Why do you normalize the feature twice?

[lwtan90]

Hi,

I am trying to understand the purpose of performing row-normalization twice in the utils.py?

• once in line 191 for preprocessing of scRNA-seq data?
• the second time in the getGraph function in line 114 (as shown in the link below)?
  https://github.com/Joye9285/scGAC/blame/893909a7e46e0ed2e52fd1800f2c1606c4cf447a/utils.py#L114

since the values will be log-transformed in each normalization function, wouldnt this pose any problem?
I dont see a difference between normalization and normalization_for_NE function.
Thank you.

Wil