4a. Calculate doublet scores

doubScores <- addDoubletScores(
  input = ArrowFiles,
  nTrials = 10,
  k = 10, # Refers to how many cells near a "pseudo-doublet" to count.
  knnMethod = "UMAP", # Refers to the embedding to use for nearest neighbor search.
  LSIMethod = 1
)

ds <- addDoubletScores(
  input = ArrowFiles,
  nTrials = 10,
  k = 10, # Refers to how many cells near a "pseudo-doublet" to count.
  knnMethod = "UMAP", # Refers to the embedding to use for nearest neighbor search.
  LSIMethod = 1
)

4b. Save initial project

proj <- ArchRProject(
  ArrowFiles = ArrowFiles, 
  outputDirectory = ".",
  copyArrows = TRUE #This is recommended so that you maintain an unaltered copy for later usage.
)

proj <- saveArchRProject(ArchRProj = proj)
save.image(paste0(sample_name, "_project1_", Sys.Date(), ".RData"))

4c. Subset by TSS score

• Visualize using cell-ranger' ./QualityControl/...-TSS_by_Unique_Frags.pdf plots
• Cut off <10

idxPass   <- which(proj$TSSEnrichment >= 10) # primarily due to P30
cellsPass <- proj$cellNames[idxPass]
proj2     <- proj[cellsPass, ]
proj2     <- filterDoublets(proj2)

proj2 <- saveArchRProject(ArchRProj = proj2)
save.image(paste0(sample_name, "_project2_", Sys.Date(), ".RData"))

Results of filterDoublets

Filtering 190 cells from ArchRProject! P5 : 14 of 1218 (1.1%) E18 : 147 of 3843 (3.8%) E12 : 15 of 1249 (1.2%) P30 : 14 of 1196 (1.2%)