Define optimal PC space for new DLPFC SCE

[mattntran]

After doing #6, edit R-batchJob_DLPFC-n3_optimalPCselxn_LAH2021.R so can be submitted as a batch job, which will:

• [x] Perform getClusteredPCs() in the corrected PC space (i.e. from fastMNN())

  • This iterative clustering generates a DataFrame that helps define the PC space optimal for working in, for clustering, visualization, etc.

• [x] Print the 'n clusters vs d PCs' diagnostic plot (really, just need to update the pdf file location/name)

  • Write into pdfs/revision/ I already have going
  • Remove the donor brain nums from the title of the graphic

• [x] Save this DataFrame with the other objects into the same .rda

[mattntran]

@lahuuki you can edit the R_batchsub.sh in the main dir, to qsub

Last time I did this for n=3 samples, I did qsub -cwd -l bluejay,mf=48G,h_vmem=56G,h_fsize=40G R_batchsub.sh and that took ~90min

THANK YOU!!! 😁 🙌