Seurat: difference between FindAllMarkers and FindMarkers

[camilliano]

Hi,
I have a question about the difference of this 2 functions.

If I want to find all of the markers for each clusters I will use FindAllMarkers right?
_exemple: all_markers <- FindAllMarkers(object = pbmc)
head(x = all_markers)_

BUT, if I use FindMarkers, and I search for each identity (without ident.2),
_exemple: markers <- FindMarkers(object = pbmc_small, ident.1 = 3)
head(markers)_

will it be the same result?
I mean if I am looking for markers for my "identity 3" in both results, I will have the same?

[leonfodoulian]

Hi,

Yes, the results should be the same. Seurat::FindAllMarkers() uses Seurat::FindMarkers(). The corresponding code can be found at lines 329 to 419 in differential_expression.R. You can also double check by running the function on a subset of your data.

Best,
Leon

[camilliano]

Thanks for your answer!

I have an other question (which is a bit related),
when you run FindMarkers, in the output, you will have _"p_val, avg_logFC, pct.1, pct.2, p_val_adj"_ etc.
_exemple: cluster0.markers <- FindMarkers(object = object, only.pos = TRUE, ident.1 = 0, min.pct = 0.25)
print(x = head(x = cluster0.markers, n = 10))_

What is the difference between pct.1 and pct.2? What do they mean?

[leonfodoulian]

Hi,

pct.1 and pct.2 correspond to the proportion of cells expressing the gene of interest. pct.1 would be for ident.1 cells, and pct.2 for ident.2 cells (if NULL, then it corresponds to the proportion of cells from all non ident.1 clusters).

Best,
Leon