Question

Improving Immgen Annotations

0

Entering edit mode

Adam • 0

@542da5c3

Last seen 3.1 years ago

United States

Hello!

Singler is wonderful, and I appreciate the ability to autoannotate cells. However, the Immgen reference is overly complex; the main dataset does not include all of the differences I need to look for, while the immgen fine dataset consists of a lot of repeating labels.

For example: the immgen.fine data labels include annotations for the following:

"T cells (T.4NVE)", "T cells (T.4NVE44-49D-11A-)", "T cells (T.4Nve)"),

Which for my purposes, can all be clumped into the group "T cells (T.4Nve)".

I cannot wrap my brain around how to condense these labels immgen is applying onto my sce data in a meaningful way, and then applying it to my data.

Of course, a tribble:

ann2 <- ann_fine
ann2 <- tribble(
    ~label.main, ~label.fine,
    "CD4 Naive", c("T cells (T.4NVE)", "T cells (T.4NVE44-49D-11A-)", "T cells (T.4Nve)"))

immgen.tc <- SingleR(test = sce, ref = immgen.tc, assay.type.test=1,
    labels = manual$label.fine)

## Filter based on manual new annotation
immgen.tc <- immgen[, immgen$label.fine %in% manual$label.fine]

## Append new label
immgen.tc$label.manual <- manual.vec[immgen.tc$label.fine]

Does not seem to work, as I recieve the following error:

Error in FUN(X[[i]], ...) : 
  failed to convert data.frame into a numeric matrix

Any ideas as to what to do?

Thank you!

Annotations • 1.0k views

ADD COMMENT • link 3.1 years ago Adam • 0

0

Entering edit mode

Just an update for posterity: i sorted it out!

Long story short, after pruning cells I did the following:


zsme$fine_groups[grepl("Tgd", zsme$fine_groups, fixed = T) ] <- "gd T Cells"
zsme$fine_groups[grepl("T.4N", zsme$fine_groups, fixed = T) ] <- "Naive T Cells"
zsme$fine_groups[grepl("T.8N", zsme$fine_groups, fixed = T) ] <- "Naive T Cells"
zsme$fine_groups[grepl("T.4EF", zsme$fine_groups, fixed = T) ] <- "Other"
zsme$fine_groups[grepl("T.4M", zsme$fine_groups, fixed = T) ] <- "Memory CD4 Cells"
zsme$fine_groups[grepl("T.8EFF", zsme$fine_groups, fixed = T) ] <- "Effector CD8 Cells"
zsme$fine_groups[grepl("T.8M", zsme$fine_groups, fixed = T) ] <- "Memory CD8 Cells"
zsme$fine_groups[grepl("T.Tregs", zsme$fine_groups, fixed = T) ] <- "T Reg Cells"
zsme$fine_groups[grepl("DC", zsme$fine_groups, fixed = T) ] <- "Other"
zsme$fine_groups[grepl("B" , zsme$fine_groups, fixed = T) ] <- "B Cells"
zsme$fine_groups[grepl("MF", zsme$fine_groups, fixed = T) ] <- "Other"
zsme$fine_groups[grepl("MO.", zsme$fine_groups, fixed = T) ] <- "Other"
zsme$fine_groups[grepl("T.CD8", zsme$fine_groups, fixed = T) ] <- "CD8 T cells"
zsme$fine_groups[grepl("T.8SP", zsme$fine_groups, fixed = T) ] <- "CD8 T cells"
zsme$fine_groups[grepl("T.CD4", zsme$fine_groups, fixed = T) ] <- "CD4 T cells"
zsme$fine_groups[grepl("T.4", zsme$fine_groups, fixed = T) ] <- "CD4 T cells"
zsme$fine_groups[grepl("ILC", zsme$fine_groups, fixed = T) ] <- "Other"
zsme$fine_groups[grepl("NK", zsme$fine_groups, fixed = T) ] <- "Other"
table(zsme$fine_groups)

Was it efficient? no, probably not. but it worked!

ADD REPLY • link 3.1 years ago Adam • 0