Group comparison bar graph error bars

support
Group comparison bar graph error bars cmwats2  2022-08-01
 

Hi. A group comparison bar graph plots the Log 2 fold change with error bars. If I right-click to "Copy Data" from the bar graph and paste in to excel, I get column headers Protein, Log 2 Fold Change, and StdErr. I have multiple comparisons so I made a custom report to export the information I want. I added a column to export "Standard Error"; however, "Standard Error" does not equal "StdErr."

What is the difference between the 2 standard error values?

Thanks in advance,
Caroline

"Copy Data" results:
Protein Log 2 Fold Change StdErr
ALDOA 0.317333532 0.097111714

Report export results:
Protein Log2 fold standard error
ALDOA 0.3173 0.049317949

 
 
Nick Shulman responded:  2022-08-01
Can you send us your Skyline document? I will figure out what is going on.

In Skyline you can use the menu item:
File > Share
to create a .zip file containing your Skyline document and supporting files including extracted chromatograms.

If that .zip file is less than 50MB you can attach it to this support request. You can upload larger files here:
https://skyline.ms/files.url

-- Nick
 
cmwats2 responded:  2022-08-01
I uploaded the Skyline file titled "GroupComparisonStdError." Thanks.
 
Nick Shulman responded:  2022-08-01
Thank you for sending that Skyline document.

The lengths of the whiskers which are drawn on the Fold Change Bar Graph is controlled by the "Confidence level" number that is specified in the Group Comparison definition.
For your group comparison, that confidence level is set to "95%", which means that the length of those whiskers will be 1.96 times the standard deviation, because, for a normal distribution, the 95% confidence interval is always 1.96 standard deviations.
If you were to change the Confidence level in the Group Comparison settings to 68%, then you would see that the size of the whiskers in the Fold Change Bar Graph were the same as the standard error value that you see in the Grid.

It would probably be better if the column that you got when you copied data from the Fold Change Bar Graph was named "95% confidence interval" instead of "StdErr". The reason that Skyline is doing what it is doing is that the "Copy Data" does not actually know anything about group comparisons. Whenever the Copy Data sees a box and whisker plot, it always calls the whisker length "StdErr", even if that's not what the whiskers represent in this case.

By the way, there is more information about how Skyline calculates fold changes using a linear regression here:
https://skyline.ms/announcements/home/support/thread.view?rowId=43007

-- Nick
 
cmwats2 responded:  2022-08-01
Thanks for clarifying!

Caroline