Metadata errors with new RawFileReader MSConvert

[FarmGeek4Life]

Metadata errors with new RawFileReader MSConvert

[alchemistmatt]

For reference, the enum values for ActivationType in ThermoFisher.CommonCore.Data.FilterEnums are:

0: CollisionInducedDissociation
1: MultiPhotonDissociation
2: ElectronCaptureDissociation
3: PQD
4: ElectronTransferDissociation
5: HigherEnergyCollisionalDissociation (hcd)
6: Any
7: SAactivation
8: ProtonTransferReaction
9: NegativeElectronTransferDissociation
10: NegativeProtonTransferReaction
11: UltraVioletPhotoDissociation

This enum is returned by the call to GetActivation as demonstrated by this code snippet

var scanFilter = mXRawFile.GetFilterForScanNumber(scan);
var reactions = scanFilter.MassCount;

var index = reactions - 1;
if (index > 0 && scanFilter.GetIsMultipleActivation(index))
{
    // The last activation is part of a ETciD/EThcD pair
    index--;
}

var activationTypeCode = scanFilter.GetActivation(index);

This snippet is from
https://github.com/PNNL-Comp-Mass-Spec/Thermo-Raw-File-Reader/blob/ThermoNuGetDLL/XRawFileIO.cs#L1130

[chambm]

Thanks for the report. I can definitely see your #2 happening because it happened with a few of the other enums when I started trying to cast directly (with MSFR all those are parsed from strings). It slipped by because apparently none of the pwiz or Skyline tests use HCD! Maybe I can figure out a way to adapt the ScanFilterTest tests to also test RawFileReader.

I cannot reproduce your #1. Can you provide a test file?

[FarmGeek4Life]

You may be able to create a scan filter from a string with RawFileReader.
I think ThermoFisher.CommonCore.RawFileReader.FilterParserFactory.CreateFilterParser().GetFilterFromString(string) would do it.

[FarmGeek4Life]

I also noticed that the "GetAverageMassList" method is now commented out, I'm assuming because it isn't used, but earlier this week I figured out how to get similar results from RawFileReader. I say "similar" because the method in MSFileReader is susceptible to the improper centroiding of FTMS data that would also occur with "GetMassList"

[FarmGeek4Life]

Also, we just ran a test with a Lumos file with EThcD spectra; it crashed. The main error was "precursor count does not match ms level[SpectrumWorkerThreads::work] error in thread". I'm assuming that, if the creating a IScanFilter object using the GetFilterFromString works, this error will probably be fixed in the process of getting the ScanFilterTests to all pass.

[FarmGeek4Life]

Also, if you call IScanFilter.ToString(), you get the scan filter string you would expect to see from MSFileReader.

[alchemistmatt]

For issue 1 (MS1 data with precursor info), it appears to happen with any MS1 file from our Q-Exactive HF instrument with scan filter FTMS + p NSI Full ms.
I'll e-mail Matt Chambers a link to our FTP site with a test .raw file (I presume 400 MB is too big to attach to an issue). For reference, this is the example erroneous output:

<spectrum index="68" id="controllerType=0 controllerNumber=1 scan=69" defaultArrayLength="3681">
  <cvParam cvRef="MS" accession="MS:1000579" name="MS1 spectrum" value=""/>
  <cvParam cvRef="MS" accession="MS:1000511" name="ms level" value="1"/>

  <precursorList count="1">
    <precursor>
      <isolationWindow>
        <cvParam cvRef="MS" accession="MS:1000827" name="isolation window target m/z" value="0.0" unitCvRef="MS" unitAccession="MS:1000040" unitName="m/z"/>
      </isolationWindow>
      <selectedIonList count="1">
        <selectedIon>
          <cvParam cvRef="MS" accession="MS:1000744" name="selected ion m/z" value="0.0" unitCvRef="MS" unitAccession="MS:1000040" unitName="m/z"/>
        </selectedIon>
      </selectedIonList>
      <activation>
        <cvParam cvRef="MS" accession="MS:1000133" name="collision-induced dissociation" value=""/>
        <cvParam cvRef="MS" accession="MS:1000045" name="collision energy" value="0.0" unitCvRef="UO" unitAccession="UO:0000266" unitName="electronvolt"/>
      </activation>
    </precursor>
  </precursorList>

[chambm]

Seems like a bug in RawFileReader. I'll report it to Jim.

[chambm]

That was a pretty severe bug. No idea how widespread that was. Fixed in #194

[chambm]

The activation type issue should be fixed in #204.

[chambm]

(if someone finds other RFR issues it can be reopened)