Single-cell Datasets Integration

Question

[manishaB]

Hi folks!

I am trying to integrate human and mouse sc data using Seurat's pipeline. There are 4 human datasets for humans and 1 for mouse. Getting an error while finding anchors for integration. Code used,

data.anchors <- FindIntegrationAnchors(object.list = list(hf1_ob,hf2_ob,hf3_ob,hf4_ob,mf1_ob), dims = 1:20)

Please help in resolving the error given below.

Finding all pairwise anchors
  |                                                  | 0 % ~calculating  
**Error in checkSlotAssignment(object, name, value) : 
  assignment of an object of class “numeric” is not valid for slot ‘data’ in an object of class “Assay”; is(value, "AnyMatrix") is not TRUE**

Many thanks !

Answer 1

[jared.andrews07]

This is probably because gene names/IDs are not shared between the mouse and human datasets, so no common genes can be used as integration anchors.

[manishaB]

Hey, this really makes sense! The code ran successfully when I removed the mouse Seurat object from the list, all the 4 human datasets got integrated. But then, how should I proceed to merge data across species (say human AND mouse as of now) ? I think I am missing something here! Would like to receive some suggestions on it.

Thanks !

[jared.andrews07]

Grab a list of mouse-human orthologues via biomart or such, remove genes from each Seurat object that aren't found in the list for both species, rename remaining genes so that they are the same for both the mouse and human sets. I'd probably concatenate them in some way, like humangene_mousegene, i.e. CDK2_Cdk2.