Brendan MacLean responded: 
20130413 
Hi John,
Unfortunately, you still need to export a report and calculate these ratios yourself in Excel, R or some other statistical tool. Skyline still has not native support for doing this type of normalization.
Sorry. It has been high on the list for several years, but not yet made to the top to be implemented.
Thanks for your feedback. Good luck with your data analysis.
Brendan 

Vahid F responded: 
20190827 
Hi Brendan,
Just a quick follow up on this since its been a few years now. Is there a way to normalize the area ratios to another surrogate standard in Skyline now in 2019? I tried looking up a few topics but this was only one I could find.
Thanks,
Vahid 

Brendan MacLean responded: 
20190827 


Vahid F responded: 
20190827 
I meant to ask if one can normalize the light to heavy ratios by another surrogate molecule i.e. light/heavy/surrogate or light/heavy/global in Skyline?
Sorry about the mix up.
Vahid 

Nick Shulman responded: 
20190827 
Vahid,
I do not understand what you are asking. Can you rephrase it or give an example?
 Nick 

Vahid F responded: 
20190828 
Hi Nick,
Some workflows need to use more than one standard to normalize the data. One such std would be an isotopically labeled heavy peptide and a second std would be a different peptide eluting at a different RT.
Is there a way in Skyline to normalize the data to both standards (heavy peptide and also another peptide treated as a global std).
Hope this clarifies.
Thanks,
Vahid 

Brendan MacLean responded: 
20190828 
Hi Vahid,
Do you want this as two separate ratios? i.e. do you mean you normalize to one and then the other and compare, or do you want a light/heavy/global, realizing that with each element you add to the normalization you are adding error?
In my experience, when results are dominated by random variance (and not systematic change) then taking the ratio to a global standard only increases the CVs (as you should expect when combining 2 random variables), but when systematic change is a problem, e.g. general signal decline on a mass spectrometer over time, then the normalization can decrease CVs, because the analyte and global standard have more correlated variance than they have random variance. And of course, with mass spectrometers values can change in one systematic jump (changing a column or even switching instruments) in ways that require normalization to be able to compare runs meaningfully.
Anyway, can you clarify again what you are looking for and why? Specifically:
1. Two normalized valued light/heavy and light/global
2. One normalized value including 3 random variables light/heavy/global
What is the expected benefit of the one you want?
Thanks for your feedback.
Brendan 

Vahid F responded: 
20190828 
Hi Brendan,
In my workflow which is a fairly complex one dealing with tissues, I have a protein level standard that is added earlier to the tissue lysate (for the lack of a SILAC) and serves to normalize sample prep variables (loss, leak etc..). A peptide from this protein std is taken as a "global" std.
The second standard however is a regular AQUA heavy peptide that is added right before digestion.
My approach is to normalize the data to both heavy peptide and the "global" peptide (light/heavy/global) as it should enhance the accuracy given the complexity of the workflow and high probability of systematic variables dominating the random variance. So to your question I am aiming to do #2. In a world that I could have a SILAC std, then that would be my only standard for normalization but without that I thought this would be best I can do to cover both systematic and random errors.
Let me know if you think this can be done in a better way.
Thanks,
Vahid 

Nick Shulman responded: 
20190828 
Vahid,
Skyline does not support normalizing to more than one number at the same time. This is something you would have to do in some other tool like Microsoft Excel.
I do not think it is a good idea to normalize against your global peptide and the heavy peptide at the same time. The usual reason for normalizing against a global standard is to compensate for variations in the concentration of the sample that you fed into the mass spectrometer. If you normalize against a heavy peptide that you spiked in, then that also has the effect of compensating for sample concentration. If you normalize against both at the same time, then you end up dividing by two numbers that are proportional to sample concentration, and that is not a valid thing to do.
One thing that is valid is to do is to calibrate normalized areas against an external standard. That is, you can have an extra well on your plate that contains a reference material, and you would look at the ratio of the normalized areas in your sample to the normalized areas in your external standard. I am not sure whether we have a good tutorial for how to do that, but it might be covered in the absolute quantification tutorial:
https://skyline.ms/wiki/home/software/Skyline/page.view?name=tutorial_absolute_quant
 Nick 

Vahid F responded: 
20190906 
Hi Nick,
Thank you for the information. I do use a calibration curve with reference material for the quantification.
The point here is that the two standards correct for two different sources of variables. One corrects for sample prep variables and and the other corrects for the LCMS variables.
Best,
Vahid 
