search

HelenaLC / CATALYST

Cytometry dATa anALYsis Tools
66 stars 31 forks source link

Filtering with CATALYST #407

Closed feanaros closed 1 month ago

feanaros commented 1 month ago

Dear @HelenaLC , thank you for your great package! I have a question for you. I have some .fcs files already normalized with beads and debarcoded. I want to keep only CD45+ cells filtering out the others. I followed your workflow but I didn't see any filtering step that allow to keep only specific cells. Could you suggest me a way to do it? BEST F.

HelenaLC commented 1 month ago

Try checking out filterSCE() (assuming you have identified a CD45+ cluster already?).

feanaros commented 1 month ago

Hi, thank you for your response. CD45 Is expressed almost everywhere, so there Is not a specific cluster for It. I Need to use It as control, removing that cells that don't express It. So I don't want to remove a specific cluster, but I Need to perform a sort of gating, excluding cells that are not viable and not express the marker.

HelenaLC commented 1 month ago

Alright. In that case I'd recommend checking out flowWorkspace, which supports gating. Alternatively, you could follow the CATALYST workflow (i.e., do unsupervised clustering); typically CD45- cells would form a separate subpopulation (say, cluster X), which you could remove via filterSCE(sce, cluster_id != "X"). See Fig. 6 in our workflow, for example, where cluster 2 appears to be CD45-, and was annotated as "surface-" downstream.

feanaros commented 1 month ago

Ok, perfect. I've already found a sort of cluster that express few CD45 cells and some Foxp3. It could be the cluster I'd like to remove, because Is not specific for anything. But on the other way, i want to do a double check doing the same analysis with those cells removed "a priori". Thank you for your help!