Creating and Editing Groups

aspect ratio pixel

I'm going to make some groups with you guys, but I'm going to go ahead and do this live because I think it's more informative when we build up a workspace from scratch using some data. What I'm going to do right now is take this PFICSComp file folder that I have on my desktop and just drag it into the samples pane. In this file folder are the 46 FCS files that I just described: the All Stain samples, my FMO controls, and then a bunch of compensation controls. You see that they load in and they create a group in the workspace automatically that has the file folder name, and those 46 files when I click on that group become displayed in the samples pane down below with their names, a blank statistics column because we haven't added any analysis, and then the number of cells or events that were measured contained within that file. What we're going to do here is just demonstrate how you can make a couple groups to parse those individual samples out.

Groups, again, just act like file folders to organize your samples. Right now I have got an All Samples group, a Compensation group, and then my file folder PFICSComp group. You can have multiple samples in different groups, so my comps automatically populated in the Compensation group, but they are also in the PFICSComp group and the All Samples group. What I'm going to do here is to delete the PFICSComp group that came in and show you that deleting a group does not delete the files from the workspace within that group. They'll still be in the workspace under the All Samples group. The only way to remove a file from the workspace is to highlight the All Samples group, highlight the sample file you want to delete, and then press the Delete button, press OK. Then that sample goes away. I'm down to 45 samples in that group now. I just got rid of one of those compensation controls. If you want to bring that sample back in, go ahead and find it and drag it into the workspace samples pane, and then that sample will repopulate. There it is, and we're back up to 46 samples.

I'm going to go ahead and double-click on the Compensation group and show you why the 12 compensation controls automatically went into this Compensation group. The Compensation group automatically parses out into this group any sample files that contain in the $FIL keyword -- that's the file name keyword -- if they contain the word "unstained" or the word "comp" in the file name, then those samples go there automatically. This group is live and it already has sample inclusion criteria setup to parse the samples out based on those two words. So in this case, all of these sample controls, the compensation controls that I create, I label in the file name with the word "comps," either Bead Comps or Cell Comps. Then the Comp part of that name gets picked up by the group inclusion criteria and automatically puts those samples in, so I don't even have to move them into that group after I load them into the workspace.

Similarly, I'm going to make a new group now and I'm going to parse some other samples out. I'm going to parse some samples out based on their naming features. The Create Group button is located up in the static toolbar of the workspace. It's also in the FlowJo tab navigate band right here, Create Group. You'll notice that when I mouse over any of the action buttons up in the ribbon of FlowJo, you get a dropdown tool tip explaining what that tool does as well as a hotkey to initiate it. I'm going to go ahead and click the Create Group button. It will open up this menu where I now am prompted to give my group a name, a color, a textile, and a role; and I can optionally include samples based on their naming features or other keyword information. So I'm going to give my group, this first group, a name called All Stain. I mentioned my All Stain samples are the ones that got all the reagents. They are my real experiment. I like blue. I'll give it a bold italic and I'll make this role the test group.

Now I need to set up some inclusion criteria if I want this group that I create to automatically get all of the samples I want based on their naming. So I'm going to select from this dropdown keywords menu the $FIL keyword. If you don't have it shown there by default, you can search this menu with a $FIL and that will isolate this file name keyword. Then I'm going to select instead of equals, I'm going to say if the file name contains the acronym LD for lab donor, because this is just my lab donor cohort of samples that I was using for this experiment. Then I want those LD samples to go into my All Stain group. When I create the group, because I've set this inclusion criteria, you can see that that All Stain group is created. It has 20 samples in it, all of which fulfill the sample inclusion criteria that I set up.

I can do the same thing with my FMO, my fluorescence minus one controls. I'm going to make an FMOs group and in this case give it a control role. I'll say this is Controls, and I'm going to say if the file name contains the acronym FMO, then I want all of the FMO samples to go into that group. When I create the group, there's my FMOs group, and all of those 14 FMOs with that in the naming scheme go there.

You can do up to four layers of choices for inclusion criteria based on keyword and file naming. I'm going to make a Merged group now that includes both the LD samples and the FMO samples, and I'm going to call this my Master Gates group. I do this so I can unify the gating structure approach for all the samples that I need to gate. In this case, I don't like puke yellow, so you can give it a different color. I'm going to go with a nice red. I like bold italic, and I'm going to call this a Gating Level group. Now the file name contains FMO as the first set of criteria that's in there, but I can add additional choices here through the More Choices menu. They can be choices that it has to fulfill this and that, or this or that. I'm going to use OR because I want to say if the file name contains FMO or if the file name contains the acronym LD in the naming scheme to merge those two groups together. When I create the group, then I get this Master Gates group with 34 samples in it.

That's just one way that you can do groups. If you do not set any group inclusion criteria, if I just go ahead and create a group and give it a name, call it My Group and create it, then the group won't have any samples in it to begin with and you can always manually drag and drop between different groups. I'm going to select these four files and I'm going to drag them and drop them into the My Group, and now those four files live there. But that group is not dynamic. If I make a template out of this that I want to reuse from week to week, it won't automatically parse any samples into My Group when I bring them in. The other groups that I've created, because I've set these sets of sample inclusion criteria when I created the group, will automatically pick up any samples that come in that satisfy that criteria. It just makes it a little bit easier when you're doing repetitive analysis. And remember, you can always delete a group. You can always change the features of that group by double-clicking on it to bring up the Modify Group window and make changes to the inclusion criteria or the naming or the role or any other feature.

I think most people will bring data in and immediately want to go to our next topic, which is the Graph window. But before I do it, I really ... Yes, Ebony. Will you be able to have multiple comp groups if you're doing multiple panels? Yes, and in that case ... 100%. Good question. You would create a new group, call it Compensation Panel 2 and give it whatever color. The deal here is that if you have multiple panels you will see multiple sets of reagents. If you've labeled your reagents in the files, then you'll see one panel, the next panel, the third panel; and you can just use the panel as the inclusion criteria. That will parse them into groups of panels. But what I would do here, actually, is make a second group called Compensation Panel 2, give it a role of Compensation. And I won't do that. I'll make it not live. Create the group, and there's my Compensation Panel 2. Whatever samples I want to have in there I can drag and drop them, and so what I would do is distribute the proper compensation controls to each compensation group for multiple panels. I do that because you can get back to the setup of your compensation matrix easier if you separate the different panels worth of controls into different compensation groups. But you can definitely do that, and you can make as many compensation matrices as you want that way.