Re: [Psidev-ms-dev] MS3+ spectra

[Chambers, M. <mat...@gm...>]

Ten years later, I'm faced with a Thermo RAW from a targeted method that only has MS3 spectra. And each MS3 is actually from "SPS" mode so 
it has multiple isolation windows contributing product ions (SPS mode has been available since the Orbitrap Velos). I need to be able to 
store both the hierarchy (because there is no precursor scan to reference) and the multiple isolation windows.

Fredrik's suggestion of using a cvParam in the <precursor> to indicate its order seems reasonable, but would require either a mapping file 
change or making the cvParam *is_a *'ion selection attribute' or 'isolation window attribute'. We could either make a new CV term or re-use 
'ms level': give it another *is_a* relationship to 'ion selection' or 'isolation window'.

Thoughts?
-Matt


On 9/22/2009 3:14 PM, Matthew Chambers wrote:
> A product ion spectrum that was produced from multiple isolation windows
> would need multiple precursors. I don't know if there's an instrument
> that can or does do this, but it's easy to imagine an instrument which
> can isolate each peak of an isotope envelope instead of isolating a wide
> window around all the isotopes (and all the noise in between).
>
> -Matt
>
>
> Fredrik Levander wrote:
>> Hi Matt,
>>
>> Are there any other examples of spectra which have more than one
>> precursor than the MS3(or higher) spectra? I think the general MS2 case
>> would be one precursor but possibly several selected ions. The only
>> situation I can  think of where there might be several possible
>> precursors selected would be some kind of merged spectrum, but that
>> could probably also in most cases be written as having one precursor
>> with several selected ions (if not the activation conditions were
>> different from each other for the original MS2 precursors).
>> A CV term that clarifies the order of a certain precursor in the
>> fragmentation hierarchy seems like a reasonable solution to clarify this
>> if needed, anyway. I guess it is not really correct in xml to just use
>> the order of apperance.
>>
>> -Fredrik
>>
>> Matthew Chambers skrev:
>>    
>>> Hi all,
>>>
>>> As far as I can tell, we don't have documentation or an official example
>>> of how to handle this case. Fredrik made an example for this and sent
>>> this example to the list on 12/28/2008:
>>> http://dev.thep.lu.se/fp6-prodac/svn/trunk/mzML/scans/5.ms3_scan.txt
>>>
>>> I'm concerned about the use of multiple precursors to represent the
>>> precursor "history." This seems quite confusing. Most of the time I
>>> would expect to see one precursor for the MS3 and if I wanted to know
>>> what the original MS precursor was, I'd look at the MS2's precursor.
>>> However, I can see that a method with a targeted MS3 (i.e. precursor and
>>> 1st generation product masses already known) might not even record the
>>> intermediate MS2 spectrum. So specifying the precursor hierarchy is
>>> sensible, but IMO we need to do it in a distinct way from the way we
>>> list multiple precursors that actually were fragmented to create the
>>> current scan. A clear solution to me seems to be to allow precursor to
>>> have a precursorList child element to explicitly indicate that the
>>> nested precursorList represents a previous generation of fragmentation.
>>> This is quite messy implementation-wise though, so another option is to
>>> use a cvParam in the precursorList to explicitly indicate what it means
>>> to have multiple precursors. This would mean that you couldn't mix
>>> multiple isolation windows with MS3+ though.
>>>
>>> Thoughts?
>>> -Matt
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry&reg; Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9&#45;12, 2009. Register now&#33;
> http://p.sf.net/sfu/devconf
> _______________________________________________
> Psidev-ms-dev mailing list
> Psi...@li...
> https://lists.sourceforge.net/lists/listinfo/psidev-ms-dev
>