Issue with Recognizing User Created Modification

support
Issue with Recognizing User Created Modification dwwilk02  2016-05-27 11:40
 
Hi,

We were trying to get Skyline-Daily (on Windows 64-bit) to create a library from a final_fragment.csv file from a Waters Synapt G2-Si MSe experiment (csv file is attached). The file has the modification GEE(Q) (an addition of H6C4O2 to Q residues) that was created in-house for a collaborator. We get the following error when loading csv files with this mod:

System.IO.IOException: ERROR: The modification 'GEE (Q)' on line 489 is not recognized.
ERROR:
ERROR: reading file GEE_Test_Mouapi_Pink01_IA_final_fragment.csv

   at pwiz.Common.SystemUtil.ProcessRunner.Run(ProcessStartInfo psi, String stdin, IProgressMonitor progress, IProgressStatus& status, TextWriter writer) in c:\proj\pwiz_x64\pwiz_tools\Shared\Common\SystemUtil\ProcessRunner.cs:line 109
   at pwiz.Common.SystemUtil.ProcessRunner.Run(ProcessStartInfo psi, String stdin, IProgressMonitor progress, IProgressStatus& status) in c:\proj\pwiz_x64\pwiz_tools\Shared\Common\SystemUtil\ProcessRunner.cs:line 53
   at pwiz.BiblioSpec.BlibBuild.BuildLibrary(LibraryBuildAction libraryBuildAction, IProgressMonitor progressMonitor, IProgressStatus& status, String[]& ambiguous) in c:\proj\pwiz_x64\pwiz_tools\Shared\BiblioSpec\BlibBuild.cs:line 171
   at pwiz.Skyline.Model.Lib.BiblioSpecLiteBuilder.BuildLibrary(IProgressMonitor progress) in c:\proj\pwiz_x64\pwiz_tools\Skyline\Model\Lib\BiblioSpecLiteBuilder.cs:line 130

I've tried several things to get the mod to work (adding the mod to Skyline, changing a current mod with H6C4O2 addition to include Q, adjusting the addition for one of the default mods at Q), but so far have had no luck. I can load the file if I delete the GEE mods, but then I miss the library matches for modified peptides. Is there anything that we could try to get Skyline to recognize the mod? Thanks for any help you can provide,

Daniel
 
 
Kaipo Tamura responded:  2016-05-27 11:58
Hi Daniel,
The library builder needs to know the mass of the modification - do you know if that information is accessible from anywhere in the .csv file?
For now, it may be easiest to construct a tab-delimited .ssl file with the two PSMs containing the modification, see: https://skyline.gs.washington.edu/labkey/wiki/home/software/BiblioSpec/page.view?name=BiblioSpec%20input%20and%20output%20file%20formats
You can then use the .ssl file in building a library.

Thanks,
Kaipo
 
dwwilk02 responded:  2016-05-27 12:27
Thanks Kaipo. I’ll try to use an .ssl file for the library. The mass of the modification doesn’t seem to be in the csv file. We submitted it to Unimod a few weeks ago; we were hoping that it would show up in the list of Skyline mods. Thanks,
 
Daniel
 
Brendan MacLean responded:  2016-06-09 10:12
Hi Daniel,
Unfortunately, we don't update from Unimod dynamically, but maybe we can pull an update soon. In the meantime, it is not necessary to rely on the Skyline auto-detection of Unimod modifications to get the modifications in your library matched. You can always pre-define your modifications of interest through the Peptide Settings - Modifications tab, and then Skyline will match to the definition before even looking at Unimod modifications.

Hopefully this is enough to get you going. You may also be interested in Skyline Tutorial Webinar #10: Working with Modifications in Skyline:

https://skyline.gs.washington.edu/labkey/webinar10.url

Good luck with your library building.

--Brendan
 
trinidad martincampos responded:  2019-03-19 06:20
Hi Brendan!!

I am also having some problems to create the library from PLGS fragments.csv. After trying to add my own modifications I decide to rename the file with the modification which are already in Skyline. My data is SILAC 3-plex analyzed in a DIA mode. The error is the following:

System.IO.IOException: ERROR: The modification 'Label:13C(6)15N(2) (C-term K)' on line 65 is not recognized.
ERROR:
ERROR: reading file 293T_SILAC_1to1_rep01_IA_final_fragment.csv

   at pwiz.Common.SystemUtil.ProcessRunner.Run(ProcessStartInfo psi, String stdin, IProgressMonitor progress, IProgressStatus& status, TextWriter writer) in C:\proj\skyline_4_2_x64\pwiz_tools\Shared\Common\SystemUtil\ProcessRunner.cs:line 59
   at pwiz.BiblioSpec.BlibBuild.BuildLibrary(LibraryBuildAction libraryBuildAction, IProgressMonitor progressMonitor, IProgressStatus& status, String[]& ambiguous) in C:\proj\skyline_4_2_x64\pwiz_tools\Shared\BiblioSpec\BlibBuild.cs:line 171
   at pwiz.Skyline.Model.Lib.BiblioSpecLiteBuilder.BuildLibrary(IProgressMonitor progress) in C:\proj\skyline_4_2_x64\pwiz_tools\Skyline\Model\Lib\BiblioSpecLiteBuilder.cs:line 137

Any idea what am I doing wrong??

Thanks a lot!!
Trini
 
trinidad martincampos responded:  2019-03-19 06:25
Hi Brendan!!

I am also having some problems to create the library from PLGS fragments.csv. After trying to add my own modifications I decide to rename the file with the modification which are already in Skyline. My data is SILAC 3-plex analyzed in a DIA mode. The error is the following:

System.IO.IOException: ERROR: The modification 'Label:13C(6)15N(2) (C-term K)' on line 65 is not recognized.
ERROR:
ERROR: reading file 293T_SILAC_1to1_rep01_IA_final_fragment.csv

   at pwiz.Common.SystemUtil.ProcessRunner.Run(ProcessStartInfo psi, String stdin, IProgressMonitor progress, IProgressStatus& status, TextWriter writer) in C:\proj\skyline_4_2_x64\pwiz_tools\Shared\Common\SystemUtil\ProcessRunner.cs:line 59
   at pwiz.BiblioSpec.BlibBuild.BuildLibrary(LibraryBuildAction libraryBuildAction, IProgressMonitor progressMonitor, IProgressStatus& status, String[]& ambiguous) in C:\proj\skyline_4_2_x64\pwiz_tools\Shared\BiblioSpec\BlibBuild.cs:line 171
   at pwiz.Skyline.Model.Lib.BiblioSpecLiteBuilder.BuildLibrary(IProgressMonitor progress) in C:\proj\skyline_4_2_x64\pwiz_tools\Skyline\Model\Lib\BiblioSpecLiteBuilder.cs:line 137

Any idea what am I doing wrong??

Thanks a lot!!
Trini
 
Nick Shulman responded:  2019-03-19 08:51
I cannot get the exact error to happen that you are seeing, but you should try removing a lot of the columns in your transition list.

If you use the menu item in Skyline:
Edit > Insert > Transition List
you can see the set of columns that Skyline looks for when importing a transition list: Peptide, Precursor m/z, Product m/z, Protein name and Protein Description.

In your transition list, it looks like you are trying specify modifications in a separate column. The only syntax that Skyline understands for modified peptide sequences is to put the modification in square brackets after the amino acid which is being modified:
FQSSAVMALQEASEAYLVGLFEDTNLC[+57.02]AIHAK[+8.01]

Skyline tries to guess which of your columns is the precursor m/z and product m/z. Skyline tries various combinations of the modifications that you have specified in "Settings > Peptide Settings > Modifications". Once Skyline thinks it knows which is the precursor m/z column and the product m/z column, based on looking at the first row in your transition list, Skyline then thinks those columns will hold true for the rest of the list.

In your case, you have two separate columns which both contain the precursor m/z of your peptide. Because of this, Skyline decides that one of those columns is your precursor m/z and one of them is the product m/z, and that the only transition you want is the intact peptide.

Hope this helps,
-- Nick
 
trinidad martincampos responded:  2019-03-19 10:03
Hi Nick!!

Thanks for your fast answer!
Well, I am trying to build a library, does it need the same parameters than a transition list?? And Skyline supports PLGS output right?

The problem is, if I keep only the columns you suggested I get consecutive errors asking for the rest of the columns... And I dont have any columns with the modify sequence, only the one called peptide.modification where it is written for example CarbamidomethylC(27) and gives no error and then CarbamidomethylC(27); Label:13C(6)15N(2) (C-term K) which is the one giving the error...

Thanks a lot again ;)
T
 
Nick Shulman responded:  2019-03-19 13:00
Trini,

Oh, I did not realize you were trying to build a spectral library. I thought you were trying to import a transition list.

Yes, Skyline supports "final_fragment.csv" files. That's what you have, right?

Can you send me your final_fragment.csv file? I will see what is going wrong.
If that file is less than 50MB, then you can attach it to this support request.
Otherwise, you can upload it here:
https://skyline.ms/files.url

-- Nick
 
trinidad martincampos responded:  2019-03-20 06:24
Hi Nick!

No worries at all ;)

I already uploaded the final_fragments with the name "final_fragments_forSkylineLib"

Thanks a lot for your help!!
Trini
 
Nick Shulman responded:  2019-03-20 12:07
The code that we wrote for reading Waters MSe search results has a list of modifications that it knows about.
The list is hard-coded here:
https://github.com/ProteoWizard/pwiz/blob/master/pwiz_tools/BiblioSpec/src/WatersMseReader.cpp#L59

It looks like your modifications have slightly different names than what BiblioSpec is expecting.
That is, BiblioSpec expects something that starts with:
SILAC 13C(6) 15N(2)
but the files that you sent me had either:
Label:13C(6)15N(2) (C-term K)
or
SILAC+13C(6)+15N(2)+(K)
BilbioSpec only cares that your modification starts with what BiblioSpec is looking for, so the "(K)" or "(C-term K)" is not a problem in terms of whether BiblioSpec understands the modification.
However, in the first case you have "Label " instead of "SILAC ", and in the second, all of the spaces were replaced by plus signs.

If you have a new modification that you would like to have added to BiblioSpec I think we can add those. But, for this one, it looks like the easiest thing for you to do use the exact same name that BiblioSpec is expecting.
-- Nick
 
trinidad martincampos responded:  2019-03-21 01:55
Hi again Nick! And thanks again ;)

Yes, I had already tried to change it in many different ways, including SILAC 13C(6) 15N(2) (K)(32) and it doesn't complain anymore at the same line but it does here:

"Cannot modify sequence AGKR, length 4, at position 8."

This happens in the line 119508 with modification = SILAC 13C(6) 15N(2) (K)(3);SILAC 13C(8) 15N(2) (R)(4)
I added a new modification in Skyline "Label: 13C(8) 15N(2) (C-term R)" by clicking 13C and 15N but the Monoisotopic mass = 10.008269 which is the one for 13C(6) 15N(4) right?? SO maybe here is the problem...

I have checked the list of modifications you have in github and I've found "mods_["SILAC 13C(8) 15N(2)"] = 10.020909;" the one I guess I need but how can I create that specific mass??

Thank you so much!!
T
 
Nick Shulman responded:  2019-03-21 08:41
That error message that you are seeing:
"Cannot modify sequence AGKR, length 4, at position 8."
sounds like it's saying that you have specified that the peptide "AGKR" has a modification at position 8. That peptide sequence is only 4 amino acids long, so there is no position 8.

In the source code that I pointed you at, you were only supposed to look at the names of the modifications. That code is a mapping between what modification name BiblioSpec will see in your final_fragment.csv, and what mass BiblioSpec will assign to it.
So, if BiblioSpec sees a modification that starts with:
SILAC 13C(8) 15N(2)
then BiblioSpec knows that its mass is 10.020909

-- Nick
 
trinidad martincampos responded:  2019-03-22 07:06

Hi Nick!

Yes sure, I understand that it complains if I have a 4 amino acids peptide and I want to modify the position 8... The thing here is that my peptide is modified at positions 3 and 4 as you can see in the modifications column: "SILAC 13C(6) 15N(2) (K)(3);SILAC 13C(8) 15N(2) (R)(4)", so the K at pos 3 and the R at position 4.

Somehow it is taking the SILAC heavy tag 13C(8) as the modification position which is not correct. It should take 3 and 4 right? So maybe I have to rewrite the modifications in a way Skyline can understand that?

I am very sorry for keeping asking, maybe I have already a mess in my mind and I am not getting the point, hahah.

Thanks a lot again!
Trini

 
Nick Shulman responded:  2019-03-22 11:08
Can you send me the final_fragment.csv file that we're talking about? I will try to figure out what is going wrong.
-- Nick
 
trinidad martincampos responded:  2019-03-24 09:23
Hi Nick!!

It is the same file I already uploaded like 2 days ago using the link for big files you sent me ;)

Thanks!!
Trini
 
trinidad martincampos responded:  2019-03-27 09:46
Hi again Nick!

Were you able to find the file back or you prefer me to upload it again??

Thanks!
Trini
 
Nick Shulman responded:  2019-03-27 15:42
Trini,

I don't think you ever gave me the file that you're encountering this error in.
The one that I found up there is named "293T_SILAC_1to1_rep01_IA_final_fragment_withSILAC.csv".

That one gives me the error:
ERROR: The modification 'SILAC+13C(6)+15N(2)+(K)(32)' on line 65 is not recognized.

You must have a different final_fragment.csv file which is giving you the "Cannot modify sequence AGKR, length 4, at position 8." error.
-- Nick
 
trinidad martincampos responded:  2019-03-28 01:11
Hi Nick!!

I just uploaded an updated version called "293T_SILAC_1to1_rep01_IA_final_fragment.csv". You should see the error here...

Thanks a lot!!
Trini
 
Nick Shulman responded:  2019-03-28 02:15
Thanks, Trini,

Yes, that seems to behave exactly as you describe. BiblioSpec seems to be looking for at the first number in parentheses in order to determine the position of the modification, instead of looking at the last number.
I cannot think of any workaround that could get you past this bug.
We will try to fix this bug soon.
-- Nick
 
trinidad martincampos responded:  2019-03-28 03:25
Hi Nick!!

Thank you very much!!! I'd really appreciate it!!

Have a super good day!
Trini
 
trinidad martincampos responded:  2019-04-05 08:56
Hi Nik!!

Do you think you could give me an approximate date or duration to fix this bug?? I am planning to do some development is this direction and I could maybe add the script to Skyline at the end ;)

Thanks a lot and have a nice weekend!

Trini
 
Nick Shulman responded:  2019-04-15 12:44
Trini,

This bug will be fixed in the next release of Skyline-Daily. We usually come out with new releases of Skyline-Daily every couple of weeks.
-- Nick
 
trinidad martincampos responded:  2019-05-17 03:14
Thanks a lot!!

Looks good now ;)