Trouble converting Waters.RAW with TPP msconvert

[Alex Gao]

Hi guys, 

So I've used TPP before to convert bottom up proteomic Thermo raw files in msconvert, comet search with custom params file, and xinteract to perform protein prophet with no issues. Earlier this year, I performed the same with Waters. Raw file, although in the beginning there was an issue with size, shrinking the size of my raw file from 25 to 5-6GB did the trick. 

However, recently, I reinstalled TPP (v6.3.3), and was trying to perform the same task to convert 5-6GB of Waters.raw but failed. msconvert didn't fail, but comet search was finished in less than 1min, which is not quite right. And out of the 200,000 spectra in the original file, only 1000 ish was used for comet search. this obviously leads to an error code of 256 at the xinteract level. 

I tried several different troubleshoots, not changing the default params file except for location of fasta file, uninstalling and reinstalling, changing a computer, renaming the FUNC0003 files from my Water.raw directory for the search, etc. etc., but none work. 

Any help would be much appreciated. Thank you!

Best,

Alex

[David Shteynberg]

Hello Alex,

Thanks for trying the TPP pipeline to solve your proteomics computational needs.  Sorry, you are having trouble converting and analyzing Waters raw data.  Have you tried using the latest version of msconvert to do the conversion to mzML?  If you are able to upload your file to an online drive and give me access to it, I can pull it down and try to convert it myself.  Let me know what works for you.

Cheers!

-David

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/84cf1e94-32b3-47f6-80a8-37150ff94d93n%40googlegroups.com.

[Alex Gao]

Hi David,

Thank you for your prompt feedback! The file can be found at the link below. 

https://drive.google.com/file/d/1QulcwZ3UP7pcMtECNwubd4DysqaDq4UN/view?usp=sharing

Looking forward to your inputs!

Best,

Alex

[David Shteynberg]

Hello Alex,

I was able to download your file, convert the data to mzML using msconvert and read a few spectra using the command line tool readMzXml (works on mzML too).  Unfortunately, I am finding many intensities these files that are encoded with negative values, which is likely causing problems for the downstream tools like comet.  Do you know why negative intensities are in these files?  It is possible this is a bug with the raw encoding.  Are you able to visualize these files using Waters software to check the intensities and compare to those output by msconvert?

Thanks,

-David

To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/d774140c-2641-48f2-b9c7-900442b10503n%40googlegroups.com.

[Alex Gao]

Hi David, 

I do not know why. The data was acquired on MassLynx by a Waters Xevo QTOF set to fast DDA, with scan times of MS1 and MS2 set to both 1s with 10 MS2 scans. The lockspray mass is also included. The settings, besides the scan times, were not changed from the last time I ran the data, which worked the last time. On MassLynx, the intensities look normal as well, with a dome of increasing intensities over the gradient, typical of bottom up proteomics. Is there a way to solve this? 

[David Shteynberg]

Hello Alex,

This is pointing to a bug in either the Waters library for access to the raw data or in msconvert. Perhaps you can contact Waters and msconvert developers to see if they have patched the problem.  My suspicion as to the nature of the problem is use of signed (as opposed to unsigned) variables internally for the representation of intensities.  Once the intensities get large enough they roll over to the negative representations of the values in C++.  Let us know when you find a solution to this issue!

Thanks,

-David

To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/6d911bb1-1bc9-4135-aafa-b168657ad219n%40googlegroups.com.

[Alex Gao]

Dear David, 

So after a long troubleshooting process with msConvert and Waters team as well as testing it by myself, I found the best way to convert Waters.raw files is to use this Waters proprietary software  https://microapps.on-demand.waters.com/home/showmarkdown/data-as-a-product to convert Waters file to mzML (convert to 32 bit). Alternatively, one could convert the file with msConvert GUI (the actual software, not the one embedded in TPP), with only the settings "write index" and "TPP compatible" checked, convert to mzML 32-bit (mzXML 32-bit for Maxquant compatibility), and most importantly, exclude the negative values (which are EMR spectra from UV or PDA detector inline with the system) by adding a subset filter for MS-Level "1-", which allows the mz(X)ML to keep only MS1 and MS2 data. Thanks for your help!

Best,

Alex

[Luis Mendoza]

Hello Alex,

Glad that you were able to get to the bottom of this issue and hope you have been able to process your data.

I  just wanted to add that when installing TPP on Windows, the entire ProteoWizard set of tools (including msConvert GUI) is also bundled -- so there is typically no need to re-download/install unless you want a newer version.  You can find the software under  C:/Program Files/ProteoWizard/  on standard installs.  Note that one is able to install multiple versions side-by-side, and you may even notice multiple ones there already if you have updated TPP on that machine before.

Cheers,

--Luis

To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/a4ed1d74-04a0-468d-b8a4-deae9bfc5e69n%40googlegroups.com.