Marco,

I apologize for the delay in responding to your questions. I will be able to answer question #1 in this message. Another one of my colleagues is researching questions #2 and #3. He will respond later this week.

For your question #1, you have found a problem in this tutorial. The behavior of this feature has changed in LabKey v11.1 and the tutorial has not yet been updated. I am sorry for the confusion that this has caused. My suggestion is to skip the section of the tutorial named "Describe the mzXML Files" and move on to the next step. I will be working with the tutorial writer to get it fixed. When the tutorial has been fixed, I will send you a note so you can try it again.

Thank you,

Dear Brian,

   thank you for the reply. I skipped indeed that step when I followed the tutorial.
   However, I found very clear the organization of the previous options in "same protocol", "all fractions from the same sample" or "each with a different protocol" (that I found in the video version and that, I think, would have generated different computational steps). I hope that in the new tutorial it will be similar.
   Best regards,


       Marco

Marco,

regarding your #2 and #3 questions.
2: We've found that the XAR.xml format is usually too complicated to use effectively as a description of procedures that occur in the "wet lab" by researchers. Most people I've talked want to record a few properties and a protocol document to cover the things that go on in sample preparation. Therefore we haven't really pushed trying to get the XAR format to have better tool support for this. Instead we created the assay framework, which is based on the the same underlying experiment model the xar.xml format is based on. But the assay framework is a much simpler way to describe research protocols and relevant annotations. Some of the problems you were having with the tutorial had to do with this shift. we changed from generating or hand-authoring XARs for describing MS2 samples to describing them via an Assay. But the tutorial didn't get updated to reflect this.

It has turned out that the xar.xml format is more useful for recording and displaying a series of linked processing steps done by computers, such as an MS2 search pipeline job. We generate those xars as the steps of a pipeline job are run, then load the xar with the resulting scored data at the end of the search. That is what you see when you click on a graph icon in the MS2 Runs grid on an MS2 dashboard.

#3. I think for starters you are missing the file PepperProteins.xls, which is the custom protein list that has the real known concentrations of each spiked peptide. For some reason the attachments to the documentation topic on label-free quantitation were hidden; i have made them visible again. I have also discovered that the pre-filtering options for the Spectra counts datasets are partly broken-- the peptideprophet cutoff is not working. There may be other problems with the label-free demo, we haven't updated that doc in a while

thanks,
Peter

Dear Peter,

   thank you for having clarified me the different aims of a XAR and an assay (being a computer scientist I started from the point of view closer to my world!); I will try to follow the tutorial for the latter and to contextualize it for an MS(2) experiment.
   As far as point n. 3 is concerned, I found that the cited missing fields (those like "*_pepperspikedpr?") in my setup have different names and I can recover them by exporting the "labkey.data" table (they have names like "protein_customannotations_pepperproteins_property_conc__in_a"). Now I have other "less obscure" errors with the scripts DoExampleRunScoring and DoSASPECT, respectively:

Error in round(score.df$BetaMix, signif = 3) :
  unused argument(s) (signif = 3)
Calls: FormatOutputData

and

unused argument(s) (proteinInfo = list(Protein = c(45, 2712, ...
Calls: RunScoringFunction -> do.call -> SASPECT

however, I think they could be just compatibility issues among R versions; I will analyze the R code in the next future.
   Best regards.


           Marco