------* *Dipl. Inform. med., Dipl. Wirtsch. **Inf. GERHARD MAYER* *PhD student* *Medizinisches Proteom-Center* *DEPARTMENT Medical Bioinformatics* *Building *ZKF E.049a | Universitätsstraße 150 | D-44801 Bochum *Fon *+49 (0)234 32-21006 | *Fax *+49 (0)234 32-14554 *E-mail***ger...@ru... <mailto:ger...@ru...> www.medizinisches-proteom-center.de <http://www.medizinisches-proteom-center.de/> |
