Feature jing

.gitignore

@@ -1,8 +1,17 @@
-# For SCWorkflow dev
+.DS_Store
 .nfs*
-*.png
+#*.png
 Rcheck.txt
-*.pdf
+#*.pdf
 tests/testthat/otherData/
+tests/testthat/_snaps
+tests/testthat/fixtures/
+tests/testthat/output/
+
 .Rproj.user
-*.txt
+
+inst/doc
+#inst/extdata/*
+docs
+*.Rds
+*.rds

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: SCWorkflow
 Title: SCWorkflow from NIDAP
 Version: 1.0.2
-Authors@R:  c(person("Maggie", "Cam", email = "maggie.cam@nih.gov", role = "aut", comment = c(ORCID = "0000-0001-8190-9766")), 
+Authors@R: c(person("Maggie", "Cam", email = "maggie.cam@nih.gov", role = "aut", comment = c(ORCID = "0000-0001-8190-9766")), 
     person("Thomas", "Meyer", email = "thomas.meyer@nih.gov", role = c("aut", "cre"), comment = c(ORCID = "0000-0002-7185-5597")), 
     person("Jing", "Bian", email = "bianjh@nih.gov", role = "aut", comment = c(ORCID = "0000-0001-7109-716X")), 
     person("Alexandra", "Michalowski", email = "michaloa@mail.nih.gov", role = "aut", comment = c(ORCID = "0000-0001-9259-6101")),
@@ -18,7 +18,7 @@ Description: A set of functions for analyzing single-cell RNA-seq data using the
 License: MIT
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.2.3
+RoxygenNote: 7.3.3
 Suggests: 
     testthat (>= 3.0.0)
 Depends:

NAMESPACE

@@ -30,7 +30,6 @@ import(cowplot)
 import(data.table)
 import(dplyr)
 import(gdata)
-import(ggExtra)
 import(ggplot2)
 import(ggpubr)
 import(ggrepel)
@@ -48,14 +47,19 @@ import(scales)
 import(tidyverse)
 import(tools)
 import(utils)
+importFrom(BiocParallel,SerialParam)
 importFrom(ComplexHeatmap,pheatmap)
 importFrom(RColorBrewer,brewer.pal)
 importFrom(RColorBrewer,brewer.pal.info)
 importFrom(Seurat,AddMetaData)
 importFrom(Seurat,CreateAssayObject)
 importFrom(Seurat,DotPlot)
+importFrom(Seurat,FindVariableFeatures)
 importFrom(Seurat,Idents)
+importFrom(Seurat,RunPCA)
 importFrom(Seurat,RunTSNE)
+importFrom(Seurat,RunUMAP)
+importFrom(Seurat,SCTransform)
 importFrom(Seurat,as.SingleCellExperiment)
 importFrom(SingleR,SingleR)
 importFrom(colorspace,RGB)
@@ -75,20 +79,18 @@ importFrom(dplyr,mutate)
 importFrom(dplyr,mutate_if)
 importFrom(dplyr,pull)
 importFrom(dplyr,relocate)
+importFrom(dplyr,rename)
 importFrom(dplyr,row_number)
 importFrom(dplyr,select)
 importFrom(dplyr,summarise)
+importFrom(ggExtra,ggMarginal)
 importFrom(ggplot2,aes)
-importFrom(ggplot2,coord_fixed)
-importFrom(ggplot2,geom_hline)
 importFrom(ggplot2,geom_point)
 importFrom(ggplot2,geom_vline)
 importFrom(ggplot2,ggplot)
 importFrom(ggplot2,ggtitle)
-importFrom(ggplot2,scale_color_identity)
 importFrom(ggplot2,scale_y_reverse)
 importFrom(ggplot2,theme)
-importFrom(ggplot2,theme_bw)
 importFrom(ggplot2,theme_classic)
 importFrom(ggplot2,xlab)
 importFrom(ggplot2,ylab)
@@ -97,8 +99,13 @@ importFrom(ggpubr,annotate_figure)
 importFrom(ggpubr,get_legend)
 importFrom(ggpubr,ggarrange)
 importFrom(grDevices,colorRampPalette)
+importFrom(grid,gTree)
 importFrom(grid,grid.draw)
+importFrom(grid,grid.newpage)
+importFrom(grid,grobHeight)
+importFrom(grid,textGrob)
 importFrom(gridExtra,arrangeGrob)
+importFrom(gridExtra,tableGrob)
 importFrom(htmlwidgets,saveWidget)
 importFrom(magrittr,"%>%")
 importFrom(plotly,as_widget)
@@ -110,9 +117,12 @@ importFrom(scales,rescale)
 importFrom(stats,as.hclust)
 importFrom(stats,hclust)
 importFrom(stats,kmeans)
+importFrom(stats,mad)
+importFrom(stats,median)
 importFrom(stats,quantile)
 importFrom(stringr,str_replace_all)
 importFrom(stringr,str_sort)
+importFrom(stringr,str_split_fixed)
 importFrom(stringr,str_to_title)
 importFrom(stringr,str_wrap)
 importFrom(tibble,deframe)

R/3D_tSNE.R

@@ -94,6 +94,7 @@ tSNE3D <- function(object,
     htmlwidgets::saveWidget(as_widget(fig), filename, selfcontained = TRUE)
   }
 
-  tsne.results <- list("plot"  = fig, "data" = tsne.df)
+  tsne.results <- list( "data" = tsne.df,
+                        "plots"  = fig)
   return(tsne.results)
 }

R/AggregateCounts.R

@@ -0,0 +1,88 @@
+##' @title Aggregate Counts (Pseudobulk)
+##' @description Compute pseudobulk expression by averaging expression across groups
+##'              defined by one or more metadata columns, and return a tidy table.
+##' @details Uses Seurat's `AverageExpression()` on the `SCT` assay to compute
+##'          group-wise average expression for each feature. Also produces a
+##'          bar plot (via `ggplot2`/`plotly`) showing the number of cells per
+##'          pseudobulk group and warns if any group contains only one cell.
+##'
+##' @param object Seurat-class object.
+##' @param var.group Character vector of metadata column names used to define
+##'                  pseudobulk groups. When multiple columns are supplied, an
+##'                  interaction of these columns defines the groups.
+##' @param slot Character name of the assay data layer passed to
+##'             `AverageExpression()` (e.g., "data", "counts", or "scale.data").
+##'
+##' @return A data.frame of pseudobulk expression with columns `Gene` followed by
+##'         one column per pseudobulk group. Column names are sanitized to
+##'         contain only alphanumeric/underscore characters.
+##'
+##' @import Seurat
+##' @import tidyverse
+##' @import ggplot2
+##' @import plotly
+##' @importFrom dplyr select
+##'
+##' @export
+
+aggregateCounts <- function(object,
+                            var.group,
+                            slot){
+
+
+  ## --------------- ##
+  ## Main Code Block ##
+  ## --------------- ##
+
+  pseudobulk <- AverageExpression(object, 
+                                  return.seurat = FALSE,
+                                  assay = "SCT",
+                                  group.by = var.group,
+                                  slot = slot)[[1]] %>% 
+                  as.data.frame.matrix() 
+
+  pseudobulk$Gene <- rownames(pseudobulk)
+  pseudobulk <- pseudobulk %>% select("Gene", everything())
+  rownames(pseudobulk) <- NULL
+
+  # Further processing of column names
+  colnames(pseudobulk) <- gsub("\\W","_",colnames(pseudobulk))
+
+  # Return Table/Figure that gives statistics on 
+  # Number of Cells in each group/new sample
+  # Distribution of cell Counts in each group/new sample
+  meta <- object@meta.data[,var.group]
+
+  # check that columns are all factors / categorical
+  char_or_factor_cols <- sapply(meta, function(x) is.character(x) || is.factor(x) || is.logical(x))
+
+  # do plots and tables if all columns are factors or characters
+  if(all(char_or_factor_cols)){
+    meta$interaction <- gsub("\\W","_",interaction(meta))
+
+    df <- as.data.frame(table(pseudobulk_group = meta$interaction)) %>% filter(Freq != 0)
+    # sort the table by the number of cells in each group
+    df <- df[order(df$Freq, decreasing = F),]
+
+    if(any(df$Freq == 1)){
+      single_counts <- df$pseudobulk_group[df$Freq == 1]
+      # sprintf, make custom warning message with %s as placeholder for single count groups
+      warning(sprintf(
+        "Some groups have only 1 cell. It is recommended to have at least 2 cells in each group.\nAffected groups: %s",
+        paste(single_counts, collapse = ", ")
+      ))
+    }
+
+    p <- ggplotly(ggplot(df, aes(x = pseudobulk_group, y = Freq)) +
+                    geom_bar(stat = "identity", position = "stack") +
+                    labs(y = "Counts", x = "Pseudobulk Groups", title = "Number of Cells in each Pseudobulk Group") +
+                    theme(axis.text.x = element_text(angle = 90, hjust = 1)))
+
+    print(p)
+
+  } else {
+    stop("All columns in var.group must be factors or characters")
+  }
+
+  return(pseudobulk)
+}

R/Annotate_Cell_Types.R

@@ -300,7 +300,7 @@ annotateCellTypes <- function(object,
   # Returning Seurat Object and 2 plots:
   return(list(
     "object" = object,
-    "p1" = p1,
-    "p2" = p2
+    "plots"=list("p1" = p1,
+                 "p2" = p2)  
   ))
 }

R/Color_by_Gene.R

@@ -67,7 +67,11 @@ colorByGene <- function(object,
 
   print(object)
   # checking for samples
-  samples = eval(parse(text = gsub('\\[\\]', 'c()', samples.to.include)))
+  if(any(grepl('c\\(|\\[\\]',samples))) {
+    samples = eval(parse(text = gsub('\\[\\]', 'c()', samples)))
+  }else{
+    samples=samples
+  }
   # if none specified, using ALL
   if (length(samples) == 0) {
     samples = unique(object@meta.data$orig.ident)
@@ -218,11 +222,13 @@ colorByGene <- function(object,
     ###    plots <- gridExtra::grid.arrange(grobs=grob,nrow=n,newpage=F)
 
     if (return.seurat.object) {
-      result.list <- list("object" = object, "plot" = grob)
+      result.list <- list("object" = object, 
+                          "plots" = grob)
       return(result.list)
     } else {
       gene = as.data.frame(gene)
-      result.list <- list("object" = gene, "plot" = grob)
+      result.list <- list("data" = list("gene_table" = gene), 
+                          "plots" = grob)
       return(result.list)
     }