Merge pull request #4 from phipsonlab/bioconductor_review_update1

Changelog:

* Moved data for vignettes to `inst/extdata` and update vignettes accordingly.
* Added scripts to generate files in `inst/extdata` and a README to inst folder.
* Updated qs object for vignettes to qs2 object.
* Minor change in `runSuperCellCyto` to use `sprintf` to print warning message.

Author: ghar1821

DESCRIPTION

@@ -33,14 +33,14 @@ Suggests:
     testthat (>= 3.0.0),
     BiocSingular,
     bluster,
-    qs,
     scater,
     scran,
     Seurat,
     SingleCellExperiment,
-    BiocStyle
+    BiocStyle,
+    magick,
+    qs2
 VignetteBuilder: knitr
 Config/testthat/edition: 3
 URL: https://phipsonlab.github.io/SuperCellCyto/
 BugReports: https://github.com/phipsonlab/SuperCellCyto/issues
-

R/runSuperCellCyto.R

@@ -257,7 +257,7 @@ runSuperCellCyto <- function(
     if (n_pc > length(markers)) {
         warning(
             sprintf(
-                "n_pc (%d) > number of markers (%d). Setting n_pc to %d.",
+                "Requested n_pc (%d) > markers (%d). Setting n_pc to %d.",
                 n_pc, length(markers), length(markers)
             )
         )

inst/README.md

@@ -0,0 +1,21 @@
+# README
+
+Data in `extdata` folder are obtained from the following sources:
+
+* `Data23_Panel3_base_NR4_Patient9.fcs` and 
+`Data23_Panel3_base_R5_Patient15.fcs` are mass cytometry data from anti-PD1 
+  immunotherapy study by [Krieg et. al](https://doi.org/10.1038/nm.4466).
+  FCS files are available from 
+  [FlowRepository](
+  http://flowrepository.org/public_experiment_representations/1124).
+* `Levine_32dim_H1_sub.csv` and `Levine_32dim_H2_sub.csv` are mass cytometry
+  data from AML study by 
+  [Levine et. al](https://doi.org/10.1016/j.cell.2015.05.047).
+* `Levine_32dim_seurat_sub.qs2` and `Levine_32dim_sce_sub.qs2` are
+  Seurat object and SingleCellExperiment object containing subsampled
+  mass cytometry data from AML study by 
+  [Levine et. al](https://doi.org/10.1016/j.cell.2015.05.047).
+  
+Scripts to generate the different formats of `Levine_32dim` dataset is stored 
+in the `inst/scripts` directory of 
+[SuperCellCyto](https://github.com/phipsonlab/SuperCellCyto) repo.

inst/extdata/Levine_32dim_H1_sub.csv

@@ -0,0 +1,66 @@
+library(HDCytoData)
+library(data.table)
+library(qs2)
+
+
+sce <- Levine_32dim_SE()
+
+# isolate data from H1 and H2 patients
+sce_H1 <- sce[rowData(sce)$patient_id == "H1", ]
+sce_H2 <- sce[rowData(sce)$patient_id == "H2", ]
+
+# remove unassigned
+sce_H1 <- sce_H1[rowData(sce_H1)$population_id != "unassigned", ]
+sce_H2 <- sce_H2[rowData(sce_H2)$population_id != "unassigned", ]
+
+# randomly select 200 cells from each patient
+set.seed(42)
+sce_H1_sub <- sce_H1[sample(nrow(sce_H1), 200), ]
+sce_H2_sub <- sce_H2[sample(nrow(sce_H2), 200), ]
+
+# convert to data.table
+dt_H1_sub <- data.table(assay(sce_H1_sub, "exprs"))
+dt_H2_sub <- data.table(assay(sce_H2_sub, "exprs"))
+
+# save them as csv files
+fwrite(dt_H1_sub, "inst/extdata/Levine_32dim_H1_sub.csv")
+fwrite(dt_H2_sub, "inst/extdata/Levine_32dim_H2_sub.csv")
+
+
+# create the sce object
+dt <- data.table(assay(sce, "exprs"))
+dt[, population_id := rowData(sce)$population_id]
+dt[, sample := rowData(sce)$patient_id]
+
+# get 50 cells from each population and sample
+set.seed(42)
+n_cells <- 50
+sampled_dt <- dt[, .SD[sample(.N, min(.N, n_cells))], by = c("population_id", "sample")]
+
+# assign cell id
+sampled_dt[, cell_id := paste0("cell_", .I)]
+
+# create SCE object
+exprs_mat <- t(as.matrix(sampled_dt[, !c("population_id", "sample", "cell_id"), with = FALSE]))
+sce_sampled <- SingleCellExperiment::SingleCellExperiment(assays = list(counts = exprs_mat))
+SummarizedExperiment::colData(sce_sampled)$population <- sampled_dt$population_id
+SummarizedExperiment::colData(sce_sampled)$sample <- sampled_dt$sample
+SummarizedExperiment::colData(sce_sampled)$cell_id <- sampled_dt$cell_id
+colnames(sce_sampled) <- sampled_dt$cell_id
+
+qs_save(sce_sampled, "inst/extdata/Levine_32dim_sce_sub.qs2")
+
+
+# create Seurat object
+seurat_obj <- Seurat::CreateSeuratObject(
+    counts = exprs_mat,
+    assay = "originalexp",
+)
+seurat_obj <- Seurat::AddMetaData(seurat_obj, metadata = sampled_dt[, .(population_id, sample, cell_id)])
+# rename patient_id to sample
+
+
+qs_save(seurat_obj, "inst/extdata/Levine_32dim_seurat_sub.qs2")
+
+
+