1
$\begingroup$

I am using Seurat to analyze my single cell data. I have 2 conditions, treated and untreated. I am trying to create a stacked bar graph in order to show the differences in cell types for each condition but need to collect the percentages of each cluster for the specific cell types. How to I put together a sheet that contains the percentages of each cluster and the make a stacked bar graph of comparison of each condition.

$\endgroup$

2 Answers 2

2
$\begingroup$
pt <- table(Idents(myobj), myobj$orig.ident)
pt <- as.data.frame(pt)
pt$Var1 <- as.character(pt$Var1)

ggplot(pt, aes(x = Var2, y = Freq, fill = Var1)) +
  theme_bw(base_size = 15) +
  geom_col(position = "fill", width = 0.5) +
  xlab("Sample") +
  ylab("Proportion") +
  scale_fill_manual(values = brewer.pal(12, "Paired")) +
  theme(legend.title = element_blank())

This shall work. myobj is your seurat object. This is my old code so if I remember correct, orig.ident is your treated and untreated conditions. Idents should give you cell clusters. If this does not work, please let me know

$\endgroup$
1
$\begingroup$

You can access these data from a Seurat object as columns of [email protected] and tabulate them.

There are functions such as “propellor” which will also implement this:

https://github.com/Oshlack/speckle

Note that a Seurat vignette also explores this topic:

https://satijalab.org/seurat/archive/v3.0/interaction_vignette.html

You need to be careful when doing this to ensure different proportions are biological as single cell data is prone to technical errors.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.