QTOF Raw to mzXML failure

[Hong]

Hi there,

I am new to TPP. I have trouble converting the MassLynx .RAW directory
with mzXML. I got the following status report message. What do I need
to change to get it work? I greatly appreciate your inputs.

Hong Li
NJ Med School

massWolf -v --mzXML c:/Inetpub/wwwroot/ISB/data/hilo08.raw

Settings:
centroid scans: 0
compress scans: 0
verbose mode: 1
lockspray: 1
MS^E: 0
threshold: 0
mzML mode: 0
mzXML mode: 1
input filename: c:/Inetpub/wwwroot/ISB/data/hilo08.raw
output filename: c:/Inetpub/wwwroot/ISB/data/hilo08.mzXML
output file as gzip: 0


Command FAILED

[Natalie Tasman]

Hello,

First, you might want to check that the file extention is indeed .raw
and not .RAW. Capitalization might make a difference.

Can you open this file in the Waters MassLynx application on the same
computer? This must be installed in order for the converter
(massWolf) to work properly.

Natalie

[Bernt]

Hi Natalie,

I'm also currently trying out MassWolf 4.2.1......it works with TD and
TZ as well, although TD appears to be a better option at generating
smaller mzxml files. But the centroiding function doesn't seem to be
working. It gives a "Warning: using EXPERIMENTAL centroiding
functionality" error msg in command prompt and Windows pops up a
"masswolf.exe has encountered a problem and needs to close" error
report. Is the centroiding function still in the process of
development?


Bernard


On Jun 25, 5:46 am, Natalie Tasman <natalie.tas...@insilicos.com>
wrote:

[Matthew Chambers]

Hi Bernard,

What is TD and TZ? MassLynx doesn't provide centroiding capability so
MassWolf has a custom centroiding algorithm. I haven't seen any
comprehensive comparisons between that custom algorithm and the MaxEnt
algorithm available from ProteinLynx Global Server (which is not
programmatically accessible even if you have access to PLGS!).

-Matt

[Dave Trudgian]

Matt,

TD filters out data points less than a given intensity, discarding them.
TZ sets them to zero. You can produce centroid spectra using the
Accurate Mass Measure function in the tools tab within Masslynx, which
can batch process raw files.

DT

--
Dr. David Trudgian
Bioinformatician in Proteomics
University of Oxford

Tel: (+44) (01865 2)87807 (CCMP - Mon-Thu)
Tel: (+44) (01865 2)75557 (Dunn Sch - Fri)

[liho2]

Hi Natalie,

The file name is nj_052108_mESC_ForBSI_Pro.raw. Do those underscores
affect the file reading?

I have MassLynx 4.0 sp2 loaded on my desk top and I am able to open
the file and see the spectra and chromatogram.

Thanks,

Hong

On Jun 24, 5:46 pm, Natalie Tasman <natalie.tas...@insilicos.com>
wrote:

[liho2]

Hi Natalie,

One thing I noticed is that under "Please choose file(s)" menu, all
the .raw directory file size is recognized as 0 k. On the other hand,
the Orbitrap .RAW file that I downloaded from MaxQuant site can be
see with the right size.

Hong


On Jun 24, 5:46 pm, Natalie Tasman <natalie.tas...@insilicos.com>
wrote:

[Natalie Tasman]

Hi Bernt,

The massWolf centroiding algorithm is indeed experimental, and I'm not
surprised that it doesn't work. In future versions of the TPP, this
program will be retired in favor of the ProteoWizard's msconvert tool,
which I believe may have centroiding options. Between that and Dave's
suggestion in this thread, hopefully you can find a way to move forward.

Best,

Natalie

[Natalie Tasman]

I wonder if there might be an issue with that file. I think that
others have suggested that sometimes saving a corrupted file as a
different name might help? I would suggest checking with Waters if
that "0 k" size does indeed represent a problem.

In the meanwhile, can you try another one of your own files?

However, I just noticed-- are you actually working with Thermo files?
The Thermo files (like the Orbitrap one that you downloaded) are a
completely different format that Waters. Thermo files should be
converted with the "readw" converter, or, you can try the ProteoWizard
project's msconvert program (included in the TPP or a possibly newer
version from their website).

Natalie

[Natalie Tasman]

Thanks, Dave, and just pointing out that there is a description of
these options in the usage statement (printed out when running the
program with no arguments).

Natalie

[Chris Hughes]

The file is Micromass data. It is not odd for the size to read 0K,
this is some sort of artifact of it being a .raw folder. The file is
not corrupt, it opens on MassLynx properly, and can be run through
PLGS with no troubles to a pkl file. It was acquired in MassLynx 4.1
however, might this make a difference as the user is using an older
version?



On Jun 25, 2:36 pm, Natalie Tasman <natalie.tas...@insilicos.com>
wrote:

> Thanks, Dave, and just pointing out that there is a description of  
> these options in the usage statement (printed out when running the  
> program with no arguments).
>
> Natalie
>
> On Jun 25, 2009, at 9:36 AM, Dave Trudgian wrote:
>
>
>
> > Matt,
>
> > TD filters out data points less than a given intensity, discarding  
> > them.

> > TZ sets them to zero. You can producecentroidspectra using the

> > Accurate Mass Measure function in the tools tab within Masslynx, which
> > can batch process raw files.
>
> > DT
>
> > Matthew Chambers wrote:
> >> Hi Bernard,
>
> >> What is TD and TZ? MassLynx doesn't provide centroiding capability so

> >>MassWolfhas a custom centroiding algorithm. I haven't seen any

> >> comprehensive comparisons between that custom algorithm and the  
> >> MaxEnt
> >> algorithm available from ProteinLynx Global Server (which is not
> >> programmatically accessible even if you have access to PLGS!).
>
> >> -Matt
>
> >> Bernt wrote:
>
> >>> Hi Natalie,
>

> >>> I'm also currently trying outMassWolf4.2.1......it works with TD  

> >>>>>centroidscans: 0

[Natalie Tasman]

Hi Chris,

Yes, especially for MassLynx, it can make a big difference if the versions don't match.  You really want to have the MassLynx version on the conversion machine exactly match the acquisition version.  It's good news that the file can be opened by MassLynx itself and thus doesn't seem corrupted.  I am curious, though, if you did that test on the conversion or acquisition machine (that is, which vendor version you tested against?).

Natalie

[Chris Hughes]

I typically acquire all my files on a machine that does only that. The
raw data is moved off right away to an external computer which is much
faster and is used specifically for processing data. They both have
the same version of MassLynx 4.1 installed. I have successfully
converted the file which Dr. Li is referring too. My best guess is
that it is not working on his machine due to the non-matching version
of MassLynx.


On Jul 15, 4:05 pm, Natalie Tasman <natalie.tas...@insilicos.com>
wrote:

> Hi Chris,
>
> Yes, especially for MassLynx, it can make a big difference if the versions
> don't match.  You really want to have the MassLynx version on the conversion
> machine exactly match the acquisition version.  It's good news that the file
> can be opened by MassLynx itself and thus doesn't seem corrupted.  I am
> curious, though, if you did that test on the conversion or acquisition
> machine (that is, which vendor version you tested against?).
>
> Natalie
>