Normalising data before running HTODemux

[jonoave]

Hello,

I'm working my way through the HTO vignette: https://satijalab.org/seurat/articles/hashing_vignette.html#hto-dataset-from-four-human-cell-lines

And I noticed that for both tutorials, there is some processing of the RNA assay prior to the HTO assay. Eg first dataset:

# Normalize RNA data with log normalization
pbmc.hashtag <- [NormalizeData](https://satijalab.org/seurat/reference/normalizedata)(pbmc.hashtag)
# Find and scale variable features
pbmc.hashtag <- [FindVariableFeatures](https://satijalab.org/seurat/reference/findvariablefeatures)(pbmc.hashtag, selection.method = "mean.var.plot")
pbmc.hashtag <- [ScaleData](https://satijalab.org/seurat/reference/scaledata)(pbmc.hashtag, features = [VariableFeatures](https://satijalab.github.io/seurat-object/reference/VariableFeatures.html)(pbmc.hashtag))

## then the HTO component is added .... 

For second dataeset

# Normalize data
hto12 <- [NormalizeData](https://satijalab.org/seurat/reference/normalizedata)(hto12)
hto12 <- [NormalizeData](https://satijalab.org/seurat/reference/normalizedata)(hto12, assay = "HTO", normalization.method = "CLR")

hto12 <- [HTODemux](https://satijalab.org/seurat/reference/htodemux)(hto12, assay = "HTO", positive.quantile = 0.99)

My question is, why do we need to run NormalizeData (or FindVariableFeaturePlot , ScaleData) on the RNA assay prior to HTODemux or exploring the HTO assay? Aren't both assays independent? How does the processing of the RNA assay influence the HTO assay?

Thank you in advance!