Fix cluster name mangling for all datasets

code/run-cepo.R

@@ -26,7 +26,18 @@ run_cepo <- function(sce, pars) {
     dplyr::bind_rows()
 
   # Fix column names.
-  result$cluster <- purrr::map_chr(result$cluster,
+  cluster <- result$cluster
+
+  # Fix for the Paul and CITE-seq datasets.
+  cluster <- ifelse(
+    substr(cluster, 1, 1) == "X" &
+    substr(cluster, 2, 2) %in% as.character(0:9),
+    substr(cluster, 2, length(cluster)),
+    cluster
+  )
+
+  # Fix for other datasets.
+  cluster <- purrr::map_chr(cluster,
     ~ switch(
       .x,
       # pbmc3k
@@ -45,10 +56,63 @@ run_cepo <- function(sce, pars) {
       "Endothelial...inflamed" = "Endothelial - inflamed",
       "Endothelial...peribronchiolar" = "Endothelial - peribronchiolar",
       "Endothelial...venule" = "Endothelial - venule",
+      # Zeisel
+      "pyramidal.CA1" = "pyramidal CA1",
+      "pyramidal.SS" = "pyramidal SS",
+      "endothelial.mural" = "endothelial-mural",
+      # Mesenchymal
+      "HAS1.High.Activated.FB" = "HAS1 High Activated FB",
+      "Matrix.FB" = "Matrix FB",
+      "MyoFB...Activated" = "MyoFB - Activated",
+      "PLIN2..FB" = "PLIN2+ FB",
+      "WNT2..FB" = "WNT2+ FB",
+      # SMART-seq3 pbmc
+      "CD14.monocytes" = "CD14 monocytes",
+      "Cytotoxic.T" = "Cytotoxic T",
+      "Cytotoxic.T.cells" = "Cytotoxic T cells",
+      "Dendritic.cells" = "Dendritic cells",
+      "Effector.CD4.T" = "Effector CD4 T",
+      "FCGR3A.monocytes" = "FCGR3A monocytes",
+      "FCGR3A.NK.cells" = "FCGR3A NK cells",
+      "HEK" = "HEK",
+      "HEK.cells" = "HEK cells",
+      "Memory.B" = "Memory B",
+      "Naive.B" = "Naive B",
+      "Naive.Memory.CD4.T" = "Naive/Memory CD4 T",
+      "Naive.Memory.CD8.T" = "Naive/Memory CD8 T",
+      "NK.cells" = "NK cells",
+      "Plasma.cells" = "Plasma cells",
+      "plasmacytoid.DC" = "plasmacytoid DC",
+      "Senescent.CD4.T" = "Senescent CD4 T",
+      # Zhao
+      "CCR7..CD4" = "CCR7+ CD4",
+      "CCR7..CD8" = "CCR7+ CD8",
+      "CD11c..mem.B" = "CD11c+ mem B",
+      "CD14..Mo" = "CD14+ Mo",
+      "CD16..Mo" = "CD16+ Mo",
+      "CD1c..DCs" = "CD1c+ DCs",
+      "Classical.mem.B" = "Classical mem B",
+      "CRCR3..CD8" = "CRCR3+ CD8",
+      "CX3CR1..CD8" = "CX3CR1+ CD8",
+      "CX3CR1..NK" = "CX3CR1+ NK",
+      "CXCR3..CD4" = "CXCR3+ CD4",
+      "CXCR6..CD8" = "CXCR6+ CD8",
+      "CXCR6..NK" = "CXCR6+ NK",
+      "cycling.ASC" = "cycling ASC",
+      "MerTK..Mo" = "MerTK+ Mo",
+      "Mo.doublet" = "Mo doublet",
+      "Naive.B" = "Naive B",
+      "NKT.doublet" = "NKT doublet",
+      "NKT.remove" = "NKT remove",
+      "noncycling.ASC" = "noncycling ASC",
+      "TCL1A..naive.B" = "TCL1A+ naive B",
+      "ZBTB32..mem.B" = "ZBTB32+ mem B",
       .x
     )
   )
 
+  result$cluster <- cluster
+
   lfcs <- calculate_log_fc(sce, genes = result$gene, clusters = result$cluster)
 
   result <- dplyr::mutate(