[nmrCV] proposals of new terms

[DJ]

Dear nmrML developers, 

Here are some terms suggested to be added within the nmrCV:

---------- Baseline correction

=> new term: Automatic baseline recognition 

=> nmrCV parent node : baseline correction

refs: 

* Golotvin & Williams (2000), Improved baseline recognition and modeling of FT NMR spectra, J. Magn. Res. 146:122-125

* Bao et al. (2013) A new automatic baseline correction method based on iterative method, J. Magn. Res. 218:35-43

----------  Solvent suppression

=> new term: post-acquisition solvent suppression

=> nmrCV parent nodes: solvent suppression, frequency domain data processing

Note:

currently only water is concerned, but there are others solvents such as Methanol (MeOD)

In the same way, one can apply Hankel singual value decomposition (this is how I proceed with the MeOD solvent and I obtain great results :-) )

---------- NMR Standard

=> new term: Trimethylsilyl propionate (TSP)

=> nmrCV parent node: 1H spectrum reference compound

Ref: http://en.wikipedia.org/wiki/Trimethylsilyl_propanoic_acid

Concerning the term 'trimethylsilane', should it really be here? (http://en.wikipedia.org/wiki/Trimethylsilane)

---------- zero calibration

Concerning the NMR plasma samples for example, how to indicate a NMR reference compound else that those given under the node '1H spectrum reference compound' 

(i.e: glucose, peak 5.23 ppm) ?

----------

Regards,

Daniel Jacob

[Philippe]

Hi Daniel (J) / (S)

Just one comment:
For all chemical compounds, It would be good (best) practice to reuse existing resources/namespaces.
Trimethylsilyl propionate should be submitted to CHEBI and imported in NMR-ML CV

in fact, this should possibly be part of the cosmos recommendations for managing semantic resources.

Best

Philippe

--
You received this message because you are subscribed to the Google Groups "nmrML-COSMOS-FP7" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nmrml+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Reza Salek]

Need to come up with more details as well as alternative reference compounds. Information such as their reference chemical shift and multiplicity used.  We can use ChEBI but some information such as multiplicity used and chemical shift calibrated to should remain here?

Best wishes,

Reza

[Daniel JACOB]

Dear nmrML developers, 

Here is another term suggested to be added within the nmrCV:

---------- Group time delay (NMR data processing)

=> new term: group time delay compensation

=> nmrCV parent node : NMR data processing or phase correction (?)

This is independent of the bruker acquisition parameter named "GRPDLY" (Group Delay), but this is a "shift" applied on the fid to calibrate its zero; method currently implemented within a home made software, and may be, is it a similar method applied in the Mnova software ?

See:

  http://nmr-analysis.blogspot.fr/2008/02/why-arent-bruker-fids-time-corrected.html

  http://nmr-analysis.blogspot.fr/2010/05/bruker-smiles.html

----------

About its need and its usage, see below an nmrML extract (generated by the JAVA nmrML converter, dev. beta version) ;

Moreover, in this same extract, see how I tried to solve the spectral referencing issue, thus following the Philippe's advice suggesting to use the CHEBI ontology.

    <softwareList>

        <software id="ID00005" version="Version 2.1" cvRef="NMRCV" accession="NMR:1400215" name="Bruker TopSpin software"/>

        <software id="ID00103" cvRef="NMRCV" accession="NMR:1400135" name="spectrum post-processing software"/>

   </softwareList>

     ...

    <referenceableParamGroupList>

        <referenceableParamGroup id="ID00104">

            <cvParam cvRef="NMRCV" accession="NMR:1000014" name="solvent suppression"/>

            <userParam name="Method" value="range"/>

        </referenceableParamGroup>

        <referenceableParamGroup id="ID00105">

            <cvParam cvRef="NMRCV" accession="NMR:1400074" name="baseline correction"/>

            <userParam name="Method" value="Automatic baseline recognition"/>

        </referenceableParamGroup>

        <referenceableParamGroup id="ID00106">

            <cvParam cvRef="NMRCV" accession="NMR:1400XXX" name="group time delay compensation"/>

            <userParam name="group time delay" value="72" unitAccession="UO_0000189" unitName="count unit" unitCvRef="UO"/>

        </referenceableParamGroup>

        <referenceableParamGroup id="ID00107">

            <cvParam cvRef="NMRCV" accession="NMR:1000075" name="spectral referencing"/>

            <userParam name="ppm" value="5.23" unitAccession="UO_0000169" unitName="parts per million" unitCvRef="UO"/>

            <userParam name="Reference Compound" value="" unitAccession="CHEBI:17234" unitName="glucose" unitCvRef="CHEBI"/>

        </referenceableParamGroup>

   </referenceableParamGroupList>

     ...

     <dataProcessingList>

        <dataProcessing id="ID00108">

            <processingMethod order="1" softwareRef="ID00103">

                <referenceableParamGroupRef ref="ID00104"/>

                <referenceableParamGroupRef ref="ID00105"/>

                <referenceableParamGroupRef ref="ID00106"/>

                <referenceableParamGroupRef ref="ID00107"/>

                <cvParam cvRef="NMRCV" accession="NMR:1000071" name="phase correction"/>

                <cvParam cvRef="NMRCV" accession="NMR:1400044" name="fourier transformation"/>

            </processingMethod>

        </dataProcessing>

   </dataProcessingList>

     ...

    <spectrum defaultDataProcessingRef="ID00108">

        <spectrum1D>

            <processingSoftwareRefList>

                <softwareRef ref="ID00103"/>

            </processingSoftwareRefList>

        ...

        </spectrum1D>

   </spectrum>

Regards,

Daniel Jacob

[DSchober]

All new requested terms are no in the CV.
I have not tackled:

Concerning the term 'trimethylsilane', should it really be here? (http://en.wikipedia.org/wiki/Trimethylsilane)

---------- zero calibration

Concerning the NMR plasma samples for example, how to indicate a NMR reference compound else that those given under the node '1H spectrum reference compound'  (i.e: glucose, peak 5.23 ppm) ?

Should I create a term "zero calibration" as a placeholder e.g. under parameters ?

Best regards,
   Daniel Schober.

[Daniel JACOB]

---------- Forwarded message ----------
From: Daniel JACOB <djac...@gmail.com>
Date: 2015-01-12 14:46 GMT+01:00
Subject: Re: [nmrml:112] Re: [nmrCV] proposals of new terms
To: DSchober <daniels...@gmail.com>

Hi Daniel,

Finaly, I think that the term "NMR:1000026, 1H spectrum reference compound" will do nicely.

In fact, it is not only the "zero calibration" that we need to define, but the calibration of the ppm scale  (including its zero).

So, following the Philippe's advice suggesting to use the CHEBI ontology, I propose to define the spectral calibration parameters, based on a reference compound, by embedding all parameters within a "referenceableParamGroup" section.

See below an nmrML extract (generated by the JAVA nmrML converter, dev. beta version):

    <softwareList>

        <software id="ID00005" version="Version 2.1" cvRef="NMRCV" accession="NMR:1400215" name="Bruker TopSpin software"/>
        <software id="ID00103" cvRef="NMRCV" accession="NMR:1400135" name="spectrum post-processing software"/>
   </softwareList>
     ...
    <referenceableParamGroupList>
        <referenceableParamGroup id="ID00104">
            <cvParam cvRef="NMRCV" accession="NMR:1000014" name="solvent suppression"/>
            <userParam name="Method" value="range"/>
        </referenceableParamGroup>
        <referenceableParamGroup id="ID00105">
            <cvParam cvRef="NMRCV" accession="NMR:1400074" name="baseline correction"/>
            <userParam name="Method" value="Automatic baseline recognition"/>
        </referenceableParamGroup>
        <referenceableParamGroup id="ID00106">

            <cvParam cvRef="NMRCV" accession="NMR:1000347" name="group time delay"/>

            <userParam name="group time delay" value="72" unitAccession="UO_0000189" unitName="count unit" unitCvRef="UO"/>
        </referenceableParamGroup>
       

<referenceableParamGroup id="ID00107">
            <cvParam cvRef="NMRCV" accession=NMR:1000026" name="1H spectrum reference compound"/>

Regards,

Daniel J.

[Philippe]

Dear All,

I was wondering if there had been a recent commit/update to the NMR-CV.
The last one seems to be 2 month old and it seems that it is still not importing chemical compounds from CHEBI, but creating new terms in the NMRCV namespace.
As Daniel Jacob also suggeted submitting terms to CHEBI in his last post, I thought we should push changes. I have documented how to do so in a pull request
https://github.com/nmrML/nmrML/pull/92

In the following ticket https://github.com/nmrML/nmrML/issues/84, Daniel Schober refers to delegating reference to CHEBI in the rule set but where is this available from?

In the meantime, I have carried out a review of the current version of the NMRCV available from github and have identified a number of issues that should be addressed in produce a production release of the terminology.

We can discuss this on the call on wednesday

Thank you for your time.

All the best

Philippe