question about grouping in report template/ intelligent reporting in Open Lab CDS 2.5

HI dwaelsve

I'm not sure what you want to do is actually possible.  Wouldn't it be easier to include the peaks in the Compound Identification table and simply group on Compound_Name?

/Andy

Hi
We never include unknown peaks in our compound identification table, this would also not be doable for most of our systems, as there can be a ton of unknown peaks.

Hi dwaelsve

Are you trying to group the unknown peaks by retention time together, using retention time (I think that's what I see).

I would use a custom calculation to create a nominal retention time (with however many decimal places you need) for each peak and use this in your grouping.

Would that work?

/Andy

Hi
I have no knowledge of custom calculations since i have never used them. So i would not know...

Regards

Hi, How would i do this?

Regards

Hi, How would i do this?

Regards

Hi dwaelsve

How to...

Open the custom calculation editor from the Processing tab in the ribbon in Data Analysis

Create a new custom calculation file and add a formula like this one (in this case, I've rounded to 1 decimal place to get consistency between all of the samples)

Save the custom calculation file and add it to the processing method.  Reprocess and you should see an extra column in the Results table

You can then use the custom calculation in the grouping in your results table.  The grouping in the report then need to be

 =IF(Compound_Name = "",CCR((Compound_CustomCalculationResults),"RoundedRT"), Compound_Name)

The CCR part comes from the Compound fields

 I've done this with all of the files in an example result set. 

Hope that makes sense.

/Andy

Hi

Thanks for the explanation.
But wouldn't this be exactly the same als just writing =IF(Compound_Name = "" , round(Peak_RetentionTime,1)   , Compound_Name ) ?
Regards

It might well be.  I'll go back to your first post... if you want to group peaks together +/- 0.01 of a retention time.  Which sample would be used as the reference, to define the centre point of the range?  Would it be the first one?  I'm wondering if the custom calculation could identify the unknown peaks and then find peaks in other samples within your +/- 0.01 minutes of the peaks in the first injection.

It's beginning to get complicated!!

"I'm wondering if the custom calculation could identify the unknown peaks and then find peaks in other samples within your +/- 0.01 minutes of the peaks in the first injection."
This is exactly something like i was aiming for!
Sadly though my colleagues shot my idea down. the peaks need to be grouped together with 100% chance of no errors. Since we cannot find a 100% foolproof way, we decided to only group identified peaks and leave the unidentified in a seperate table.