Downsampling reads vs. Seurat normalisation

Question

[Lucy]

Hi,

I am performing combined analysis of three 10x Genomics scRNA-seq samples using Cell Ranger and Seurat. I am trying to understand the purpose of downsampling the matrix/reads with DropletUtils or with Cell Ranger aggregate to account for differences in library size if you are then going to normalize for sequencing depth within the sctransform step in Seurat? Is there any advantage to performing both of these normalization steps or would you perform one or the other? In what situations would DropletUtils/Cell Ranger aggregate downsampling be advantageous over Seurat normalization and vice versa?

Many thanks,

Lucy