Empower custom calc. question(s)...

[NYYFan]

Hello all. So I've been doing some custom calculations for a little bit now, but mostly basic stuff. Recently, I've come across a problem using a more "complex" formula. I just got a new job and I'm the Empower "guru" (aka I'm the only person that's used Empower in my whole group of 60 people and they think I'm better than I actually am), however, the Custom calcs they are giving me are over my head so far.

Basically what I'm looking to do is take the amount from a single component in a standard which contains 4 components, and divide that by the amounts generated by all other samples in the sample set. I've had no success doing this and I've tried several ways. The label for that particular standard is S0102 and the component which I want to reference is Cholesterol. I have tried the following:

S0102.1.%CCompRef1[Amount]/[Amount] - Where CCompRef1 is designated as Cholesterol in my processing method. With this field I only get the amounts divided by my component peak for that spefic sample (S0102) and not in all my samples.

S0102.1%Cholesterol[Amount]/[Amount] and S0102.1%Cholesterol[Amount]/[%.%.%[Amount] - Didn't work..

Any help is apprecaited!

[NYYFan]

15 views and no one to Empower me

[benW]

Hello NYYFan,
You have a point!

As you have an intersample label you need bracketing to specify to what the label refers.

so try :
S0102.1.%(CCompRef1[Amount])/Amount

best regards,
Ben

[NYYFan]

Hello NYYFan,
You have a point!

As you have an intersample label you need bracketing to specify to what the label refers.

so try :
S0102.1.%(CCompRef1[Amount])/Amount

best regards,
Ben

That's one of the iterations I've tried the last few days too, with no success unfortunately That formula only gives my cholesterol peak amount divided by amounts for ONLY the sample labeled S0102. Basically my Sample Set will include 5 standards (S0102-S0106) and two unknowns (U0101 and U0102). I want to take the amount generated for the cholesterol component in sample S0102 and divide that by the amounts generated for all other standards and samples, in other words I will have this custom field populate in all samples and standards, whereas it's currently only populating for S0102. Each standard and sample has four components, each which generates an amount after processing, it's these amounts I want to divide the cholesterol component in sample S0102 from. I hope this is somewhat clear in what I'm asking. Thanks so much!

[benW]

Hi NYYfan,
So you are still having problems with your custom field. The custom field suggested will return the functionality you are after; it looks like you may have some other problem.

The field:
S0102.1.%(CCompRef1[Amount])/Amount
uses the value S0102.1.%(CCompRef1[Amount]) (which will be the Amount value for cholesterol in the standard labelled S0102 (injection 1) divided by the Amount calculated for each component in each injection. This field will be populated for all injections and peaks within your search area (where the variables are satisfied).

It sounds like you are not getting any values for the other components in other injections (“I will have this custom field populate in all samples and…”), is this correct?

If so, some suggestions as to why:
1) your calibration has not completed and are not returning values for Amounts for the other fields. – check you have Amount calculated for the components where your custom field is blank.
2) The CCompRef1 Amount value has not been calculated for sample S0102 when Empower calculates for the other components – use background processing to process the entire sample set in one go
3) CCompRef1 has not been entered for the other components.
4) ….

[NYYFan]

Hi NYYfan,

It sounds like you are not getting any values for the other components in other injections (“I will have this custom field populate in all samples and…”), is this correct?

Yup!

If so, some suggestions as to why:
1) your calibration has not completed and are not returning values for Amounts for the other fields. – check you have Amount calculated for the components where your custom field is blank.
2) The CCompRef1 Amount value has not been calculated for sample S0102 when Empower calculates for the other components – use background processing to process the entire sample set in one go
3) CCompRef1 has not been entered for the other components.
4) ….

Hopefully it's not reason #4 haha.
1) The calibration is complete as normal amounts are populating in all other samples.
3) CCompRef1 is labeled as Cholesterol for all other components in the processing method.
2) I like your explanation here, it sounds like a defintiely possiblity. How would I go about background processing and/or testing this? Basically I'd want to make sample S0102 is processed before all other injections were processed, correct?

Thanks again!

[NYYFan]

Hi NYYfan,

It sounds like you are not getting any values for the other components in other injections (“I will have this custom field populate in all samples and…”), is this correct?

Yup!

If so, some suggestions as to why:
1) your calibration has not completed and are not returning values for Amounts for the other fields. – check you have Amount calculated for the components where your custom field is blank.
2) The CCompRef1 Amount value has not been calculated for sample S0102 when Empower calculates for the other components – use background processing to process the entire sample set in one go
3) CCompRef1 has not been entered for the other components.
4) ….

Hopefully it's not reason #4 haha.
1) The calibration is complete as normal amounts are populating in all other samples.
3) CCompRef1 is labeled as Cholesterol for all other components in the processing method.
2) I like your explanation here, it sounds like a defintiely possiblity. How would I go about background processing and/or testing this? Basically I'd want to make sample S0102 is processed before all other injections were processed, correct?

Thanks again!

It was indeed #2! Awesome. I basically "told it" to process S0102 first and everything else fell into place. I could see this causing problems however... What if I wanted to use a field from a sample processed after S0102 to calculate something that I wanted to appear into S0102?

[NYYFan]

After basking in my glory of solving that last problem, and even after solving a few more! I have run into another roadblock...

I have a check standard calculation which will be the peak area of my check standard (label C0106 - 1 injection, 5 components) divided by the average area of 5 injections of my standard (label S0105). I've tried several things here, result type calc, peak type calc. I think what makes this more complicated is my sample has 5 components, and I only want the average of each component in those 5 injections (i.e. I want the average area of 5 cholesterol areas from 5 injections of the same sample, and the average areas of 5 L-608 areas from 5 injections etc.).

I've tried: C0106.%.%Area/S0105.%.%.AVE(Area) - result and peak type with and without "summarize custom fields"
C0106.%.%(CCompRef1[AREA])/SAME.%.%.AVE(CCompRef1[Area)) - where CCompRef1 referenced it's own component in the processing method.

Almost everything I've tried has given me numbers at least, which is kind of a good feeling, but they're wrong numbers, and I can't even figure out how they were calculated...

So, any help, yet again, would be greatly appreciated!

[jp_datasolutions]

You may have already solved this, but here is something that should work:

Peak Type custom fields will give you a result per component; result type will give you one result for the entire chromatogram. I assume you want a result for each component, so start with a Peak Type. Have you set the Sample Type for your Check Standards as Controls? If so, you can set the Sample Type as Control for your custom field. Then the custom field will only generate a value for the controls. What is the Sample Type for the standard injections? I assume Standard or they won't contribute to the calibration curve.

I would recommend the following custom fields to get the result you want.
StdAvg=S%.%.%.AVE(Area*EQ(SampleType,"Standard"))
This will get you an average of all standard injections (using the EQ (equal to) boolean)) where the label is S% (%=wildcard) for each component; Field Type=Peak; Sample Type=Standards & Controls Only.

Then divide the Area by this value:
ChkStd=Area/StdAvg
This will calculate the area of each Check Std by the average std area for the same component; Field Type=Peak; Sample Type=Controls Only

I hope this helps