Running hdWGCNA on merged dataset and harmonizing on "study"

smorabit / hdWGCNA

High dimensional weighted gene co-expression network analysis

https://smorabit.github.io/hdWGCNA/

Other

315 stars 31 forks source link

Running hdWGCNA on merged dataset and harmonizing on "study" #225

Closed mitchTwoTimes closed 2 months ago

mitchTwoTimes commented 2 months ago

Hi,

Do you see any issue running the standard hdWGCNA workflow on a harmony-corrected dataset? My approach is similar to your approach in Fig 6. Instead of the consensus approach (I found it didn't give me enough modules), I used "study" as my group.by.vars to harmonize in ModuleEigengenes. I did trait correlation and maybe 1-2 out of 17 correlated with study, but that was it.

Thank you

smorabit commented 2 months ago

Yes this sounds like a reasonable approach to me!