including Skyline version # in exported report template?

support
including Skyline version # in exported report template? dgmccaskill  2015-03-13
 
Is it possible to include the version number of Skyline in a report template for exporting results from Skyline? I'm looking at ways of automating some downstream summary reports of exported results using R markdown. With your continued rapid development of Skyline it would be really useful if I could include the version number used to generate the results in the template.

Thanks so much for your continued rapid development of Skyline. ;)
 
 
Brendan MacLean responded:  2015-03-14
Hi Dave,
Thanks for raising this issue. As I mull this over, it seems like a complicated issue to me. There could actually be a number of Skyline versions involved in creating the data exported from any single Skyline document. As you note, there is the version that actually exported the report, but it may be that other versions of Skyline and its component tools that have touched the document content are even more important. I will start the list at:

1. All versions of Skyline that have touched the .sky document file.
2. All versions of Skyline that have contributed to the .skyd chromatogram and peak statistics data file.
3. All versions of Skyline/Bibliospec that have contributed .blib spectral library files.
4. Perhaps, all versions of Skyline that have contributed other library files, e.g. iRT and parameter optimization.
5. Perhaps, versions of Panorama and Skyline that were involved in creating any chromatogram libraries in use.
6. The version of Skyline that actually exported a report.

We do now write the last version of Skyline that saved the .sky file into the document, but that clearly falls short of being able to deliver full version information as described above. While it might be possible to deliver #6, I am not sure how useful it is and worry that it might deliver the false impression that the version is the sole source of all the exported report data. Arguably #2 is among the most important version data, and it could, in theory contain a different version for each imported raw data file. This is where the chromatograms and peak statistics get calculated.

I have no simple solution for you at the moment, beyond working the actual .sky file from which the report was produced into your workflow. It is human readable XML format. As I mentioned it does contain the Skyline version of the software that last saved the file. It would not be difficult to write a tool that parsed that out of the file and attached it to your report.

Intriguing problem. I am also interested in what you had in mind when you wrote the request. Where would you put the version number for #6? As a new row before the header row? That would obviously require a new option (off by default) in the UI to avoid breaking existing reports for people that have worked them into their workflows.

Thanks for your feedback. Please share more of our ideas and any solution you come up with by appending to this thread.

--Brendan
 
dgmccaskill responded:  2015-03-16
Hi Brendan,

Thanks for the detailed reply. I thought I was asking for something simple, but as you point out it's considerably more involved than I appreciated.

For what I was thinking, #2 and 3 would be the most important information I'm looking for, but I can appreciate it's not that simple.

I'm basically looking at how to use templates with R markdown that will pull data out of multiple text files (including those generated from Skyline) to produce a summary document of an analysis, including boilerplate text and things like a list of the raw data files, when they were acquired, fasta file(s) used in the database search, search algorithm(s) used, summary statistics of the data (e.g distribution of idotp, fwmh, peak areas etc)...

It could end up being a relatively long document, but it would be useful to for documenting *how* the results were generated from what is a fairly long data analysis process.

Thanks again.
 
axel ducret responded:  2015-03-17
Hi Brendan,

Thank you for a very detailed answer to Dave's question, which I can forecast will come more and more often. As Skyline is used increasingly in the community, it is bound at some point to enter a more "regulated" environment where software versionning is going to play a more important role. As you know, there are some other software packages to look at SRM (or other targeted) data but they are certainly not as convenient as Skyline. However, from a regulatory point of view, we would not be able to use Skyline as it is not compliant - yet.

This is now obviously future music --- but not that far fetched as QQQ instruments are starting to penetrate this market.

Best regards,

Axel