search

GregorySchwartz / too-many-cells

Cluster single cells and analyze cell clade relationships with colorful visualizations.
https://gregoryschwartz.github.io/too-many-cells/
GNU General Public License v3.0
105 stars 19 forks source link

Normalization section in too-many-cells differential -h #23

Closed kuang-da closed 1 year ago

kuang-da commented 4 years ago

Hi,

Thanks for building such a nice tool. But I am a bit confused about the normalization in the differential analysis component. The help page does not help much since it basically identical to the help page of make-tree.

It seems so far the too-many-cell pipeline run two normalization algorithm: one before clustering and the other before DE analysis. By default, the normalization before clustering is TfIdfNorm and the normalization is NoneNorm so that the program will revoke edgeR's normalization algorithm. Is my understanding right?

I guess my question is to double-check the function of --normalization argument in too-many-cells differential. May I assume it is used to choose the normalization algorithm before DE and by default is should be NoneNorm?

Currently, the Normalization section in the help page of too-many-cells differential is identical, which makes me feel like I am choosing an algorithm for clustering.

Thanks for your time!

GregorySchwartz commented 4 years ago

This issue is due to a limitation in optparse-generic unfortunately https://github.com/Gabriel439/Haskell-Optparse-Generic-Library/issues/60.

kuang-da commented 1 year ago

Mark as resolved. Thank you!