Skip to content

Latest commit

 

History

History
12 lines (8 loc) · 786 Bytes

README.md

File metadata and controls

12 lines (8 loc) · 786 Bytes

Protobothrops mucrosquamatus gene expression networks

Data files

  • ercc.csv.gz spike in controls
  • factors.csv library description
  • rsem.csv.gz rsem data on gene expression and gene names

You can donwload the pre-computed WGCNA results (strongly recommended, they take a long time) here. The script looks for them in the data folder

To do

I don't think the chunk module_trait and anything below it works. There are good tutorials on the WGCNA website about how to conduct this type of analysis, particularly network preservation statistics. I strongly advise running those on Sango in parallel, and loading pre-computing results into RStudio, since they will take a long time, with lots of replicates.