This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Defining the threshold for reads and features

I have some Indrop scRNA-seq

The number of genes per cell and number of read seems like this

enter image description here

enter image description here

Here what I should put to get a good Seurat object?

I have tried this but I think I am missing some cell types

pbmc <- subset(pbmc, subset = nFeature_RNA  1000 & nFeature_RNA < 5000 & percent.mt < 25)

Please look at these plots to give me an intuition which numbers give better results

scrna-seq
• 3,549 views • link
written by zizigolu zizigolu

1 answer

Read this chapter of the OSCA book. Nobody is going to be able to tell you hard and fast cutoffs that will just "work". Every experiment is different, as are different cell types. You will have to experiment and find a balance between stringency and retaining potentially real cell populations on your own.

• 0 views • link
written by jared.andrews07 jared.andrews07

Whatever I am trying there is no B cells in the data

How this is possible :(

That is why I am asking for help

I want to know how I can interpret this plot

• 0 views • link
written by zizigolu zizigolu

I have answers on these metrics here and here.

Regardless, if the cells aren't there (how do you know there aren't any B cells, have you annotated?), then they aren't there.

• 0 views • link
written by jared.andrews07 jared.andrews07

I have used these markers but likely nothing here very weird

>  VlnPlot(pbmc, features = "CD20")
Error in FetchData(object = object, vars = features, slot = slot) : 
  None of the requested variables were found: CD20
>  VlnPlot(pbmc, features = "CD22")
>  VlnPlot(pbmc, features = "CD19")
>  VlnPlot(pbmc, features = "CD79A")
>  VlnPlot(pbmc, features = "CD79B")
>

enter image description here

enter image description here

enter image description here

enter image description here

• 0 views • link
written by zizigolu zizigolu

Log in to answer this question.

More posts like this