reporting pooled results with unknown number of samples

Hi Marty,

Thanks for your reply and suggested CC approach. I will try to clarify what i want the report to do. i have attached an excel file showing an example but it doesn't open properly for me so i have inserted the table below.

I want to report the pooled RSD for Precision 2 (1injection) + Precision (6 injections) = count 7; then report separately the pooled RSD for Precision 3 (1 injection) + Precison 2 (injection) + Precision (6 injections) = count 8. And so on for each check run every 5 samples. So if the pooled RSD for Precision 3 failed (injected after samples 6-10) but Precision 2 passed (injected after samples 1-5), samples 1-5 would be valid but samples 6-10 would not be valid. If there were a fixed number of samples run each time i could do it but  i don't know the number of samples going to be run and therefore i dont know the number of these Precision (single injections that need to be pooled with the preceding precision data (6 injections + single injection of each subsequent precision injection). Apologies for struggling to explain this rather odd requirement. I look forward to any further comment you may have. 

Hi Marty,

Thanks for your reply and suggested CC approach. I will try to clarify what i want the report to do. i have attached an excel file showing an example but it doesn't open properly for me so i have inserted the table below.

I want to report the pooled RSD for Precision 2 (1injection) + Precision (6 injections) = count 7; then report separately the pooled RSD for Precision 3 (1 injection) + Precison 2 (injection) + Precision (6 injections) = count 8. And so on for each check run every 5 samples. So if the pooled RSD for Precision 3 failed (injected after samples 6-10) but Precision 2 passed (injected after samples 1-5), samples 1-5 would be valid but samples 6-10 would not be valid. If there were a fixed number of samples run each time i could do it but  i don't know the number of samples going to be run and therefore i dont know the number of these Precision (single injections that need to be pooled with the preceding precision data (6 injections + single injection of each subsequent precision injection). Apologies for struggling to explain this rather odd requirement. I look forward to any further comment you may have. 

In this scenario is the next bracket also a count of 8? Would the results be the first 6, 3 and 4? If so you could do something like I have below. The first CC creates an index of all runs with the sample name CCC. Then the second CC references the first and uses that index plus a filter on sample type cal standard to do the calculation. You can change the filtering to other fields as needed. If you want to roll up all of the bracket injections up to the current one, you could change the query expression to be Where(function x.Injection.GetIntegerCC("CSSIndex") <= CurrentInjection.GetIntegerCC("CSSIndex") OR x.Injection.Sample.Sample_Name = "SSS". I have attached a notepad file with the expressions shown because the editor is displaying "(x)" as a .

Scope: Injection Type: Integer

CurrentSequence.AllInjections.Where(function x.Sample.Sample_Name = "CSS").OrderBy(function x.Sample.Sample_OrderNumber).Select(function x.Sample.Sample_OrderNumber).ToList.FindIndex(function x=Sample_OrderNumber)+1

Scope: Peak or Group Type: Double

RelStdDev("Peak_Area",CurrentSequence.AllIdentifiedPeaks.Where(function x.Injection.Sample.Sample_Type = 1).Where(function x.Injection.GetIntegerCC("CSSIndex") = CurrentInjection.GetIntegerCC("CSSIndex") OR x.Injection.Sample.Sample_Name = "SSS" OR x.Injection.GetIntegerCC("CSSIndex") = CurrentInjection.GetIntegerCC("CSSIndex")-1).Where(function x.Compound_Name = CurrentPeakOrGroup.Compound_Name))

Hi Marty

The next bracket would be 9 - each subsequent check would increase by a count of 1 as it would be pooling that single injection with the previous SS checks. Thanks for attaching the CC's in notepad. It will take a while for me to test this out as I'm working from home at the minute and unfortunately don't have access to the software. I have attached an example showing where in the run I want to report the RSD results. If you have any further comments on this please let me know.  Thanks very much for your input I will get back to you when I can.

Cheers

Okay if that is the way you are doing it then you can forget custom calculator although that would work with the <= index. The way you want to do this can be done in IR. The example below shows all the standards in one repeating table. The report shows 2 ways to present the values in the table or as fields below. The template relies on a filter of sample type as calibration standards but that could be changed to something else. It also assumes that the sample names of the first injection set are all the same and that the other standards have unique names. You could alter this to use another method to repeat them as needed. This method stores the peak_area as an aggregator indexed on compound_name in case you have more than on compound.  I am attaching the example template so you can see the way I did it. The template was done in 2.4.

That looks exactly how it needs to be done. Will the template which you have created in 2.4 be compatible with 2.3?

No but try this one, I changed the version in the raw template it should work.

Marty

Hi Marty,

Thanks for your time and expertise.  As I am unable to get into work yet (and its been a while) I haven't been able to look at the template you kindly sent. I will mark the answer as correct though as I have checked the report output shown and the calculations do what I need so I am confident that it will work. 

Many thanks