We use an FCS file keyword to hold the specimen id, and use a study specimen repo. When I import FCS files into a flow folder, I can then Customize View of FCSFiles table to show info about the Specimen (like ParticipantId, SequenceNum, DrawTimestamp, etc.) because I set the SpecimenId Column setting in the ICS Meta Data screen to by the appropriate keyword, and chose the appropriate target study during import. In Customize View, there is a SpecimenId property which has a Specimen property which has ParticipantId, SequenceNum properties, etc. Nice.

Now I import a FlowJo workspace, and have it resolve to existing FCS files - the ones I imported above. That works fine. When I customize view of the resulting FCSAnalyses, there is an "FCSFile" property which itself has an "Original FCSFile" property. As I understand it, the "FCSFile" property refers to the info about the file found in the FlowJo workspace, while the "Original FCSFile" property refers to the originally imported FCS file. The latter is the same as what shows up in the FCSFiles table (same wellId in the URL).

However, when I customize view of FCSAnalyses, and try to use FCSFile.OriginalFCSFile.SpecimenId.Specimen.*, those all show up blank - even though they are properly resolved when I use SpecimenId.Specimen.* in the FCSFiles table. If I instead use FCSFile.SpecimenId.Specimen.*, the data show up. The problem with this is the following:

Sometimes the operator entered the wrong value as the keyword value for specimenId. To fix this, we modify the keyword value in the FCSFile, and (re)import the FCSFile run. However, the FCSFile used in the FlowJo workspace might be the version with the wrong specimenId. This is why we prefer to use FCSFile.OriginalFCSFile.SpecimenId.Specimen.* rather than FCSFile.SpecimenId.Specimen.* - the former has been corrected, while the latter might not have been.

Note that when I customize view of FCSAnalyses to show FCSFile.OriginalFCSFile.Keyword.*, those show up fine, and have the values from the "corrected" file. It just seems to be the SpecimenId that does not show up.

This also affects Copy to Study: When I try to copy-to-study the analysis results, it uses the "bad" specimenIds found in the FlowJo workspace, not the corrected ones found in the FCSFiles table.

Do I understand things correctly regarding how this should work? Why are the FCSFile.OriginalFCSFile.SpecimenId.Specimen.* values showing up blank when I customize view of FCSAnalyses, but showing up fine when I customize view of FCSFiles? Is there any way to use the proper specimenId values during copy to study?

Thanks.

Andy Straw
University of Rochester