Manual Integration Question

<p>I have looked the following tech note (TECN10195095) that speaks to manual integration.  It doesn't speak to cases</p><p>where you need to modify the processing method as part of the manual integration.  Say your retention time in your</p><p>samples has shifted (stability samples for example) and the retention time window for your sample doesn't align</p><p>with your standard so you need to modify the window for that particular sample.  As I understand it you would need</p><p>to modify the window>integrate>quantitate>save processing method>save result.</p><p></p><p>When you do this (particularly with samples that are associated with a curve) the amount value is no longer calculated.</p><p>When you reprocess the result set (which is required to get your custom calculations to come out with the new result) you</p><p>will not see an amount value as the curve has been removed from that sample/processing method. </p><p></p><p>Does anyone know how to handle these situations where you have a curve and you need to apply a different processing method/integration</p><p>to your samples/standards?  How can you do this and still get the correct amount values for the sample?</p><p></p><p>Thanks!</p>

Answers

  • HI Chris

    Please review the attached paper describing how to handle manually adjusted results.

    best regards

    Rune

  • Hi Rune:

    This is the exact same thing that mentioned in my post.  It's the same as the tech note and this paper doesn't cover cases

    where you need to change the processing method.  It specifically talks about NOT changing the processing method.

    This is why I already mentioned that I read this document in my post.

    Thanks anyway,

    Chris

  • HI Chris

    Sorry for not understanding what you were looking for. Why can´t you just use the "manually identify peaks" function, to get the rigth ID on the peaks. This will not require changing or saving the processing method and will automatically calculate amounts ?

    best regards

    Rune

  • Empower insists that the processing methods are not changed and are identical for all standards and samples. So first, get as many injections integrated right using ONE single processing method. Then you can set about doing real manual changes, either integration or peak identification, but using the manualy identify peaks tool and not changing the processing method.

    H

  • Actually you don’t have to use the same processing methods. You can have different names and it works fine. I tested it.

    You can’t use manually identify peaks when using custom fields. The custom field calculations are not updated

    when batch processing because the manually identified peaks are not maintained. I appreciate the input though.

    I have figured out workarounds for this issue.

    Thanks,

    Chris

    Manual Integration Question

    reply from heather<http://forums.waters.com/people/heather&gt; in Informatics - View the full discussion<http://forums.waters.com/message/6050#6050

  • Hi Chris,

    If you are working in a GxP environment this could case some issues .

    Here is a way to get all of you're custom fields calculated, even intersample custom fields and run this with one processing method version and keep your maunal integration!

    1. Optimize your processing method

    2. Process your complete sample set with this processing method ( Process Only)

    3 Take the Result Set in Review and examine your integration.

         - manually adjust your integration:

              - For Standars : DO NOT hit the calibrate button !!!

              - For Unknowns : Hit the Quantify button ...

         - At the end of the Result Set : Save all changes by selecting : File - Save All

    4. Select the same Result Set

         - Process this Result Set by slecting Keep Manual Integration & Calibrate - Quantitate

    5 You will no have created a new Result Set with the manual integration settings AND with all custom fields calculated, even intersample custom fields

    With this you will see that all of your unknowns will be processed with the same processing method version ( Id) and the same calibration version ( Id) and calibration curves.

    Try it, you'll love it!

    Stefaan