scRNA-seq reduction plot for individual group

Question

[Morris_Chair]

Hello everyone and happy new year !

I am analyzing scRNA-seq for three replicates of 2 conditions in total 6 libraries using SURAT pipeline. They are from 3 PDX and I want to see if regardless of the treatment all them three overlap in a dimension reduction plot such as a UMAP.

In the metadata I have the column "library" containing each library number from 1 to 6. Considering that 1-2, 3-4 and 5-6 belong to the same PDX, can you please guys help me out to visualize with just three different colour this three PDX in a UMAP plot?

Thank you

[skjobs]

You can use this tutorial

[http://cole-trapnell-lab.github.io/monocle-release/monocle3/][1]

[Morris_Chair]

Thanks for the link, do you know if there is a code that can be used with Seurat script?

[Morris_Chair]

Hi Arup, Thanks for the answer, I do have a subset in the metadata object called library, which contains all my libraries from 1 to 6. here is a screenshot

https://ibb.co/fSQyK0f

and here is my code for visualizing the colour clusters

seurat_object <- RunPCA(seurat_object, verbose = FALSE)
seurat_object <- RunUMAP(seurat_object, dims = 1:30, verbose = FALSE)

seurat_object <- FindNeighbors(seurat_object, dims = 1:30, verbose = FALSE)
seurat_object <- FindClusters(seurat_object, resolution = 0.09)
DimPlot(seurat_object, label = TRUE) + NoLegend()

it's not clear to me how can I use the SetIdent function, the idea is to show on a UMAP the three replicates with three different colours

can you please tell me more?

thank you

[Arup Ghosh]

Can you provide some information about how you are handling the 6 libraries, like are you creating separate Seurat objects then merging them for UMAP or merging them prior to loading in R.

[Morris_Chair]

I'm using three files received from 10X already processed with cellranger so the libraries are merged together in a unique data.frame prior loading in R. I'm following the Seurat pipeline and to distinguish cells of each libraries I'm first taking the last number after the UMI

library <- substr(Cells(seurat_object), 18,18)

adding the column library in metadata which now contains the libraries codes (from 1 to 6)

seuratobj <- AddMetaData( seurat_object, library, "library")

Thanks Arup

[Arup Ghosh]

Check the metadata has been applied to all the cells by length(seuratobj$library) if it returns the number of cells you are expecting, use seuratobj <- SetIdent(seuratobj, value = 'library') to change the active.ident . Now create the UMAP plot with DimPlot function. If possible move the comments to my answer as suggested by @genomax.

[Morris_Chair]

thank you Arup it worked :)

[GenoMax]

Please use ADD COMMENT/ADD REPLY when responding to existing posts to keep threads logically organized. This comment belongs under @Arup's answer.

SUBMIT ANSWER is for new answers to original question.

Answer 1

[Arup Ghosh]

Seurat sets the color of the UMAP plot based on the active.ident factor, if you have the metadata for each group/cells somewhere in Seurat object set it as the active.ident using SetIdent function and the plot color will change accordingly.