Why do you need SAME or the % for the channel position of a custom field?

Empower2018

I just tested a CF called A.%.SAME.AVE(Area) on PDA data that had 4 derived channel single wavelengths and when I processed the data I got results for every single channel. Then when I edited the formula to A.%..AVE(Area) I still got the same results for the channels! I don't understand why you need the SAME function if Empower will automatically cycle through each channel anyway to give you separate results..

Also when I entered a % in for the channel (A.%.%.AVE(Area)) it still gave me separate results for each channel when I presumed that the % would mean Empower adds up the areas for ALL channels ie the area of active in all injections of the first, second, third and fourth channels then divide the result by the total number of channels ie (200000/4) but this didn't happen- I got separate blocks of results per channel. Its probably something small but I don't understand how % and SAME are in any way different.

When I use a CF like %.%.%.AVE(Area) and I have several different labels in the sample set (for example, A, B, C, D etc) then Empower DID add up the area value for every single labelled injection which I presumed the % would do for all the channels. Any ideas??

MJS

I've put together a file with an embedded excel file to give an example...hopefully this is more clear. See the attachment.

ydan1977

I'm no Cf expert but % and SAME different

SAME = with same label , inj number or same channel name for intersample calcs

% = injection

More useful for label ie

Label.injection.channel. Function(statement)

as long as 3 dots appear before CF formula

Empower2018

Hi ydan, the % and ? symbols can be used in the place of Label Injection or Channel according to Empower Help. So for any injections from 10-20 you put 1? or for channels that have a common root name of TUV you put TUV* as channel position. 

The % for injection always uses "all" the injections so A.%..AVE(Area) adds up area for ALL injections for samples labelled as A. But there is no difference between using SAME or % for channel such as A.%.%.AVE(Area) or A.%.SAME.AVE(Area). I presumed % would add up areas for samples labelled as A for EVERY channel and not just each individual channel so I don't see the difference or logic for using SAME in CFs for channels because whether you use %, SAME or just leave it out you get the same result!

ydan1977

Good point Empower 2018

your logic seems correct

Guess if you have channels in processing method this will override CF

MJS

Leaving the placeholder as a period or using % would be equivalent as they are wildcards for any number of characters.  Using a ? would represent a wildcard for only a single character.

So you've got 4 labels, 4 channels each...A1, A2, A3, A4, B1, B2, etc.  Probably something you've addressed, but just in case: Do you have the same peak names in each channel/PM or across the labels (exactly the same spelling for the peak)?  If you had API1 in channel1, but API2 in channel2, for example, Empower wouldn't see them as the same and would not do a summary function on them for a peak field like area.  If you had API1 in both A1 and B1, however, Empower would see the peak as the same and perform a summary function where possible.  If this is the case, that would explain why you get individual values when using SAME or wildcard % in the channel designation because you have it locked to sample label A and there isn't anything similar so that leaves you with individual results.  When you moved the wildcard to the sample label and have peaks named the same, you do get an actual sum.

Another quick thing to verify is that you configured the search order/peak type similarly and didn't inadvertently set them incorrectly between your different configurations.

Empower2018

Hi MJS, to answer your questions. 

No, actually, I don't have the same peak names across the two channels or processing methods attached to the two derived channels. Its a good point. So I have 2 actives, active A in the first channel, and active B in channel B and only active A is in Component section of PM1 and only active B is in Component section of PM2. So I inject the sample 6 times. Are you saying that if I use a formula of A.%.%.AVE(Area) or A.%.SAME.AVE(Area) it will still give the same answer because there are different peak areas in each channel? So A.%.%.AVE(Area) is asking Empower to return average area of peaks in all injections of samples labelled as A- if only Active A in first channel then you get one answer for the 6 x injections of first channel and another answer for Active B for the 6 x injections of the second channel? 

But...if Active A and Active B were present in BOTH channels then a formula like A.%.%.AVE(Area) would return the average area for active A and active B in BOTH channels over all injections of A, is that correct? So if I had 2 injections of sample labelled A and for injection 1, Active A had an Area of 109000 in channel 1 and an area of 110000 in channel 2, and for injection 2, Active A had area of 108000 in channel 1 and 111000 in channel 2 then the value populated for the custom field A.%.%.AVE(Area) would be (109000+110000+108000+111000)/4?  And for A.%.SAME.AVE(Area) it would be (109000+108000)/2 at channel 1 and (110000+111000)/2 at channel 2 because you are locking it to specific channels?


I don't understand your point about the sample label of %. This is something I am struggling with- if you had a sample set with lots of different labels (A,B,F7,John etc) and different channels per sample, what exactly does a formula of %.%.%.AVE(Area) return? Is it the average area of peaks taking into account every single peak in every single labelled sample or is it blocked into labels so the average area of all injections with Label A, with Label B etc?

Empower2018

Hi MJS, thanks so much for taking the time to explain the syntax to me. That's easily the best answer I have ever read on here! I was getting confused with the whole % as a label and I wasn't sure what the difference between % and SAME was with channels of the same and different peaks but after studying your reply and Excel sheet I understand that the same peaks in more than one channel are all included with % or separated with SAME. 

I run a few methods where the peaks at channel 1 are not in channel 2 and vice versa so assay and calculations have to take that into account. I also run a method where the exact same peaks are in both channels so I know what to include now. I blame Empower 3 Help for some of this confusion, a lot of the articles have typos, are confusing to follow or are just plain wrong so its no wonder it can lead to confusion. It needs a complete revamp and spell check in my view. Cheers for the help.