FCS Express compatibility

[malcook]

Has any testing been performed with .fcs files as generated by FCS Express. Specifically, should load_cytoset_from_fcs work as it does with FloJo .fcs files?

[mikejiang]

As long as fcs file generated conforms to the standard, cytolib should have no problem to parser it. Feel free to report it when it does

[malcook]

Apologies, my question was only poorly stated. I am really most interested in applying a gating strategy developed in FCSExpress under control of R, something akin to what flowWorkspace enables for flowjo workspaces. So far, I am successful in importing into an environment the .gatingML files (which recent versions of FCSExpress can export). I can identify the imported filters and apply them individually to a flowFrame or flowSet.

I am now building a GatingSet and using gs_pop_add to add to it each imported filter. In appreciation of your TIP that the standard gatingML mainly focuses on the definition of individual gates in a flat structure, which is really not suitable for representing the gating hierarchy from GatingSet, I am employing a naming convention for the gates to encode the hierarchy (or simple order) of gating (e.g. liveCells.1.0, singleCells.2.1, giving each named gate an index and the index of its parent) and thus how to structure the various calls to gs_pop_add .

Having done this, when I then recompute the GatingSet I (appropriately) get an error

Error in cpp_gating(x@pointer, y, alwaysLoadData, verbose, leaf.bool) : 
  colname not found: TransformsingleSS02-W.fcs.SS02-W

which makes sense because I have not (yet) learned how to add the transforms present in the gatingML to the GatingSet.

I am having trouble figuring out what my next steps should be. Perhaps examining the source for how the flowJo workspace import will give me the secret sauce I need at this point. But I would be grateful for any worked examples.

I am attaching an example .gatingML (which does not yet adhere to my proposed gate naming convention) and an .fcs file file if it helps you to help me.
21cy147.zip

Thanks

[altrabio]

Hi Malcolm,
I am wondering if you finally succeeded in importing an FCSExpress gating in R? If so, would you be able to share your code, it would so much helpful!
Thanks

[malcook]

Alas I stopped pursuing this line of approach but would be interested in learning of your or anyone else’s successes in such endeavors, or if perhaps the CytoML framework has improved support or documentation for such capabilities. _edit: remove email cruft_

[PedroMilanezAlmeida]

hey @mikejiang, just checking whether are there any plans to support FCSExpress layouts in near future?

[gfinak]

FCS express doesn't open their format (last time we checked), so no we have no plans to support it.

[PedroMilanezAlmeida]

thank you @gfinak!

[malcook]

@gfinak re:

FCS express doesn't open their format (last time we checked), so no we have no plans to support it.

IMO - no opening of format should be needed - FCS express exports gatingML files

@altrabio - I'm taking this up again in #86 - still hoping to find satisfadction - please alert me with any insights/progress you may have!

[gfinak]

Last I had contacted the FCS express folks about collaborating to support their exports in R there was no interest on their side. This was years ago, but even if they are exporting gating ml, there are always peculiarities and I have no desire to spend my time trying to reverse engineer another undocumented file format. Greg

…

On Tue, Jun 3, 2025, 18:03 Malcolm Cook ***@***.***> wrote: *malcook* left a comment (RGLab/CytoML#139) <#139 (comment)> @gfinak <https://github.com/gfinak> re: FCS express doesn't open their format (last time we checked), so no we have no plans to support it. IMO - no opening of format should be needed - FCS express exports gatingML files @altrabio <https://github.com/altrabio> - I'm taking this up again in #86 <#86> - still hoping to find satisfadction - please alert me with any insights/progress you may have! — Reply to this email directly, view it on GitHub <#139 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKSI6K4FRVFFSX3DM67MND3BYZWHAVCNFSM6AAAAAB6RDTOV6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMZXG44DMMRUGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[malcook]

reverse engineer another undocumented file format.

@gfinak I am sympathetic but hoped that FCS Express exporting using the reasonably well documented gatingML format would alleviate anyone's need for having to do so.

It seems to me that the effort is learning how to convert a naked (outside of a flowJo workspace) gatingML file into a form that can be applied to a gating set.

In #86 you will see I am trying to undertake this effort by using CytoML:::read.gatingML to read the xml into an R environment, and the writing setAs('environment','graphGML') as an approach to converting such an environment into a graphGML object which I hope to be apply to apply to a gating set using CytoML:::gating_graphGML.

Might you (or anyone) be able and willing to provide some guidance, perhaps in the form of discussion (in your choice of forum), as to whether this approach is reasonable and perhaps some insight into what read.gatingML.cytobank is doing that needs a parallel treatment in my setAs approach?