From e65e2846d9554bcd19567bd753081b810f0645db Mon Sep 17 00:00:00 2001 From: phoman14 Date: Tue, 17 Feb 2026 14:08:54 -0500 Subject: [PATCH 01/42] UsingClaude added Compare_Cell_Pop from NIDAPjson. passed AI tests --- CHANGELOG.md | 9 + DESCRIPTION | 1 + NAMESPACE | 20 ++ R/Compare_Cell_Populations.R | 269 ++++++++++++++++++ README.md | 16 ++ ...ompare_Cell_Populations.code-template.json | 103 +++++++ man/compareCellPopulations.Rd | 88 ++++++ man/violinPlot_mod.Rd | 8 +- .../helper-Compare_Cell_Populations.R | 71 +++++ .../testthat/test-Compare_Cell_Populations.R | 211 ++++++++++++++ 10 files changed, 792 insertions(+), 4 deletions(-) create mode 100644 R/Compare_Cell_Populations.R create mode 100644 inst/extdata/NIDAPjson/Compare_Cell_Populations.code-template.json create mode 100644 man/compareCellPopulations.Rd create mode 100644 tests/testthat/helper-Compare_Cell_Populations.R create mode 100644 tests/testthat/test-Compare_Cell_Populations.R diff --git a/CHANGELOG.md b/CHANGELOG.md index 1fe1d48..a3438a3 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,6 +1,15 @@ # CHANGELOG +## v1.0.3 (in development) +### Feature + +* feat: Add compareCellPopulations() function for comparing cell population distributions across experimental groups + - Visualizes cell population frequencies or absolute counts across multiple groups + - Generates alluvial flow bar plots and faceted box plots + - Supports custom group ordering and color palettes + - Added ggalluvial dependency for flow visualizations + - Generated from JSON template using json2r.prompt.md instructions ## v1.0.2 (2024-02-01) diff --git a/DESCRIPTION b/DESCRIPTION index 0a26fcb..fdd109e 100755 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -39,6 +39,7 @@ Imports: gargle (>= 1.2.0), ggplot2 (>= 3.3.6), ggpubr (>= 0.4.0), + ggalluvial, globals (>= 0.16.1), harmony (>= 0.1.1), hdf5r (>= 1.3.5), diff --git a/NAMESPACE b/NAMESPACE index 96b1ac4..0d3834c 100755 --- a/NAMESPACE +++ b/NAMESPACE @@ -3,15 +3,19 @@ export(aggregateCounts) export(annotateCellTypes) export(appendMetadataToSeuratObject) +export(build_modscore_plots) export(colorByGene) export(colorByMarkerTable) export(combineNormalize) +export(compareCellPopulations) +export(compute_modscore_data) export(degGeneExpressionMarkers) export(dotPlotMet) export(dualLabeling) export(filterQC) export(filterSeuratObjectByMetadata) export(heatmapSC) +export(launch_module_score_app) export(modScore) export(nameClusters) export(object) @@ -40,12 +44,14 @@ import(gridExtra) import(harmony) import(httr) import(jsonlite) +import(magrittr) import(parallel) import(plotly) import(quantmod) import(reshape2) import(rlang) import(scales) +import(tibble) import(tidyverse) import(tools) import(utils) @@ -76,6 +82,7 @@ importFrom(dplyr,arrange) importFrom(dplyr,case_when) importFrom(dplyr,desc) importFrom(dplyr,filter) +importFrom(dplyr,group_by) importFrom(dplyr,if_else) importFrom(dplyr,mutate) importFrom(dplyr,mutate_if) @@ -86,14 +93,26 @@ importFrom(dplyr,row_number) importFrom(dplyr,select) importFrom(dplyr,summarise) importFrom(ggExtra,ggMarginal) +importFrom(ggalluvial,geom_flow) importFrom(ggplot2,aes) importFrom(ggplot2,coord_fixed) +importFrom(ggplot2,element_blank) +importFrom(ggplot2,element_text) importFrom(ggplot2,geom_hline) +importFrom(ggplot2,geom_line) importFrom(ggplot2,geom_point) +importFrom(ggplot2,geom_segment) +importFrom(ggplot2,geom_violin) importFrom(ggplot2,geom_vline) importFrom(ggplot2,ggplot) importFrom(ggplot2,ggtitle) +importFrom(ggplot2,guide_legend) +importFrom(ggplot2,guides) +importFrom(ggplot2,scale_color_gradientn) importFrom(ggplot2,scale_color_identity) +importFrom(ggplot2,scale_x_continuous) +importFrom(ggplot2,scale_y_continuous) +importFrom(ggplot2,scale_y_log10) importFrom(ggplot2,scale_y_reverse) importFrom(ggplot2,theme) importFrom(ggplot2,theme_bw) @@ -128,6 +147,7 @@ importFrom(stats,kmeans) importFrom(stats,mad) importFrom(stats,median) importFrom(stats,quantile) +importFrom(stats,setNames) importFrom(stringr,str_replace_all) importFrom(stringr,str_sort) importFrom(stringr,str_split_fixed) diff --git a/R/Compare_Cell_Populations.R b/R/Compare_Cell_Populations.R new file mode 100644 index 0000000..36e2c17 --- /dev/null +++ b/R/Compare_Cell_Populations.R @@ -0,0 +1,269 @@ +#' @title Compare Cell Populations +#' @description Compare cell population distributions across different groups +#' using bar plots and box plots. Creates visualizations showing cell type +#' frequencies or counts across user-defined groupings. +#' +#' @details This function generates comparative visualizations of cell +#' populations from a Seurat object. It can display data as either frequency +#' percentages or absolute counts, and creates both stacked bar plots +#' (with alluvial flow connections) and grouped box plots for comparison +#' across samples and conditions. +#' +#' @param object A Seurat object containing the single-cell data +#' @param metadata.table A data.frame containing metadata (typically from +#' Seurat object's meta.data slot) +#' @param annotation.column Character string specifying the metadata column +#' containing cell type annotations to summarize in the bar plot +#' @param group.column Character string specifying the metadata column +#' defining groups to compare (e.g., treatment conditions) +#' @param sample.column Character string specifying the metadata column +#' containing sample identifiers. Default is "orig.ident" +#' @param counts.type Character string specifying plot data type: +#' "Frequency" (percentages) or "Counts" (absolute numbers). Default is "Frequency" +#' @param group.order Character vector specifying the order of groups in plots. +#' If NULL, uses natural order from data. Default is NULL +#' @param seurat.object.filename Character string for the Seurat object +#' filename. Default is "seurat_object.rds" +#' @param wrap.ncols Integer specifying number of columns for facet wrapping +#' in box plots. Default is 5 +#' +#' @import Seurat +#' @import ggplot2 +#' @import ggpubr +#' @import RColorBrewer +#' @import tibble +#' @import reshape2 +#' @import data.table +#' @import dplyr +#' @import magrittr +#' @import cowplot +#' @import gridExtra +#' @import grid +#' +#' @importFrom ggalluvial geom_flow +#' @importFrom stats setNames +#' @importFrom grDevices colorRampPalette +#' +#' @export +#' +#' @return A list containing: +#' \itemize{ +#' \item \code{Plots} - A list with two ggplot objects: +#' \itemize{ +#' \item \code{Barplot} - Stacked bar plot with alluvial flows +#' \item \code{Boxplot} - Faceted box plots by cell type (only if counts.type="Frequency") +#' } +#' \item \code{Table} - A data.frame with cell counts and percentages +#' } +#' +#' @examples +#' \dontrun{ +#' # Compare cell populations by treatment group +#' results <- compareCellPopulations( +#' object = seurat_obj, +#' metadata.table = seurat_obj@meta.data, +#' annotation.column = "cell_type", +#' group.column = "treatment", +#' sample.column = "sample_id", +#' counts.type = "Frequency" +#' ) +#' +#' # Display plots +#' plot(results$Plots$Barplot) +#' plot(results$Plots$Boxplot) +#' +#' # View summary table +#' head(results$Table) +#' } + +compareCellPopulations <- function( + object, + metadata.table, + annotation.column, + group.column, + sample.column = "orig.ident", + counts.type = "Frequency", + group.order = NULL, + seurat.object.filename = "seurat_object.rds", + wrap.ncols = 5 +) { + + ## -------------------------------- ## + ## Input Validation ## + ## -------------------------------- ## + + # Validate object + if (!inherits(object, "Seurat")) { + stop("Error: 'object' must be a Seurat object") + } + + # Validate metadata columns exist + required.cols <- c(annotation.column, group.column, sample.column) + missing.cols <- setdiff(required.cols, colnames(object@meta.data)) + if (length(missing.cols) > 0) { + stop("Error: The following columns are missing from metadata: ", + paste(missing.cols, collapse = ", ")) + } + + # Validate counts.type + if (!counts.type %in% c("Frequency", "Counts")) { + stop("Error: 'counts.type' must be either 'Frequency' or 'Counts'") + } + + ## --------- ## + ## Functions ## + ## --------- ## + + createAnnoTable <- function(SO, AnnoCol, GroupCol) { + ## Extract annotation data for each group + cntTble <- unique(SO@meta.data[[AnnoCol]]) %>% as.matrix() + + for (s in unique(SO@meta.data[[GroupCol]])) { + expr <- FetchData(object = SO, vars = GroupCol) + subSO <- SO[, which(x = expr == s)] + cntTble <- cbind(cntTble, table(subSO@meta.data[[AnnoCol]])) + } + + colnames(cntTble) <- c(AnnoCol, unique(SO@meta.data[[GroupCol]])) + cntTble <- cntTble[, -1, drop = FALSE] + cntTble <- data.frame( + apply(cntTble, 2, function(x) as.numeric(as.character(x))), + check.names = FALSE, + row.names = rownames(cntTble) + ) + + freqTble <- apply(cntTble, 2, FUN = function(x) { + return(x / sum(x)) + }) + freqTble <- (freqTble * 100) + + outTbl <- merge(cntTble, as.data.frame(freqTble), + by = 'row.names', + suffixes = c('_CellCounts', '_Percent')) + outTbl <- dplyr::rename(outTbl, 'Clusters' = "Row.names") + + return(list( + 'CellFreq' = freqTble, + 'CellCounts' = cntTble, + 'OutTable' = outTbl + )) + } + + ## --------------- ## + ## Main Code Block ## + ## --------------- ## + + # Replace dots with underscores in column names + colnames(object@meta.data) <- gsub("\\.", "_", colnames(object@meta.data)) + + # Update column names if they were modified + annotation.column <- gsub("\\.", "_", annotation.column) + group.column <- gsub("\\.", "_", group.column) + sample.column <- gsub("\\.", "_", sample.column) + + # Set up ordering + ordr <- object@meta.data[[annotation.column]] %>% + unique() %>% + sort() + + if (is.null(group.order)) { + group.order <- unique(object@meta.data[[group.column]]) + } + + # Set up colors + numColors <- max( + length(unique(object@meta.data[[annotation.column]])), + 20 + ) + colpaired <- colorRampPalette(brewer.pal(12, "Paired")) + cols <- c( + "#e6194B", "#3cb44b", "#4363d8", "#f58231", "#911eb4", "#42d4f4", + "#f032e6", "#bfef45", "#fabebe", "#469990", "#e6beff", "#9A6324", + "#800000", "#aaffc3", "#808000", "#000075", + colpaired(numColors) + ) + names(cols) <- ordr + + object@meta.data[[annotation.column]] <- factor( + object@meta.data[[annotation.column]], + levels = ordr + ) + + # Create tables + ColTables <- createAnnoTable(object, annotation.column, group.column) + BoxTables <- createAnnoTable(object, annotation.column, sample.column) + + metaGroups <- object@meta.data[, c(group.column, sample.column)] + rownames(metaGroups) <- NULL + metaGroups <- metaGroups %>% unique() + + ## Create plots based on counts type + if (counts.type == 'Frequency') { + ptbl <- melt(ColTables$CellFreq) + ptblBox <- melt(as.matrix(BoxTables$CellFreq)) + ptblBox <- merge(ptblBox, metaGroups, + by.x = 'Var2', by.y = sample.column, all.x = TRUE) + + labelCol <- 'PerValue' + ylab <- 'Frequency of each cell type (100%)' + } else if (counts.type == "Counts") { + ptbl <- melt(as.matrix(ColTables$CellCounts)) + ptblBox <- melt(as.matrix(BoxTables$CellCounts)) + ptblBox <- merge(ptblBox, metaGroups, + by.x = 'Var2', by.y = sample.column, all.x = TRUE) + + labelCol <- 'value' + ylab <- 'Cell Counts' + } + + # Format bar plot data + ptbl$Var1 <- factor(ptbl$Var1, levels = ordr) + ptbl$value <- round(ptbl$value, 1) + ptbl$PerValue <- paste0(ptbl$value, '%') + ptbl$PerValue <- gsub('^%$', "_", ptbl$PerValue) + ptbl[ptbl$value < 1, 'PerValue'] <- "" + ptbl$Var2 <- factor(ptbl$Var2, levels = group.order) + + # Create bar plot with alluvial flows + p2 <- ptbl %>% + ggplot(aes_string(y = 'value', x = 'Var2', fill = 'Var1', label = labelCol)) + + geom_flow(aes(alluvium = Var1), alpha = .2, + lty = 2, color = "black", + curve_type = "linear", + width = .5) + + geom_col(aes(fill = Var1), width = .5, color = "black") + + geom_text(size = 3, position = position_stack(vjust = 0.5)) + + theme_classic() + + ylab(ylab) + + xlab("") + + scale_x_discrete(guide = guide_axis(angle = 45)) + + scale_fill_manual(annotation.column, values = cols) + + # Create box plot + ptblBox$value <- round(ptblBox$value, 1) + ptblBox$PerValue <- paste0(ptblBox$value, '%') + ptblBox$PerValue <- gsub('^%$', "_", ptblBox$PerValue) + ptblBox[ptblBox$value < 1, 'PerValue'] <- "" + ptblBox[[group.column]] <- factor(ptblBox[[group.column]], levels = group.order) + + p2_Box <- ptblBox %>% + ggboxplot(y = 'value', x = group.column, add = "jitter", color = "Var1") + + facet_wrap(~Var1, ncol = wrap.ncols, scales = 'fixed') + + ylab(ylab) + + xlab("") + + theme(legend.title = element_blank()) + + # Return results + result <- list( + 'Plots' = list('Barplot' = p2, 'Boxplot' = p2_Box), + 'Table' = ColTables$OutTable + ) + + return(result) +} + +# Add global variables to avoid R CMD check NOTEs +utils::globalVariables(c( + "Var1", "Var2", "value", "PerValue", "alluvium", + ".", "CellFreq", "CellCounts", "OutTable" +)) diff --git a/README.md b/README.md index f2e1e03..83f6795 100644 --- a/README.md +++ b/README.md @@ -17,6 +17,22 @@ R package for Single Cell analysis The Single Cell Workflow streamlines the analysis of multimodal Single Cell RNA-Seq data produced from 10x Genomics. It can be run in a docker container, and for biologists, in user-friendly web-based interactive notebooks (NIDAP, Palantir Foundry). Much of it is based on the Seurat workflow in Bioconductor, and supports CITE-Seq data. It incorporates a cell identification step (ModScore) that utilizes module scores obtained from Seurat and also includes Harmony for batch correction. +## Key Functions + +### Sequential Workflow +1. **processRawData()** - Process H5 files into Seurat objects +2. **filterQC()** - Quality control and filtering +3. **combineNormalize()** - Merge samples, normalize, dimension reduction +4. **Harmony integration** (optional) - Batch correction +5. **annotateCellTypes()** - Automatic cell type annotation via SingleR + +### Analysis & Visualization +- **compareCellPopulations()** - Compare cell population distributions across groups +- **degGeneExpressionMarkers()** - Differential expression analysis +- **reclusterSeuratObject()** / **reclusterFilteredSeuratObject()** - Subset and re-cluster +- **colorByGene()**, **heatmapSC()**, **violinPlot_mod()** - Visualization functions +- **plotMetadata()**, **dotPlotMet()** - Metadata visualization +
For further documentation see our detailed [Docs Website](https://nidap-community.github.io/SCWorkflow/) diff --git a/inst/extdata/NIDAPjson/Compare_Cell_Populations.code-template.json b/inst/extdata/NIDAPjson/Compare_Cell_Populations.code-template.json new file mode 100644 index 0000000..e9ae00e --- /dev/null +++ b/inst/extdata/NIDAPjson/Compare_Cell_Populations.code-template.json @@ -0,0 +1,103 @@ +{ + "codeTemplate": "unnamed_21 <- function({{{Object}}},{{{Metadata_Table}}}) {\n\n## --------- ##\n## Libraries ##\n## --------- ##\nlibrary(Seurat)\nlibrary(ggplot2)\nlibrary(ggpubr)\nlibrary(RColorBrewer)\nlibrary(tibble)\nlibrary(reshape2)\nlibrary(ggalluvial)\n#library(plotly)\nlibrary(data.table)\nlibrary(dplyr)\nlibrary(magrittr)\nlibrary(cowplot)\nlibrary(gridExtra)\n#library(EnhancedVolcano)\nlibrary(grid)\nlibrary(nidapFunctions)\n \n #nidapLoadPackages(\"SCWorkflow\")\n\n## -------------------------------- ##\n## User-Defined Template Parameters ##\n## -------------------------------- ##\n\n\nseurat_object={{{Object}}}\nMetaData={{{Metadata_Table}}}\n\n#Basic Parameters:\nAnnoCol='{{{Annotation_Column}}}'\nGroupCol='{{{Group_Column}}}'\nSampleCol='{{{Sample_Column}}}'\nplotType='{{{Counts_Type}}}'\ncolor='custom'\ngroup_order=c({{{Group_Order}}})\nwrap_ncols=5\n\n#Filesave Parameters:\n seurat_object_filename <- \"{{{Seurat_Object_Filename}}}\"\n\n##--------------- ##\n## Error Messages ##\n## -------------- ##\n\n## --------- ##\n## Functions ##\n## --------- ##\n\n## --------------- ##\n## Main Code Block ##\n## --------------- ##\n\n\n\n ## --------------- ##\n ## Main Code Block ##\n ## --------------- ##\n\n path <- nidapGetPath(seurat_object,seurat_object_filename)\n SO <- readRDS(path)\n\n colnames(SO@meta.data) <- gsub(\"\\\\.\",\"_\",colnames(SO@meta.data))\n\n\nout=ComapreCellPop(SO=SO,\n AnnoCol=AnnoCol,\n GroupCol=GroupCol,\n SampleCol=SampleCol,\n group_order=group_order,\n plotType=plotType,\n color=color,\n wrap_ncols=wrap_ncols)\n\n#plotType%>%print\n#print(out)\n#print(out[['freqTble']])\n\nplot(out$Plots$Barplot)\n\nif(plotType=='Frequency'){\n plot(out$Plots$Boxplot)\n}else if (plotType=='Counts'){\n\n}\n\n#return(as.matrix(out$Table))\nreturn(NULL)\n}\n#################################################\n## Global imports and functions included below ##\n#################################################\n\n\nComapreCellPop=function(SO,\n AnnoCol,\n GroupCol='Group',\n SampleCol='orig_ident',\n group_order=NULL,\n plotType='Frequency',\n color='custom',\n wrap_ncols=5){\n \n if (is.null(group_order)) {\n group_order=unique(SO@meta.data[[GroupCol]])\n } \n # SO=AnnoOut_out$object\n ordr=SO@meta.data[[AnnoCol]]%>%unique%>%sort\n numColors = max(length(unique(SO@meta.data$mouseRNAseq_main)), length(unique(SO@meta.data$immgen_main)))\n colpaired = colorRampPalette(brewer.pal(12, \"Paired\"))\n cols = c(\n \"#e6194B\",\n \"#3cb44b\",\n \"#4363d8\",\n \"#f58231\",\n \"#911eb4\",\n \"#42d4f4\",\n \"#f032e6\",\n \"#bfef45\",\n \"#fabebe\",\n \"#469990\",\n \"#e6beff\",\n \"#9A6324\",\n \"#800000\",\n \"#aaffc3\",\n \"#808000\",\n \"#000075\",\n colpaired(numColors)\n )\n \n names(cols)=ordr\n SO@meta.data[[AnnoCol]]=factor(SO@meta.data[[AnnoCol]],levels=ordr)\n \n \n ### Create cnt Table by any group\n CreateAnnoTable=function(SO,AnnoCol,GroupCol){\n \n ## extract annotation data for each group\n cntTble=unique(SO@meta.data[[AnnoCol]])%>%as.matrix\n for (s in unique(SO@meta.data[[GroupCol]]) ){\n expr <- FetchData(object = SO, vars = GroupCol)\n subSO=SO[, which(x = expr ==s)]\n cntTble=cbind(cntTble,\n table(subSO@meta.data[[AnnoCol]])\n )\n }\n colnames(cntTble)=c(AnnoCol,unique(SO@meta.data[[GroupCol]]))\n cntTble=cntTble[,-1]\n cntTble=data.frame(apply(cntTble, 2, function(x) as.numeric(as.character(x))),check.names=F, row.names = rownames(cntTble))\n \n \n freqTble=apply(cntTble,2,FUN = function(x){\n return(x/sum(x))\n })\n freqTble=(freqTble*100)\n \n \n outTbl=merge(cntTble,as.data.frame(freqTble),by='row.names',suffixes = c('_CellCounts','_Percent'))\n outTbl=dplyr::rename(outTbl,'Clusters'=\"Row.names\")\n # colSums(outTbl[,2:ncol(outTbl)])\n return(list(\n 'CellFreq'=freqTble,\n 'CellCounts'=cntTble,\n 'OutTable'=outTbl))\n }\n \n ColTables=CreateAnnoTable(SO,AnnoCol,GroupCol) \n BoxTables=CreateAnnoTable(SO,AnnoCol,SampleCol) \n metaGroups=SO@meta.data[,c(GroupCol,SampleCol)]\n rownames(metaGroups)=NULL\n mataGroups=metaGroups%>%unique\n SampleCol=colnames(mataGroups)[ncol(mataGroups)]\n \n \n ####################################\n ## Create Annotation Column Plot\n if (plotType=='Frequency') {\n ptbl=melt(ColTables$CellFreq)\n ptblBox=melt(as.matrix(BoxTables$CellFreq))\n ptblBox=merge(ptblBox,metaGroups,by.x='Var2',by.y=SampleCol,all.x=T)\n \n labelCol='PerValue'\n ylab='Frequency of each cell type (100%)'\n }else if (plotType==\"Counts\") {\n ptbl=melt(as.matrix(ColTables$CellCounts))\n ptblBox=melt(as.matrix(BoxTables$CellCounts))\n ptblBox=merge(ptblBox,metaGroups,by.x='Var2',by.y=SampleCol,all.x=T)\n \n labelCol='value'\n ylab='Cell Counts'\n }\n \n ptbl$Var1=factor(ptbl$Var1,levels=ordr)\n ptbl$value=round(ptbl$value,1)\n # ptbl$PerValue=round(ptbl$value,0)\n ptbl$PerValue=paste0(ptbl$value,'%')\n ptbl$PerValue=gsub('^\\\\%$',\"_\",ptbl$PerValue)\n ptbl[ptbl$value<1,'PerValue']=\"\"\n \n ptbl$Var2=factor(ptbl$Var2,levels=group_order)\n \n \n p2=ptbl%>%ggplot(\n aes_string(y = 'value', x = 'Var2',fill='Var1',label = labelCol)) +\n geom_flow(aes(alluvium = Var1), alpha= .2, \n lty = 2, color = \"black\",\n curve_type = \"linear\", \n width = .5) +\n geom_col(aes(fill = Var1), width = .5, color = \"black\") +\n geom_text(size = 3, position = position_stack(vjust = 0.5))+\n theme_classic()+\n ylab(ylab)+\n xlab(\"\")+\n scale_x_discrete(guide = guide_axis(angle = 45))\n if (color!='orig') {\n p2=p2+scale_fill_manual(AnnoCol,values = cols) \n }\n ptbl$Group=gsub('_[1-9]+$','',ptbl$Var2)\n \n \n ####################################\n ## Create Annotation Box Plot by sample\n \n \n # ptblBox$Var1=factor(ptblBox$Var1,levels=ordr)\n ptblBox$value=round(ptblBox$value,1)\n ptblBox$PerValue=paste0(ptblBox$value,'%')\n ptblBox$PerValue=gsub('^\\\\%$',\"_\",ptblBox$PerValue)\n ptblBox[ptblBox$value<1,'PerValue']=\"\"\n \n ptblBox[,GroupCol]=factor(ptblBox[,GroupCol],levels=group_order)\n \n p2_Box=ptblBox%>%ggboxplot(y = 'value', x = GroupCol,add = \"jitter\",color = \"Var1\") + \n # stat_compare_means(method = \"t.test\")+\n facet_wrap(~Var1,ncol = wrap_ncols,scales = 'fixed')+ylab(ylab)+xlab(\"\")+\n theme(legend.title=element_blank())\n \n \n return(list('Plots'=list('Barplot'=p2,'Boxplot'=p2_Box),\"Table\"=ColTables$OutTable))\n \n \n}", + "columns": [ + { + "key": "Annotation_Column", + "displayName": "Annotation Column", + "description": "Column to summarize in Barplot", + "paramGroup": "Basic", + "sourceDataset": "Metadata_Table", + "defaultValue": null, + "columnType": "STRING", + "isMulti": null + }, + { + "key": "Group_Column", + "displayName": "Group Column", + "description": "Column to split Barplot into separate populations to compare", + "paramGroup": "Basic", + "sourceDataset": "Metadata_Table", + "defaultValue": null, + "columnType": "STRING", + "isMulti": null + }, + { + "key": "Sample_Column", + "displayName": "Sample Column", + "description": "Column in Seurat Metadata that contains sample names", + "paramGroup": "Basic", + "sourceDataset": "Metadata_Table", + "defaultValue": null, + "columnType": "STRING", + "isMulti": null + } + ], + "condaDependencies": [], + "description": "", + "externalId": "Compare_Cell_Populations", + "inputDatasets": [ + { + "key": "Object", + "displayName": "Object", + "description": "Seurat object", + "paramGroup": null, + "anchorDataset": false, + "dataType": "R_TRANSFORM_INPUT", + "tags": [] + }, + { + "key": "Metadata_Table", + "displayName": "Metadata Table", + "description": "Seurat Metadata table", + "paramGroup": null, + "anchorDataset": false, + "dataType": "R_NATIVE_DATAFRAME", + "tags": [] + } + ], + "vectorLanguage": "R", + "codeLanguage": "R", + "parameters": [ + { + "key": "Counts_Type", + "displayName": "Counts Type", + "description": "What type of data do you want to plot", + "paramType": "SELECT", + "paramGroup": "Basic", + "paramValues": [ + "Frequency", + "Counts" + ], + "defaultValue": "Frequency", + "condition": null, + "content": null, + "objectPropertyReference": null + }, + { + "key": "Seurat_Object_Filename", + "displayName": "Seurat Object Filename", + "description": "", + "paramType": "STRING", + "paramGroup": "Filesave", + "paramValues": null, + "defaultValue": "seurat_object.rds", + "condition": null, + "content": null, + "objectPropertyReference": null + }, + { + "key": "Group_Order", + "displayName": "Group Order", + "description": "", + "paramType": "VECTOR", + "paramGroup": "Basic", + "paramValues": null, + "defaultValue": "c(\"\")", + "condition": null, + "content": null, + "objectPropertyReference": null + } + ], + "title": "Compare Cell Populations", + "templateApiVersion": "0.1.0" +} \ No newline at end of file diff --git a/man/compareCellPopulations.Rd b/man/compareCellPopulations.Rd new file mode 100644 index 0000000..4cf427c --- /dev/null +++ b/man/compareCellPopulations.Rd @@ -0,0 +1,88 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/Compare_Cell_Populations.R +\name{compareCellPopulations} +\alias{compareCellPopulations} +\title{Compare Cell Populations} +\usage{ +compareCellPopulations( + object, + metadata.table, + annotation.column, + group.column, + sample.column = "orig.ident", + counts.type = "Frequency", + group.order = NULL, + seurat.object.filename = "seurat_object.rds", + wrap.ncols = 5 +) +} +\arguments{ +\item{object}{A Seurat object containing the single-cell data} + +\item{metadata.table}{A data.frame containing metadata (typically from +Seurat object's meta.data slot)} + +\item{annotation.column}{Character string specifying the metadata column +containing cell type annotations to summarize in the bar plot} + +\item{group.column}{Character string specifying the metadata column +defining groups to compare (e.g., treatment conditions)} + +\item{sample.column}{Character string specifying the metadata column +containing sample identifiers. Default is "orig.ident"} + +\item{counts.type}{Character string specifying plot data type: +"Frequency" (percentages) or "Counts" (absolute numbers). Default is "Frequency"} + +\item{group.order}{Character vector specifying the order of groups in plots. +If NULL, uses natural order from data. Default is NULL} + +\item{seurat.object.filename}{Character string for the Seurat object +filename. Default is "seurat_object.rds"} + +\item{wrap.ncols}{Integer specifying number of columns for facet wrapping +in box plots. Default is 5} +} +\value{ +A list containing: +\itemize{ +\item \code{Plots} - A list with two ggplot objects: +\itemize{ +\item \code{Barplot} - Stacked bar plot with alluvial flows +\item \code{Boxplot} - Faceted box plots by cell type (only if counts.type="Frequency") +} +\item \code{Table} - A data.frame with cell counts and percentages +} +} +\description{ +Compare cell population distributions across different groups +using bar plots and box plots. Creates visualizations showing cell type +frequencies or counts across user-defined groupings. +} +\details{ +This function generates comparative visualizations of cell +populations from a Seurat object. It can display data as either frequency +percentages or absolute counts, and creates both stacked bar plots +(with alluvial flow connections) and grouped box plots for comparison +across samples and conditions. +} +\examples{ +\dontrun{ +# Compare cell populations by treatment group +results <- compareCellPopulations( + object = seurat_obj, + metadata.table = seurat_obj@meta.data, + annotation.column = "cell_type", + group.column = "treatment", + sample.column = "sample_id", + counts.type = "Frequency" +) + +# Display plots +plot(results$Plots$Barplot) +plot(results$Plots$Boxplot) + +# View summary table +head(results$Table) +} +} diff --git a/man/violinPlot_mod.Rd b/man/violinPlot_mod.Rd index df2fcd5..ee42877 100644 --- a/man/violinPlot_mod.Rd +++ b/man/violinPlot_mod.Rd @@ -7,7 +7,7 @@ violinPlot_mod( object, assay, - slot, + layer, genes, group, facet_by = "", @@ -23,7 +23,7 @@ violinPlot_mod( \item{assay}{Assay to extract gene expression data from (Default: SCT)} -\item{slot}{Slot to extract gene expression data from (Default: scale.data)} +\item{layer}{Slot to extract gene expression data from (Default: scale.data)} \item{genes}{Genes to visualize on the violin plot} @@ -49,7 +49,7 @@ Create violin plot of gene expression data across groups } \details{ Takes in a list of genes inputted by the user, displays violin plots -of genes across groups from a slot-assay with (optional) outliers +of genes across groups from a layer-assay with (optional) outliers removed. Can also choose to scale or transform expression data. } \examples{ @@ -57,7 +57,7 @@ removed. Can also choose to scale or transform expression data. violinPlot_mod( object = seurat, assay = "SCT", - slot = "data", + layer = "data", genes = c("Cd4", "Cd8a"), group = "celltype", facet_by = "orig.ident", diff --git a/tests/testthat/helper-Compare_Cell_Populations.R b/tests/testthat/helper-Compare_Cell_Populations.R new file mode 100644 index 0000000..7699a99 --- /dev/null +++ b/tests/testthat/helper-Compare_Cell_Populations.R @@ -0,0 +1,71 @@ +# Helper functions for Compare_Cell_Populations tests + +# Load real Seurat objects from fixtures +getParamCCP <- function(data) { + if (data == "TEC") { + object <- selectCRObject("TEC") + annotation.column <- "seurat_clusters" + group.column <- "Status" + sample.column <- "orig.ident" + counts.type <- "Frequency" + group.order <- NULL + + } else if (data == "Chariou") { + object <- selectCRObject("Chariou") + annotation.column <- "seurat_clusters" + group.column <- "Status" + sample.column <- "orig.ident" + counts.type <- "Frequency" + group.order <- NULL + + } else if (data == "PBMC") { + object <- selectSRObject("pbmc-single") + annotation.column <- "HPCA_main" + group.column <- "Phase" + sample.column <- "orig.ident" + counts.type <- "Frequency" + group.order <- NULL + + } else if (data == "NSCLC") { + object <- selectCRObject("nsclc-multi") + annotation.column <- "seurat_clusters" + group.column <- "Phase" + sample.column <- "orig.ident" + counts.type <- "Frequency" + group.order <- NULL + + } else if (data == "BRCA") { + object <- selectCRObject("BRCA") + annotation.column <- "seurat_clusters" + group.column <- "Phase" + sample.column <- "orig.ident" + counts.type <- "Frequency" + group.order <- NULL + } + + return( + list( + "object" = object, + "metadata.table" = object@meta.data, + "annotation.column" = annotation.column, + "group.column" = group.column, + "sample.column" = sample.column, + "counts.type" = counts.type, + "group.order" = group.order + ) + ) +} + +# Helper function to save ggplot objects for snapshot testing +.drawCCPFig <- function(x, width = 10, height = 10) { + path <- tempfile(fileext = ".png") + ggplot2::ggsave(path, x, width = width, height = height) + path +} + +# Helper function to save data tables for snapshot testing +.saveCCPTable <- function(x) { + path <- tempfile(fileext = ".rds") + saveRDS(x, file = path) + path +} diff --git a/tests/testthat/test-Compare_Cell_Populations.R b/tests/testthat/test-Compare_Cell_Populations.R new file mode 100644 index 0000000..d13dbdf --- /dev/null +++ b/tests/testthat/test-Compare_Cell_Populations.R @@ -0,0 +1,211 @@ +# Test 1: Standard parameters - TEC dataset +test_that("compareCellPopulations returns correct structure with TEC data", { + params <- getParamCCP("TEC") + result <- do.call(compareCellPopulations, params) + + # Check result structure + expect_type(result, "list") + expect_named(result, c("Plots", "Table")) + expect_named(result$Plots, c("Barplot", "Boxplot")) + + # Check plot types + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + # Check table structure + expect_true(is.data.frame(result$Table)) + expect_true("Clusters" %in% colnames(result$Table)) + + # Snapshot tests for plots and table + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Barplot), + "TEC_Standard_Barplot.png" + ) + expect_snapshot_file( + .drawCCPFig(result$Plots$Boxplot), + "TEC_Standard_Boxplot.png" + ) + expect_snapshot_file( + .saveCCPTable(result$Table), + "TEC_Standard_Table.rds" + ) +}) + +# Test 2: Standard parameters - Chariou dataset +test_that("compareCellPopulations works with Chariou data", { + params <- getParamCCP("Chariou") + result <- do.call(compareCellPopulations, params) + + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Barplot), + "Chariou_Standard_Barplot.png" + ) + expect_snapshot_file( + .drawCCPFig(result$Plots$Boxplot), + "Chariou_Standard_Boxplot.png" + ) +}) + +# Test 3: Standard parameters - PBMC dataset with annotated cell types +test_that("compareCellPopulations works with PBMC annotated cell types", { + params <- getParamCCP("PBMC") + result <- do.call(compareCellPopulations, params) + + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + skip_on_ci() + # Note: PBMC has only one sample, which creates issues with alluvial flow visualization + # Skip barplot snapshot for single-sample dataset + expect_snapshot_file( + .drawCCPFig(result$Plots$Boxplot), + "PBMC_Standard_Boxplot.png" + ) +}) + +# Test 4: Standard parameters - NSCLC dataset +test_that("compareCellPopulations works with NSCLC multi data", { + params <- getParamCCP("NSCLC") + result <- do.call(compareCellPopulations, params) + + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Barplot), + "NSCLC_Standard_Barplot.png" + ) + expect_snapshot_file( + .drawCCPFig(result$Plots$Boxplot), + "NSCLC_Standard_Boxplot.png" + ) +}) + +# Test 5: Standard parameters - BRCA dataset +test_that("compareCellPopulations works with BRCA data", { + params <- getParamCCP("BRCA") + result <- do.call(compareCellPopulations, params) + + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Barplot), + "BRCA_Standard_Barplot.png" + ) + expect_snapshot_file( + .drawCCPFig(result$Plots$Boxplot), + "BRCA_Standard_Boxplot.png" + ) +}) + +# Test 6: Counts type parameter - TEC dataset +test_that("compareCellPopulations works with Counts type on TEC data", { + params <- getParamCCP("TEC") + params$counts.type <- "Counts" + + result <- do.call(compareCellPopulations, params) + + # Check result structure + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Barplot), + "TEC_Counts_Barplot.png" + ) +}) + +# Test 7: Custom group order - Chariou dataset +test_that("compareCellPopulations handles custom group order on Chariou", { + params <- getParamCCP("Chariou") + params$group.order <- c("1", "0") # Status levels + + result <- do.call(compareCellPopulations, params) + + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Barplot), + "Chariou_CustomOrder_Barplot.png" + ) +}) + +# Test 8: Custom wrap columns - PBMC dataset +test_that("compareCellPopulations handles custom wrap.ncols on PBMC", { + params <- getParamCCP("PBMC") + params$wrap.ncols <- 3 # Change from default 5 to 3 columns + + result <- do.call(compareCellPopulations, params) + + expect_type(result, "list") + expect_s3_class(result$Plots$Barplot, "gg") + expect_s3_class(result$Plots$Boxplot, "gg") + + skip_on_ci() + expect_snapshot_file( + .drawCCPFig(result$Plots$Boxplot), + "PBMC_CustomWrap_Boxplot.png" + ) +}) + +# Test 9: Input validation - non-Seurat object +test_that("compareCellPopulations validates input object", { + expect_error( + compareCellPopulations( + object = list(), + metadata.table = data.frame(), + annotation.column = "cell_type", + group.column = "treatment" + ), + "must be a Seurat object" + ) +}) + +# Test 10: Missing column validation - TEC dataset +test_that("compareCellPopulations validates metadata columns on TEC", { + params <- getParamCCP("TEC") + params$annotation.column <- "nonexistent_column" + + expect_error( + do.call(compareCellPopulations, params), + "missing from metadata" + ) +}) + +# Test 11: Invalid counts.type parameter +test_that("compareCellPopulations validates counts.type parameter", { + params <- getParamCCP("TEC") + params$counts.type <- "Invalid" + + expect_error( + do.call(compareCellPopulations, params), + "must be either 'Frequency' or 'Counts'" + ) +}) + +# Test 12: Table output validation - BRCA dataset +test_that("compareCellPopulations table contains expected columns on BRCA", { + params <- getParamCCP("BRCA") + result <- do.call(compareCellPopulations, params) + + # Check for _CellCounts and _Percent suffixed columns + expect_true(any(grepl("_CellCounts$", colnames(result$Table)))) + expect_true(any(grepl("_Percent$", colnames(result$Table)))) +}) + From ec5d4852a6ddef0c23bc318a3a95d632eed9dc01 Mon Sep 17 00:00:00 2001 From: phoman14 Date: Tue, 17 Feb 2026 14:26:53 -0500 Subject: [PATCH 02/42] fixing ModScoreHelpers_function. Error with Roxygen Format --- R/ModuleScoreHelpers.R | 10 ++++++---- R/ModuleScoreHelpers_011726.R | 10 ++++++---- man/modScore.Rd | 1 + man/modscore-imports-011726.Rd | 9 +++++++++ man/modscore-imports.Rd | 9 +++++++++ 5 files changed, 31 insertions(+), 8 deletions(-) create mode 100644 man/modscore-imports-011726.Rd create mode 100644 man/modscore-imports.Rd diff --git a/R/ModuleScoreHelpers.R b/R/ModuleScoreHelpers.R index 539b6d9..462e582 100644 --- a/R/ModuleScoreHelpers.R +++ b/R/ModuleScoreHelpers.R @@ -1,12 +1,14 @@ #' @title Helpers for ModuleScore Shiny app #' @description Precompute module scores per celltype and build plots from cached data. +#' @name modscore-imports #' @keywords internal #' @importFrom dplyr mutate group_by summarise arrange select #' @importFrom ggplot2 ggplot aes theme_bw theme element_blank -#' @importFrom geom_point scale_color_gradientn guides guide_legend -#' @importFrom xlab ylab element_text geom_violin theme_classic geom_hline -#' @importFrom scale_y_continuous geom_line geom_segment scale_y_log10 scale_x_continuous -#' @importFrom gridExtra arrangeGrob grid textGrob gpar +#' @importFrom ggplot2 geom_point scale_color_gradientn guides guide_legend +#' @importFrom ggplot2 xlab ylab element_text geom_violin theme_classic geom_hline +#' @importFrom ggplot2 scale_y_continuous geom_line geom_segment scale_y_log10 scale_x_continuous +#' @importFrom gridExtra arrangeGrob +#' @importFrom grid textGrob gpar NULL #' @export diff --git a/R/ModuleScoreHelpers_011726.R b/R/ModuleScoreHelpers_011726.R index 539b6d9..6a76384 100644 --- a/R/ModuleScoreHelpers_011726.R +++ b/R/ModuleScoreHelpers_011726.R @@ -1,12 +1,14 @@ #' @title Helpers for ModuleScore Shiny app #' @description Precompute module scores per celltype and build plots from cached data. +#' @name modscore-imports-011726 #' @keywords internal #' @importFrom dplyr mutate group_by summarise arrange select #' @importFrom ggplot2 ggplot aes theme_bw theme element_blank -#' @importFrom geom_point scale_color_gradientn guides guide_legend -#' @importFrom xlab ylab element_text geom_violin theme_classic geom_hline -#' @importFrom scale_y_continuous geom_line geom_segment scale_y_log10 scale_x_continuous -#' @importFrom gridExtra arrangeGrob grid textGrob gpar +#' @importFrom ggplot2 geom_point scale_color_gradientn guides guide_legend +#' @importFrom ggplot2 xlab ylab element_text geom_violin theme_classic geom_hline +#' @importFrom ggplot2 scale_y_continuous geom_line geom_segment scale_y_log10 scale_x_continuous +#' @importFrom gridExtra arrangeGrob +#' @importFrom grid textGrob gpar NULL #' @export diff --git a/man/modScore.Rd b/man/modScore.Rd index 824f921..c96c277 100644 --- a/man/modScore.Rd +++ b/man/modScore.Rd @@ -7,6 +7,7 @@ modScore( object, marker.table, + group_var = "orig.ident", use_columns, ms_threshold, general.class, diff --git a/man/modscore-imports-011726.Rd b/man/modscore-imports-011726.Rd new file mode 100644 index 0000000..344bac7 --- /dev/null +++ b/man/modscore-imports-011726.Rd @@ -0,0 +1,9 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/ModuleScoreHelpers_011726.R +\name{modscore-imports-011726} +\alias{modscore-imports-011726} +\title{Helpers for ModuleScore Shiny app} +\description{ +Precompute module scores per celltype and build plots from cached data. +} +\keyword{internal} diff --git a/man/modscore-imports.Rd b/man/modscore-imports.Rd new file mode 100644 index 0000000..20de3c0 --- /dev/null +++ b/man/modscore-imports.Rd @@ -0,0 +1,9 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/ModuleScoreHelpers.R +\name{modscore-imports} +\alias{modscore-imports} +\title{Helpers for ModuleScore Shiny app} +\description{ +Precompute module scores per celltype and build plots from cached data. +} +\keyword{internal} From c3c6152c559d6b0da92791d732544624260d78eb Mon Sep 17 00:00:00 2001 From: phoman14 Date: Tue, 17 Feb 2026 14:52:38 -0500 Subject: [PATCH 03/42] Update Vignette --- docs/CHANGELOG.html | 15 +- docs/LICENSE-text.html | 23 +- docs/articles/SCWorkflow-DEG.html | 2 +- docs/articles/SCWorkflow-Overview.html | 292 ------------------------- docs/authors.html | 4 +- docs/index.html | 40 +++- docs/pkgdown.yml | 2 +- docs/reference/colorByMarkerTable.html | 21 ++ docs/reference/index.html | 18 ++ docs/reference/modScore.html | 25 +++ docs/reference/object.html | 13 ++ docs/reference/violinPlot_mod.html | 86 +++----- docs/search.json | 2 +- docs/sitemap.xml | 9 +- vignettes/SCWorkflow-QC.Rmd | 2 +- vignettes/SCWorkflow-Usage.Rmd | 2 +- 16 files changed, 180 insertions(+), 376 deletions(-) delete mode 100644 docs/articles/SCWorkflow-Overview.html diff --git a/docs/CHANGELOG.html b/docs/CHANGELOG.html index cdf2cd9..d7caeea 100644 --- a/docs/CHANGELOG.html +++ b/docs/CHANGELOG.html @@ -44,6 +44,19 @@

CHANGELOG

+
+

v1.0.3 (in development)

+
+

Feature

+
  • feat: Add compareCellPopulations() function for comparing cell population distributions across experimental groups +
    • Visualizes cell population frequencies or absolute counts across multiple groups
      • +
    • Generates alluvial flow bar plots and faceted box plots
      • +
    • Supports custom group ordering and color palettes
      • +
    • Added ggalluvial dependency for flow visualizations
      • +
    • Generated from JSON template using json2r.prompt.md instructions
      • +
    • +
+

v1.0.2 (2024-02-01)

@@ -74,7 +87,7 @@

Documentation8b5cc98)

-

Feature

+

Feature

  • feat: Update test-annotation to supress warnings (3d5cf8f)

  • feat: test (4c4cee7)

  • feat: test (c8274f9)

    • diff --git a/docs/LICENSE-text.html b/docs/LICENSE-text.html index 5519705..f0f2a04 100644 --- a/docs/LICENSE-text.html +++ b/docs/LICENSE-text.html @@ -42,27 +42,8 @@

    License

-
MIT License
-
-Copyright (c) 2024 NIDAP Community
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
+
YEAR: 2024
+COPYRIGHT HOLDER: NIDAP Community
diff --git a/docs/articles/SCWorkflow-DEG.html b/docs/articles/SCWorkflow-DEG.html index 96afdea..84fc771 100644 --- a/docs/articles/SCWorkflow-DEG.html +++ b/docs/articles/SCWorkflow-DEG.html @@ -208,7 +208,7 @@

Aggregate Seurat Counts 
-aggregateCounts(object=so,
+aggregateCounts(object=so,
                var.group=var_group,
                slot=slot)

diff --git a/docs/articles/SCWorkflow-Overview.html b/docs/articles/SCWorkflow-Overview.html deleted file mode 100644 index 3336cb2..0000000 --- a/docs/articles/SCWorkflow-Overview.html +++ /dev/null @@ -1,292 +0,0 @@ - - - - - - - -Import Data and Quality Control • SCWorkflow - - - - - - - - - - Skip to contents - - -
- - - - -
-
- -

Import Data and Quality Control

- - - -
SCWorkflow-Overview.Rmd
-
- - - -


-
-

Process Input Data -

-

This package is designed to work with the general Seurat Workflow[1]. -To begin using the SCWorkflow tools you will have to process the h5 -files generated by the Cell Ranger[Reference] software from the 10x -genomics platform to create a list of Seurat Objects corresponding to -each h5 file. A Seurat Object is the basic data structure for Seurat -Single Cell analysis

-

This tool supports standard scRNAseq, CITE-Seq, and TCR-Seq assays. -Samples prepared with a cell hashing protocol (HTOs) can also be -processed to produce a Seurat Object split by the corresponding -experimental design strategy. h5 files containing multiple samples can -also be processed to create Seurat objects that will be split based on -the values in the orig.ident column.

-

A corresponding Metadata table can be used to add sample level -information to the Seurat object. The table format should have Sample -names in the first Column and any sample metadata in additional columns. -The Metadata table can also be used to rename samples by including an -alternative sample name Column in the metadata table.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Sample_NameRenameTreatment
SCAF1713_1_1PBSWT
SCAF1714_2_1ENTEntinostat
SCAF1715_3_1NHSIL12NHS-IL12
SCAF1716_4_1ComboEntinostat + NHS-IL12
SCAF1717_5_1CD8depEntinostat + NHS-IL12
-

Samples can also be excluded from the final Seurat object using a -REGEX strategy to identify the samples to be included/excluded. explain -based on newnames

-

The final Seurat Object will contain an assay slot with log2 -normalized counts. QC figures for individual samples will also be -produced to help evaluate samples quality.

-
-
-SampleMetadataTable <- read.table(file = "./images/Sample_Metadata.txt",  sep = '\t',header = T)
-files=list.files(path="../tests/testthat/fixtures/Chariou/h5files",full.names = T)
-
-SOlist=processRawData(input=files,
-               organism="Mouse",
-               sample.metadata.table=SampleMetadataTable,
-               sample.name.column='Sample_Name',
-               rename.col='Rename',
-               keep=T,
-               file.filter.regex=c(),
-               split.h5=F,
-               cell.hash=F,
-               do.normalize.data=T                
-)
-

-
-
-

Filter Low Quality Cells -

-

This function will filter genes and cells based on multiple metrics -available in the Seurat Object metadata slot. A detailed guide for -single cell quality filtering can be found from Xi and Li, 2021 [2]. -First, genes can be filtered by setting the minimum number of cells -needed to keep a gene or removing VDJ Add descriptiopn of VDJ genes. -Next, cells can be filtered by setting thresholds for each individual -metric. Cells that do not meet any of the designated criteria will be -removed from the final filtered Seurat Object . Filter limits can be set -by using absolute values or median absolute deviations (MADs) for each -criteria. If both absolute and MAD values are set for a single filter, -the least extreme value (i.e. the lowest value for upper limits or the -highest value for lower limits) will be selected. The filter values used -for each metric will be printed in the log output. All filters have -default values and can be turned off by setting limits to NA.

-

The individual filtering criteria used in this tool are listed -below.

-
    -
  1. The total number of molecules detected within each cell -(nCount_RNA)
    2. -
  2. The number of genes detected in each cell (nFeature_RNA)
    3. -
  3. The complexity of of genes ( -log10(nFeature_RNA)/log10(nCount_RNA)
    4. -
  4. Percent of mitochondrial Genes
    5. -
  5. Percent counts in top 20 Genes
    6. -
  6. Doublets calculated by scDblFinder (using package default -parameters) [3]
    7. -
-

The function will return a filtered Seurat Object and various figures -showing metrics before and after filtering. These figures can be used to -help evaluate the effects of filtering criteria and whether filtering -limits need to be adjusted.

-
-
-SO_filtered=filterQC(object=SOlist$object,
-                     ## Filter Genes
-                     min.cells = 20,
-                     filter.vdj.genes=F,
-                     
-                     ## Filter Cells
-                     nfeature.limits=c(NA,NA),
-                     mad.nfeature.limits=c(5,5),
-                     ncounts.limits=c(NA,NA),
-                     mad.ncounts.limits=c(5,5),
-                     mitoch.limits = c(NA,25),
-                     mad.mitoch.limits = c(NA,3),
-                     complexity.limits = c(NA,NA),
-                     mad.complexity.limits = c(5,NA),
-                     topNgenes.limits = c(NA,NA),
-                     mad.topNgenes.limits = c(5,5),
-                     n.topgnes=20,
-                     do.doublets.fitler=T
-                                 
-            )
-

-

-
-
-

Combine, Normalize, and Cluster Data -

-

This functions combines multiple sample level Seurat Objects into a -single Seurat Object and normalizes the combined dataset. The -multi-dimensionality of the data will be summarized into a set of -“principal components” and visualized in both UMAP and tSNE projections. -A graph-based clustering approach will identify cell clusters with in -the data.

-
-
-Comb_SO=combineNormalize(
-                         object=SO_filtered$object,
-                     # Nomralization variables
-                         npcs = 21,
-                         SCT.level="Merged",
-                         vars.to.regress = c("percent.mt"),
-                     # FindVariableFeatures
-                         nfeatures = 2000,
-                         low.cut = 0.1,
-                         high.cut = 8,
-                         low.cut.disp = 1,
-                         high.cut.disp = 100000,
-                         selection.method = 'vst',
-                     # Dim Reduction
-                         only.var.genes = FALSE,
-                         draw.umap = TRUE,
-                         draw.tsne = TRUE,
-                         seed.for.pca = 42,
-                         seed.for.tsne = 1,
-                         seed.for.umap = 42,
-                    # Clustering Varables
-                         clust.res.low = 0.2,
-                         clust.res.high = 1.2,
-                         clust.res.bin = 0.2,
-                    # Select PCs
-                         methods.pca = NULL,
-                         var.threshold = 0.1,
-                         pca.reg.plot = FALSE,
-                         jackstraw = FALSE,
-                         jackstraw.dims=5,
-                    # Other                         
-                         exclude.sample = NULL,
-                         cell.count.limit= 35000,
-                         reduce.so = FALSE,
-                         project.name = 'scRNAProject',
-                         cell.hashing.data = FALSE
-)
-

-



1. Hao Y et al. Integrated analysis of multimodal -single-cell data. Cell. 2021 Jun 24;184(13):3573-3587.e29. doi: -10.1016/j.cell.2021.04.048. Epub 2021 May 31. PMID: 34062119; PMCID: -PMC8238499. 2. Heumos, L., Schaar, A.C., Lance, C. et al. Best practices -for single-cell analysis across modalities. Nat Rev Genet (2023). https://doi.org/10.1038/s41576-023-00586-w 3. Germain P, -Lun A, Macnair W, Robinson M (2021). “Doublet identification in -single-cell sequencing data using scDblFinder.” f1000research. doi:10.12688/f1000research.73600.1.

-
-
-
- - - -
-

Created by the .

-
- -
-

Site built with pkgdown 2.1.3.

-
- -
-
- - - - - - - diff --git a/docs/authors.html b/docs/authors.html index 5f19539..538b72b 100644 --- a/docs/authors.html +++ b/docs/authors.html @@ -78,14 +78,14 @@

Authors

Citation

-

Cam M, Meyer T, Bian J, Michalowski A, Lobanov A, Homan P, He R (2025). +

Cam M, Meyer T, Bian J, Michalowski A, Lobanov A, Homan P, He R (2026). SCWorkflow: SCWorkflow from NIDAP. R package version 1.0.2. 

@Manual{,
   title = {SCWorkflow: SCWorkflow from NIDAP},
   author = {Maggie Cam and Thomas Meyer and Jing Bian and Alexandra Michalowski and Alexei Lobanov and Philip Homan and Rui He},
-  year = {2025},
+  year = {2026},
   note = {R package version 1.0.2},
 }
diff --git a/docs/index.html b/docs/index.html index aff5c7d..ac0f806 100644 --- a/docs/index.html +++ b/docs/index.html @@ -68,10 +68,47 @@



The Single Cell Workflow streamlines the analysis of multimodal Single Cell RNA-Seq data produced from 10x Genomics. It can be run in a docker container, and for biologists, in user-friendly web-based interactive notebooks (NIDAP, Palantir Foundry). Much of it is based on the Seurat workflow in Bioconductor, and supports CITE-Seq data. It incorporates a cell identification step (ModScore) that utilizes module scores obtained from Seurat and also includes Harmony for batch correction.

+
+

Key Functions +

+
+

Sequential Workflow +

+
    +
  1. +processRawData() - Process H5 files into Seurat objects
    2. +
  2. +filterQC() - Quality control and filtering
    +
    3. +
  3. +combineNormalize() - Merge samples, normalize, dimension reduction
    4. +
  4. +Harmony integration (optional) - Batch correction
    5. +
  5. +annotateCellTypes() - Automatic cell type annotation via SingleR
    6. +
+
+
+

Analysis & Visualization +

+
    +
  • +compareCellPopulations() - Compare cell population distributions across groups
    • +
  • +degGeneExpressionMarkers() - Differential expression analysis
    • +
  • +reclusterSeuratObject() / reclusterFilteredSeuratObject() - Subset and re-cluster
    • +
  • +colorByGene(), heatmapSC(), violinPlot_mod() - Visualization functions
    • +
  • +plotMetadata(), dotPlotMet() - Metadata visualization
    • +


For further documentation see our detailed Docs Website



Future Developments include addition of support for multiomics (TCR-Seq, ATAC-Seq) single cell data and integration with spatial transcriptomics data.

+
+