How to edit peak picking in AutoQC? danielacgranato  2019-06-06
 

Hello Dear,

A System Suitability workflow is being implemented in the lab but we are experience an issue with the data uploaded to Panorama using AutoQC. A few peptides are being inaccurately picked over different runs (screenshot enclosed). We double checked opening the files into Skyline and the correct peak was always present. Is it possible to provide the correct peak boundaries or scheduling the acquisition is the best option? Can we edit peak picking on AutoQC online?

Best regards, Daniela

 
 
Brendan MacLean responded:  2019-06-06

Hi Daniela,
First, I don't understand the assertion about Panorama AutoQC being different from Skyline. The AutoQC Loader uses Skyline command-line for its data processing and the peak picking you see in the .sky file it leaves on local disk should be identical to what you see on Panorama, or to what you see if you open your template document and import the raw data manually. This is what I see when I download the .sky.zip file from your AutoQC folder. (see attached)

The primary issue is that you are relying on a constant slope and intercept applied to SSRCalc sequence-specific hydrophobicity scores for retention time prediction and the predictions are not very good. Secondarily, the peaks for LSSEAPALFQFDLK are just not very good in several samples (20190425_SysS1, 20190429_SysS1, 20190506_SysS1, 20190506_2_SysS1). And, finally, you don't have any library data for Skyline to go off of for expected fragment ion relative abundance.

I would probably remove the LSSEAPALFQFDLK peptide, calibrate iRTs based on the remaining 14 PRTC peptides and include the 4 BSA peptides in a new iRT calculator. This is probably enough, and I have included that iRT library here also. I think if you take your document, remove the LSS peptide and use the attached .irtdb file, you will find that Skyline and AutoQC do much better at reproducibly integrating the standards targeted in your system suitability runs.

This is really the goal and you generally want track relatively stable peptides in these runs to avoid false-positive system suitability events.

I also uploaded your runs fully curated with the iRT calculator to a PanoramaWeb.org chromatogram library folder just to see how that would work. I think it worked pretty well. I will send you your document using the chromatogram library in email to avoid making your data public. Though, of course, these runs only contain highly available standards.

Here is the documentation on Panorama support for Chromatogram Libraries in case anyone else reading is interested in creating a CLIB for their system suitability runs:

https://panoramaweb.org/wiki/home/page.view?name=chromatogram_libraries

--Brendan