AVG DIA Tobi  2019-08-06 04:33
 

Dear Skyline Team,

could you please take a look at an issue with Avant-Garde DIA? Sadly the forum for Avant-Garde DIA does not seem to be maintained.

When trying to run it the execution is halted. The full error message is attached, but the main point seems to be

" cannot open file 'S:/avant garde test/AvantGardeDIA/DataFormatting/TempFiles/Report_GR_ColNamesName_Tag.csv': No such file or directory Execution halted"

The named file does not seem to exist. The preperation was done according to the AVG Manual, but it might still be an issue. Did anyone else experience an error like this one?

Best regards,
tobi

 
 
S.Vaca responded:  2019-08-06 06:29

Hi Tobi,
I'm sorry that you an into this error. What version of AvG are you using?
This error appeared due to the update of a dependency package that created a bug in the code. I have corrected it and the new version of Avant-garde (0.0.1.3) is now available. This should resolve the issue. If not, please let me know.

You can update the External tool directly in Skyline through the Tool Store, the R package should update automatically after running the first preparation script (Prep_1_CreateParamsFile). You can check that the software is updated using the Check_AvG_Version in the help module of the Avant-garde in Skyline.

Best,
Sebastian

 
Tobi responded:  2019-08-06 07:08

Dear Sebastian,

thank you very much for the very swift response. AVG Version 0.0.1.3 (freshly installed from tool store).

I also had other issues when R was not available for a few days (now fixed on its own) and also when the program was denied access to the R folders because of windows admin issues. These problems seem solved but there still can be sth wrong on my side.

It doesnt hurt having transitions for decoys right? Are some files in the AVG folder supposed to contain "Name_Tag" or is this meant contain the same name as the skyline document?

With the newest AVG and Skyline on Win 7 the issue remains that the temp file mentioned below is manually not observable in the folder and the program cannot continue. I noticed that AVG installed R-3.5.2 while some packages where built under R version 3.5.3

LaunchParallelTasks_ReadFromDB(ParamsFile,RefinementWorkflow)
Warning messages:
1: In connection_release(conn@ptr) :
[1] "Parallel: Done!"
There are 1 result in use. The connection will be released when they are closed
Read_AndFormatResults()
2: In connection_release(conn@ptr) : Already disconnected
Error in file(file, "rt") : cannot open the connection
Calls: Read_AndFormatResults -> read.csv -> read.table -> file
In addition: Warning message:
In file(file, "rt") :
cannot open file 'S:/avant garde test/AvantGardeDIA/DataFormatting/TempFiles/Report_GR_ColNamesName_Tag.csv': No such file or directory
Execution halted

 
S.Vaca responded:  2019-08-06 07:45

Hi Tobi,

It doesn't hurt to have decoys. AvG can handle that.
When you run the first preparation script a .R file is created. You can change the value of the Name_Tag parameter to anything you want to. This was done so you can annotate the results in any way you want. If you haven't changed that value then all the temporary files should have "Name_Tag" as the default.
I'll look at the files you sent today and report back.

Best,
Sebastian

 
S.Vaca responded:  2019-08-06 10:25

Hi Tobi,

I noticed that you are trying to run AvG on a Skyline file containing only one file. AvG was developed to improve the transition selection and peak boundaries determination by using the data from an entire dataset. It uses among others, the median profile of relative areas (MPRA) subscore. This is a dataset-wide score that measures how similar the signals in a given run are to all other runs in the entire dataset. Because you are analyzing only one file, the calculation of this metric is producing an error.

Could you try running AvG on a Skyline file that has at least two runs? Does this work or do you keep seeing the same error?
In the meantime, I'll correct this case in the upcoming version.

Best,
Sebastian

 
Tobi responded:  2019-08-06 23:52

Dear Sebastian,

thank you very much, i imported a differently named copy of the same raw file and it worked, I get the output and especially the refined peak picking looks really nice on the first view.

The "entire dataset" you mentioned, does it refer to having Tech and Bio replicates in a single skyline document? Or does it just mean its good to have lots of data for the program to run properly?

So far thank you very much for the help, its highly appreciated.

Best regards,
tobi

 
Tobi responded:  2019-09-13 03:02

Dear Sebastian,

I would like to adress one important question/issue of AVG. Does it treat MS1 and MS2 chromatograms as transitions in a similar/same fashion? I was surprised to see that also MS1 chromatograms get stripped of their quantitative state when going for MinimalNumberOfTransitionsAfterOptimization=3.

This might be solved by setting this value to 6, yet it would be nice having at least the option to tell avg to ignore MS1 for refinement and only refine MS2 chromatograms. (not sure if including/excluding MS1 for peak picking and scoring would make sense but optionally excluding it from refinement certainly does, especially for DIA). The issue is that any dotp (idotp and spectral dotp) require min 3 values, but MS1 usually has only 3 values to begin with so any refinement would automatically remove the idotp.

Could you please shortly comment on this and what would be the best way to run AVG on fragments while tagging MS1 along (import MS1 after AVG usage?)

Best,
tobias

 
S.Vaca responded:  2019-09-13 14:08

Hi Tobias,

This is a fair point. I normally don't curate the data based on MS1. I only do it using the MS2 data. The way I do it is, as you suggested, to only extract the data for MS2 and I add the MS1 traces later if I ever need them.

If you want to be able to do this now you can try, as you suggested, importing the MS1 after using AvG. Or you can modify the report template used by AvG. The report is called "AvantGardeDIA_Export" and should be under the External tools folder. You can edit it and add a filter so you don't include data from MS1 into the report. In the "Edit report" window, go to the "Filter" and select the "Fragment Ion" field. Add a filter so that the "Fragment Ion" does not contain the value "precursor" (see attached image). This will create a report that contains only MS2 data and when you will import the annotations it won't change the MS1 data.

Thanks for your feedback and let me know if this worked for you. I can add a filter to ignore the MS1 data in the next version of AvG.

Best,
Sebastian

 
Tobi responded:  2019-09-15 00:51

Dear Sebastian,

thank you very much, both workarounds do the job.

Best regards,
tobi