7 Replies Latest reply on May 7, 2018 7:20 AM by ryoboyle

    Calibration table importing incorrect peak area values?

    jtaylor7

      OpenLab CDS Rev. C.01.07.SR1 {113}

       

      We ran a sequence with five standards (calibration levels 1-5) with update RF/RT set to "replace". When the sequence completed, the generated calibration table had incorrect peak area values; the intelligent report clearly showed two different sets of data between the sample summary and calibration curve sections. When we deleted the calibration table and added each injection/level manually, it populated the area cells with correct values and the two sections on the intelligent report aligned. Is there another setting I'm missing that would affect this?

        • Re: Calibration table importing incorrect peak area values?
          ryoboyle

          Hello,

           

          I added tags to this post to increase visibility.

           

          How reproducible is this issue? Have you had this problem with multiple sequences, or just this one?

           

          Was there a significant retention time shift for the peaks in your calibration standards?

            • Re: Calibration table importing incorrect peak area values?
              jtaylor7

              It's happened a few times in the past but I can only find one other example at the moment. In this other example bad mobile phase was used which resulted in an unstable baseline and variable analyte retention times; however, the calibration table still populated with viable peak areas supposedly found at the normal retention time of the analyte. When I deleted individual levels and attempted to re-add them manually, nothing was found and the level did not repopulate with data, as expected.

               

              In the latest occurrence, there was no retention time shifts or baseline issues, but the same issue occurred with the calibration table populating with incorrect peak areas. Again, this only occurs with the automatically-generated table. If I do it manually, it's fine.

                • Re: Calibration table importing incorrect peak area values?
                  ryoboyle

                  Are any calibrated peaks missing or are not being integrated in any of your calibration sample data files?

                   

                  If you reprocess the result set in the Data Analysis view with the sequence still set to replace the RF/RT for the calibration samples, do you see the "correct" area reported?

                    • Re: Calibration table importing incorrect peak area values?
                      jtaylor7

                      All peaks were automatically identified and integrated without issue. I believe that, yes, when I reprocessed the data set the calibration table populated with the correct values.

                       

                      I also found some additional information that may help narrow down the issue. We looked back at an old sequence (the first use of the calibration curve) and noticed the peak areas in this sequence were the same as the ones improperly populating in the new sequences' calibration tables. Originally, after the calibration table for this old sequence was created, the master method's data analysis parameters were updated with the thinking that, going forward, the "amounts" column would remain pre-populated while the "areas" column would be replaced with the new values. However, it looks like it's simply importing both the X and Y values from the older sequence without updating them. I can send you a PDF with a few of our reports if needed.

                       

                      Ultimately, our goal is for the software to automatically generate a calibration curve based on the parameters pre-set in the sequence template. If we deleted the calibration table from one of the sequences and updated the master method, would it still be created automatically upon a new sequence's completion? Or would we, at the very least, need to have a master method with a calibration table with the intended X values saved?

                        • Re: Calibration table importing incorrect peak area values?
                          ryoboyle

                          A calibration table needs to be saved to the Master Method with valid retention time windows and level amounts in order for you to take advantage of the automatic calibration update functionality in ChemStation.

                           

                          It seems like something odd might have happened when the Master Method was updated with the older Sequence Method. If you are identifying an injection as a calibration sample and selecting to update the RT and RF then the new result set's Sequence Method should be updated with each calibration run.

                           

                          If this continues to happen, my first inclination would be that there might be a corruption in the Master Method or Sequence Template and I would recommend you try recreating both from the default method/sequence.

                            • Re: Calibration table importing incorrect peak area values?
                              jtaylor7

                              So I looked back at the old sequence and I think the retention time window was the issue. There's a 0.2min difference between the old and new retention time, so the calibration table probably wasn't recognizing the new peak (our window is +/-5%) and thus didn't update any of the areas.

                               

                              I updated the master method calibration table to only have retention time, compound name, and calibration levels, and then ran another sequence. It populated with correct areas this time. Curiously, it didn't populate the "amounts" column, so I just saved those to the master method as well. Is there a reason this parameter wasn't brought in from the sequence table?

                               

                              I'll go ahead and mark this answered since the primary issue seems to be rectified. Thanks again for your help.

                                • Re: Calibration table importing incorrect peak area values?
                                  ryoboyle

                                  That's great that it looks like you found the root cause.

                                   

                                  The amount needs to be entered in directly on the calibration table and cannot be updated via the sequence table as there is no corresponding sequence table field. There is a sequence table column for Sample Amount, but this is a value that is used to calculate a relative amount value if you are using a quantification mode of ESTD% or ISTD%.

                                   

                                  Only the response and retention times on the calibration table can be updated through the sequence table.