hvg selection in final embedding/clustering

[Nusob888]

Hi,

Congrats on the work and many thanks for uploading the code.
This is just a question regarding hvg selection for the final clustering and embeddings, which I am struggling to follow on the markdowns. I am mainly curious as you seem to achieve great separation in CD4 and CD8 subsets despite not using the ADT in the neighbourhood graph calculation (eg. with Seurat's WNN).

Specifically, for the "umap_harmony_RNA_wVDJ_30pcs_6000hvgs" embedding, I can't quite figure out where in the preprocessing markdown you ran code that corresponds to 6000hvgs? And were the hvgs re-calculated after the iterative clustering and removal of spurious/doublet cells?

Thanks in advance for the insight