diff --git a/DESCRIPTION b/DESCRIPTION
index de1cc47ca88da1aecff1e86bbbdc1d0406d43eba..fac926bb5a3ea04db89510c6035123c60265c151 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -10,7 +10,7 @@ Version: 0.1.0
BugReports: https://git.scicore.unibas.ch/ovca-research/drugsens/-/issues
SystemRequirements: QuPathâ„¢ 4.0.0 or higher
Authors@R: c(
- person("Flavio", "Lombardo", , "flavio.lombardo@unibas.ch", role = c("aut", "cre", "cph")),
+ person("Flavio", "Lombardo", , "flavio.lombardo@unibas.ch", role = c("aut", "cre", "cph"), comment = c(ORCID = "0000-0002-4853-6838")),
person("Ricardo Coelho", role = c("cph")),
person("Ovarian Cancer Research", role = c("cph")),
person("University of Basel and University Hospital Basel", role = c("cph"))
@@ -19,16 +19,16 @@ URL: https://git.scicore.unibas.ch/ovca-research/drugsens/
Maintainer: Flavio Lombardo <flavio.lombardo@unibas.ch>
License: MIT + file LICENSE
Imports:
- utils,
dplyr,
- tidyr,
- readr,
- stringr,
- knitr,
ggplot2,
ggpubr,
+ knitr,
roxygen2,
+ stats,
+ stringr,
+ tidyr,
tidyselect,
+ utils,
testthat (>= 3.0.0)
Depends:
R (>= 4.2)
diff --git a/NAMESPACE b/NAMESPACE
index 1bec287b8e48a624996bca6aaa64e8b52d92607d..b73a5960e5d797af505889204edec8d9b92ac989 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -15,7 +15,18 @@ import(roxygen2)
import(testthat)
importFrom(dplyr,filter)
importFrom(dplyr,select)
-importFrom(readr,write_excel_csv)
+importFrom(ggplot2,aes)
+importFrom(ggplot2,aes_string)
+importFrom(ggplot2,element_text)
+importFrom(ggplot2,facet_wrap)
+importFrom(ggplot2,geom_boxplot)
+importFrom(ggplot2,geom_jitter)
+importFrom(ggplot2,geom_violin)
+importFrom(ggplot2,ggplot)
+importFrom(ggplot2,labs)
+importFrom(ggplot2,position_jitter)
+importFrom(ggplot2,stat_summary)
+importFrom(ggplot2,theme)
importFrom(stats,setNames)
importFrom(stringr,str_count)
importFrom(stringr,str_extract)
diff --git a/R/generate_qu_path_script.R b/R/generate_qupath_script.R
similarity index 83%
rename from R/generate_qu_path_script.R
rename to R/generate_qupath_script.R
index f006a2c632230f281f2b216eb85e00c636564af6..487b2548bed216c39e8e3ec4fedc773c3d762e81 100644
--- a/R/generate_qu_path_script.R
+++ b/R/generate_qupath_script.R
@@ -2,13 +2,24 @@
#' @description
#' Generate a useful script to consistently save the output data from QuPath in .csv format following the naming conventions
#' followed during the package development.
-#' @return `script_for_qupath.txt` in local working directory.
+#' @param output_dir Directory where the script should be saved. If NULL, uses tempdir()
+#' @return Invisibly returns the path to the generated script file.
#' @export
#' @examples
#' \dontrun{
+#' # Generate script in a temporary directory
#' generate_qupath_script()
+#'
+#' # Generate script in a specific directory
+#' output_dir <- tempdir()
+#' generate_qupath_script(output_dir = output_dir)
#' }
-generate_qupath_script <- function() {
+generate_qupath_script <- function(output_dir = NULL) {
+ if(is.null(output_dir)) {
+ output_dir <- tempdir()
+ }
+ output_file <- file.path(output_dir, "script_for_qupath.txt")
+
write(
x = paste0('
//This code script was tested with QuPath 4
@@ -53,9 +64,9 @@ def exporter = new MeasurementExporter()
print "Done!"
'),
- file = paste0(path.expand(getwd()), "/script_for_qupath.txt")
+ file = output_file
)
message("You can now take the script and personalize it to your needs")
- message(paste0(Sys.time(), " The script file was generated here: ", getwd(), "/"))
+ message(paste0(Sys.time(), " The script file was generated here: ", output_file, "/"))
message(paste0(Sys.time(), " Please make sure to follow the name convention here proposed, or it might fail to get all the information"))
}
diff --git a/R/get_QC_plots.R b/R/get_QC_plots.R
index 8a5336f3235507a92e5e8bf075c99aa550c3f70e..7bd93af3305a15d7ead8ec4cb90d4037270fa4ad 100644
--- a/R/get_QC_plots.R
+++ b/R/get_QC_plots.R
@@ -1,85 +1,157 @@
#' Plot some QC plots to define that everything ran correctly
#' @description
-#' Plot data to visualize immediate trends
-#' @param .data The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer()) merged data.frame that should be visualized
+#' Plot data to visualize immediate trends. This function expects data that has been processed
+#' through make_count_dataframe() and change_data_format_to_longer() to ensure the correct
+#' data structure for plotting.
+#' @param .data The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer())
+#' merged data.frame that should be visualized
#' @param patient_column_name The PID's column name in the merged data.frame (defaults to "PID")
-#' @param colors A list of colors to supply to personalize the plot, as default 4 colors c("dark green", "red", "orange", "pink")
-#' @param save_plots A Boolean value indicating if the plots should be saved or not, TRUE for saving in the current working directory, FALSE to not. Default is FALSE
-#' @param folder_name A string indicating the name of the folder where to save the plots in case that save_plots = TRUE
+#' @param colors A list of colors to supply to personalize the plot, defaults to c("darkgreen", "red", "orange", "pink")
+#' @param save_plots A Boolean value indicating if the plots should be saved or not (default is FALSE)
+#' @param folder_name A string indicating the name of the folder where to save the plots if save_plots is TRUE
#' @param isolate_a_specific_patient A string indicating the patient name to isolate for single plot case (default is NULL)
#' @param x_plot_var A string indicating the treatment's full name for the QC plots (default is "Treatment_complete")
#' @import ggplot2
#' @import ggpubr
#' @importFrom dplyr filter
-#' @return A `dataframe`/`tibble`.
+#' @return Invisibly returns NULL, but saves plots to disk if save_plots is TRUE
#' @examples
#' \dontrun{
-#' get_QC_plots(longer_format_dataframe, patient_column_name = "PID",
-#' save_plots = TRUE, folder_name = "figures")
+#' # First process example data
+#' example_path <- system.file("extdata/to_merge/", package = "drugsens")
+#' raw_data <- data_binding(path_to_the_projects_folder = example_path)
+#' count_data <- make_count_dataframe(raw_data)
+#' processed_data <- change_data_format_to_longer(count_data)
+#'
+#' # Create and save plots to temporary directory
+#' temp_dir <- file.path(tempdir(), "qc_plots")
+#' get_QC_plots(
+#' processed_data,
+#' save_plots = TRUE,
+#' folder_name = temp_dir
+#' )
+#'
+#' # Create plots for a specific patient
+#' get_QC_plots(
+#' processed_data,
+#' isolate_a_specific_patient = "B39",
+#' save_plots = TRUE,
+#' folder_name = temp_dir
+#' )
#' }
#' @export
get_QC_plots <- function(.data,
patient_column_name = "PID",
colors = c("darkgreen", "red", "orange", "pink"),
save_plots = FALSE,
- folder_name = "figures",
+ folder_name = NULL,
x_plot_var = "Treatment_complete",
isolate_a_specific_patient = NULL) {
- if (!is.null(isolate_a_specific_patient)) .data <- .data[.data[[patient_column_name]] == isolate_a_specific_patient, ]
- if (nrow(.data) < 1) stop("The data cannot be empty")
+ # Input validation
+ if (!is.data.frame(.data)) {
+ stop("Input must be a data frame")
+ }
- # run for every unique PID the QC plot
- for (i in unique(.data[patient_column_name])) {
- message(paste0("Running the QC plot function for PID: ", i))
+ # Check required columns exist
+ required_cols <- c(patient_column_name, x_plot_var, "marker_positivity", "marker_positivity_ratio")
+ missing_cols <- setdiff(required_cols, colnames(.data))
+ if (length(missing_cols) > 0) {
+ stop("Missing required columns: ", paste(missing_cols, collapse = ", "),
+ ". Please ensure data has been processed with make_count_dataframe() and change_data_format_to_longer()")
+ }
- QC_plot <- .data |>
- dplyr::filter(.data[[patient_column_name]] == i) |>
- ggplot(aes(x = .data[[x_plot_var]],
- y = .data$marker_positivity_ratio,
- col = .data$marker_positivity)) +
- geom_boxplot(
- position = position_dodge(width = 1.0),
- ) +
- facet_wrap(~marker_positivity) +
- geom_jitter(width = 0.15) +
- theme_light() +
- labs(title = paste0("Cell marker ratios for PID: ", i), color = "Cell marker") +
- ylab("Percentage of expression marker (marker-positive-cells/total_cell_count)") +
- xlab("Drugs") +
- theme(axis.text.x = element_text(angle = 45, hjust = 1.0)) +
- scale_color_manual(values = colors) +
- stat_summary(
- fun = "median", geom = "pointrange",
- mapping = aes(xend = after_stat(x) - 0.25, yend = after_stat(y)),
- size = 1.5, alpha = 1.0,
- position = position_dodge(width = 1)
- ) +
- stat_summary(
- geom = "line", fun = "median", position = position_dodge(width = 1),
- size = 1, alpha = 0.3, aes(group = marker_positivity)
+ # Filter for specific patient if requested
+ if (!is.null(isolate_a_specific_patient)) {
+ .data <- .data[.data[[patient_column_name]] == isolate_a_specific_patient, ]
+ if (nrow(.data) < 1) {
+ stop("No data found for patient: ", isolate_a_specific_patient)
+ }
+ }
+
+ # Set up output directory if saving plots
+ if (save_plots) {
+ if (is.null(folder_name)) {
+ folder_name <- file.path(tempdir(), "figures")
+ }
+ dir.create(folder_name, showWarnings = FALSE, recursive = TRUE)
+ }
+
+ # Process each patient
+ for (current_pid in unique(.data[[patient_column_name]])) {
+ message("Processing patient: ", current_pid)
+
+ # Filter data for current patient
+ patient_data <- dplyr::filter(.data, .data[[patient_column_name]] == current_pid)
+
+ # Create the plot
+ QC_plot <- ggplot2::ggplot(patient_data,
+ ggplot2::aes(x = .data[[x_plot_var]],
+ y = marker_positivity_ratio,
+ color = marker_positivity)) +
+ ggplot2::geom_boxplot(position = ggplot2::position_dodge(width = 1.0)) +
+ ggplot2::facet_wrap(~marker_positivity) +
+ ggplot2::geom_jitter(width = 0.15) +
+ ggplot2::theme_light() +
+ ggplot2::labs(
+ title = paste0("Cell marker ratios for PID: ", current_pid),
+ color = "Cell marker",
+ y = "Percentage of expression marker (marker-positive-cells/total_cell_count)",
+ x = "Drugs"
) +
- theme(
- axis.title.x = element_blank(),
- plot.title = element_text(hjust = 0.5),
- axis.ticks.x = element_blank(),
- panel.grid = element_blank(),
- strip.background = element_rect(
+ ggplot2::theme(
+ axis.text.x = ggplot2::element_text(angle = 45, hjust = 1.0),
+ axis.title.x = ggplot2::element_blank(),
+ plot.title = ggplot2::element_text(hjust = 0.5),
+ axis.ticks.x = ggplot2::element_blank(),
+ panel.grid = ggplot2::element_blank(),
+ strip.background = ggplot2::element_rect(
colour = "black",
fill = "grey1"
)
+ ) +
+ ggplot2::scale_color_manual(values = colors) +
+ ggplot2::stat_summary(
+ fun = "median",
+ geom = "point",
+ size = 3,
+ position = ggplot2::position_dodge(width = 1)
+ ) +
+ ggplot2::stat_summary(
+ geom = "line",
+ fun = "median",
+ position = ggplot2::position_dodge(width = 1),
+ linewidth = 1,
+ alpha = 0.3,
+ aes(group = marker_positivity)
)
+ # Save the plot if requested
if (save_plots) {
- if (!dir.exists(paths = paste0(getwd(), "/", folder_name, "/"))) dir.create(path = paste0(getwd(), "/", folder_name, "/"), showWarnings = F, recursive = T)
+ plot_filename <- file.path(
+ folder_name,
+ paste0("patients_QC_box_plots_",
+ current_pid,
+ "_median",
+ format(Sys.Date(), "%Y-%m-%d"),
+ ".pdf"
+ )
+ )
- ggsave(QC_plot,
- filename = paste0(folder_name, "/", "patients_QC_box_plots_", i, "_", "median", Sys.Date(), ".pdf"),
+ ggplot2::ggsave(
+ filename = plot_filename,
+ plot = QC_plot,
device = "pdf",
height = 12,
width = 12
)
}
}
- message(paste0("If save_plots = TRUE, the plots will be saved here:", paste0(folder_name, "/", "patients_QC_box_plots_", "median", Sys.Date(), ".pdf")))
+
+ # Final message about plot locations
+ if (save_plots) {
+ message("Plots have been saved in: ", folder_name)
+ }
+
+ invisible(NULL)
}
diff --git a/R/get_QC_plots_and_stats.R b/R/get_QC_plots_and_stats.R
index 4daf74c4c6a2ac93dd9dd24347405dd2329f85e5..35ea9d7016eb89e9044f04bf176ad28cc3e150d8 100644
--- a/R/get_QC_plots_and_stats.R
+++ b/R/get_QC_plots_and_stats.R
@@ -1,37 +1,60 @@
-#' Plot some QC plots for the bound data
+#' Plot QC plots and calculate statistics for bound data
#' @description
-#' This plot can show trends within the dataset and run some basic statistics.
-#'
-#' @param .data The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer()) merged data.frame that should be visualized
-#' @param list_of_columns_to_plot The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer()) merged data.frame that should be visualized
-#' @param save_plots Boolean, TRUE if plots should be saved (default is FALSE)
-#' @param saving_plots_folder String indicating the folder where the plots should be stored (default is "figures")
-#' @param PID_column_name String, indicating the name of the sample to subset (default is "PID")
-#' @param isolate_specific_drug String, indicating if there should be a Treatment specific data subset (default is NULL)
-#' @param isolate_specific_patient String, indicating a spacific sample to plot only (default is NULL)
-#' @param PID_column_name String, indicating the name of the sample to subset (default is "Treatment")
-#' @param save_list_of_plots Boolean, if TRUE returns a named list of all the plots ran (default is TRUE), this can be usefult to isolate specific plots
-#' @param save_plots_in_patient_specific_subfolders Boolean, if TRUE the plots will be saved (if `save_plots` TRUE) in sample specific folders (default is TRUE)
-#' @param fill_color_variable Boolean, String, indicating the name of the variable (discrete) to use for the plot's filling
-#' @param p_height Integer, indicate the plot's height (default is 10 inches)
-#' @param p_width Integer, indicate the plot's width (default is 10 inches)
-#' @param drug_column_name String, indicate the column indicating the Drug/Treament (default is "Treatment")
-#' @import ggplot2
-#' @import ggpubr
-#' @importFrom readr write_excel_csv
-#' @importFrom dplyr filter
-#' @return A `list`/`NULL`.
+#' This function creates quality control plots and calculates basic statistics for microscopy data.
+#' The plots provide visual insights into marker expression patterns and data quality.
+#' @param .data The preprocessed data frame to analyze
+#' @param list_of_columns_to_plot Columns to include in plots. If NULL, all numeric columns are used.
+#' @param save_plots Logical, whether to save plots to files. Defaults to FALSE.
+#' @param saving_plots_folder Directory for saving plots. If NULL and save_plots=TRUE, uses a subdirectory of tempdir().
+#' @param save_plots_in_patient_specific_subfolders Logical, whether to create patient subdirectories. Defaults to TRUE.
+#' @param fill_color_variable Variable name for plot color filling
+#' @param PID_column_name Column name for patient IDs. Defaults to "PID".
+#' @param isolate_specific_drug Drug name to subset data
+#' @param isolate_specific_patient Patient ID to subset data
+#' @param drug_column_name Column name for drug information. Defaults to "Treatment".
+#' @param save_list_of_plots Logical, whether to return list of plot objects. Defaults to TRUE.
+#' @param p_height Plot height in inches. Defaults to 10.
+#' @param p_width Plot width in inches. Defaults to 10.
+#' @param verbose Logical, whether to show progress messages. Defaults to TRUE.
+#' @return If save_list_of_plots=TRUE, returns a named list of ggplot objects. Otherwise returns invisible(NULL).
+#' @importFrom ggplot2 ggplot aes geom_violin geom_boxplot facet_wrap theme element_text labs
+#' geom_jitter position_jitter stat_summary aes_string
#' @examples
#' \dontrun{
-#' qc <- get_QC_plots_parsed_merged_data(bind_data, save_plots = TRUE,
-#' save_list_of_plots = FALSE)
+#' # First load and process example data
+#' example_path <- system.file("extdata/to_merge/", package = "drugsens")
+#' raw_data <- data_binding(path_to_the_projects_folder = example_path)
+#' count_data <- make_count_dataframe(raw_data)
+#' processed_data <- change_data_format_to_longer(count_data)
+#'
+#' # Basic usage - create plots for all patients
+#' plots <- get_QC_plots_parsed_merged_data(processed_data)
+#'
+#' # Save plots to a temporary directory
+#' temp_dir <- file.path(tempdir(), "qc_plots")
+#' plots <- get_QC_plots_parsed_merged_data(
+#' processed_data,
+#' save_plots = TRUE,
+#' saving_plots_folder = temp_dir
+#' )
+#'
+#' # Focus on a specific patient
+#' plots <- get_QC_plots_parsed_merged_data(
+#' processed_data,
+#' isolate_specific_patient = "B39"
+#' )
+#'
+#' # Color plots by tissue type
+#' plots <- get_QC_plots_parsed_merged_data(
+#' processed_data,
+#' fill_color_variable = "Tissue"
+#' )
#' }
#' @export
-
get_QC_plots_parsed_merged_data <- function(.data,
list_of_columns_to_plot = NULL,
save_plots = FALSE,
- saving_plots_folder = "figures",
+ saving_plots_folder = NULL,
save_plots_in_patient_specific_subfolders = TRUE,
fill_color_variable = NULL,
PID_column_name = "PID",
@@ -40,127 +63,153 @@ get_QC_plots_parsed_merged_data <- function(.data,
drug_column_name = "Treatment",
save_list_of_plots = TRUE,
p_height = 10,
- p_width = 10) {
- # List where plots could be stored
- list_plottos <- list()
+ p_width = 10,
+ verbose = TRUE) {
+
+ # Define the helper function for creating individual QC plots
+ create_qc_plot <- function(data, metric, fill_var, pid, drug) {
+ p <- ggplot2::ggplot(data, ggplot2::aes(x = marker_positivity, y = unlist(data[[metric]])))
+
+ if (!is.null(fill_var)) {
+ p <- p + ggplot2::geom_violin(trim = FALSE,
+ ggplot2::aes_string(fill = fill_var),
+ color = NA)
+ } else {
+ p <- p + ggplot2::geom_violin(trim = FALSE,
+ fill = "#A4A4A4",
+ color = "darkred")
+ }
- if (!is.data.frame(.data) | nrow(.data) < 1) stop("ERROR: the data provided must be not empty of dataframe type.")
+ p <- p +
+ theme_minimal() +
+ ggplot2::geom_boxplot(width = 0.1, fill = "white") +
+ ggplot2::facet_wrap(~Treatment) +
+ ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 45, hjust = 1)) +
+ ggplot2::labs(
+ title = metric,
+ x = "Cell Markers",
+ y = paste0("Intensity of ", metric, " (log2)"),
+ subtitle = paste0(pid, ".", drug)
+ ) +
+ ggplot2::geom_jitter(shape = 16,
+ position = ggplot2::position_jitter(0.01)) +
+ ggplot2::stat_summary(geom = "crossbar",
+ fun = mean,
+ colour = "red",
+ width = 0.21)
+
+ return(p)
+ }
- # get the number of possible plotting variables
- if (is.null(list_of_columns_to_plot)) {
- list_of_columns_to_plot <- colnames(.data)[which(sapply(.data, is.numeric))]
+ # Input validation
+ if (!is.data.frame(.data) || nrow(.data) < 1) {
+ stop("Input must be a non-empty data frame")
}
- # check that the fill_color_variable is in the dataset and not null
- if (!is.null(fill_color_variable) & !fill_color_variable %in% colnames(.data)) stop("ERROR: the fill_color_variable must be in the colum names variables.")
+ # Initialize plot storage
+ list_plottos <- list()
- # if the user decides to isolate a specific sample only
- if (!is.null(isolate_specific_patient)) .data <- .data[.data[[PID_column_name]] == isolate_specific_patient, ]
+ # If user requested to isolate specific patient, filter the data
+ if (!is.null(isolate_specific_patient)) {
+ .data <- .data[.data[[PID_column_name]] == isolate_specific_patient, ]
+ if (nrow(.data) < 1) {
+ stop("No data found for specified patient: ", isolate_specific_patient)
+ }
+ }
+
+ # Set up output directory if saving plots
+ if (save_plots) {
+ # Initialize the base directory for plots
+ saving_plots_folder <- if (is.null(saving_plots_folder)) {
+ file.path(tempdir(), "drugsens_plots")
+ } else {
+ saving_plots_folder
+ }
+ dir.create(saving_plots_folder, showWarnings = FALSE, recursive = TRUE)
+ }
+
+ # Determine columns to plot
+ if (is.null(list_of_columns_to_plot)) {
+ list_of_columns_to_plot <- colnames(.data)[sapply(.data, is.numeric)]
+ }
+ # Process each patient
for (pid in unique(.data[[PID_column_name]])) {
+ if (verbose) {
+ message("Processing patient: ", pid)
+ }
+
+ # Subset data for current patient
subset_data <- .data[.data[[PID_column_name]] == pid, ]
- for (i in list_of_columns_to_plot) {
- if (!is.null(isolate_specific_drug)) subset_data <- subset_data[subset_data[[drug_column_name]] %in% isolate_specific_drug, ]
+ # Apply drug filter if specified
+ if (!is.null(isolate_specific_drug)) {
+ subset_data <- subset_data[subset_data[[drug_column_name]] %in% isolate_specific_drug, ]
+ }
- if (nrow(subset_data) < 1) {
- message(unique(subset_data[[PID_column_name]]))
- message(unique(subset_data[[drug_column_name]]))
- stop("ERROR: Your filtering query has returned no observations")
+ # Skip if no data after filtering
+ if (nrow(subset_data) < 1) {
+ if (verbose) {
+ message("No data found for PID: ", pid)
}
+ next
+ }
- # browser()
+ # Create patient directory if needed
+ if (save_plots && save_plots_in_patient_specific_subfolders) {
+ patient_dir <- file.path(saving_plots_folder, pid)
+ dir.create(patient_dir, showWarnings = FALSE, recursive = TRUE)
+ }
+
+ # Process each metric
+ for (i in list_of_columns_to_plot) {
+ # Create plot
+ p <- create_qc_plot(subset_data, i, fill_color_variable,
+ pid, isolate_specific_drug)
- # Function to dynamically add layers to a ggplot object based on conditions
- add_violin_layers <- function(p, fill_color_variable) {
- if (!is.null(fill_color_variable)) {
- p <- p + geom_violin(trim = FALSE, aes_string(fill = fill_color_variable), color = NA) +
- geom_boxplot(width = 0.1, fill = "white")
+ # Save plot if requested
+ if (save_plots) {
+ plot_file <- sprintf("%s_%s_%s_%s.pdf",
+ format(Sys.Date(), "%Y%m%d"),
+ pid,
+ ifelse(is.null(isolate_specific_drug), "all", isolate_specific_drug),
+ make.names(i))
+
+ # Determine save directory
+ save_dir <- if (save_plots_in_patient_specific_subfolders) {
+ file.path(saving_plots_folder, pid)
} else {
- p <- p + geom_violin(trim = FALSE, fill = "#A4A4A4", color = "darkred") +
- geom_boxplot(width = 0.1, fill = "white")
+ saving_plots_folder
}
- return(p)
- }
- # Initialize ggplot
- p <- ggplot(subset_data, aes(x = Name, y = log2(unlist(subset_data[[i]]))))
-
- # Add violin and boxplot layers
- p <- add_violin_layers(p, fill_color_variable)
-
- # More layers on top
- p <- p + facet_wrap(~Treatment) +
- theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
- labs(
- title = colnames(subset_data[i]),
- x = "Cell Markers",
- y = paste0("Intensity of ", colnames(subset_data[i]), " (log2)"),
- subtitle = paste0(pid, ".", isolate_specific_drug)
- ) +
- geom_jitter(shape = 16, position = position_jitter(0.01)) +
- stat_summary(geom = "crossbar", fun = mean, colour = "red", width = 0.21)
-
- # Conditionally add to list of plots
- if (save_list_of_plots) {
- list_plottos[[paste0(isolate_specific_drug, ".", pid, ".", i)]] <- p
- }
+ plot_path <- file.path(save_dir, plot_file)
+ ggplot2::ggsave(plot_path,
+ plot = p,
+ width = p_width,
+ height = p_height,
+ dpi = 600)
- if (save_plots) {
- if (save_plots_in_patient_specific_subfolders) {
-
- if (!dir.exists(paste0(saving_plots_folder, "/", pid))) dir.create(paste0(saving_plots_folder, "/", pid), showWarnings = F, recursive = T)
- ggsave(
- plot = p,
- filename = paste0(
- paste0(saving_plots_folder, "/", pid),
- "/",
- Sys.Date(),
- "_",
- pid,
- ".",
- isolate_specific_drug,
- ".",
- colnames(.data[i]),
- ".pdf"
- ),
- device = "pdf",
- dpi = 600
- )
- } else {
- # Saving plots in .pdf at 600 dpi
- if (!dir.exists(saving_plots_folder)) dir.create(saving_plots_folder, showWarnings = F, recursive = T)
- ggsave(
- plot = p,
- width = p_width,
- height = p_height,
- filename = paste0(
- saving_plots_folder,
- "/",
- Sys.Date(),
- "_",
- pid,
- ".",
- isolate_specific_drug,
- ".",
- colnames(.data[i]),
- ".pdf"
- ),
- device = "pdf",
- dpi = 600,
- )
+ if (verbose) {
+ message("Saved plot to: ", plot_path)
}
+ }
- message(paste0(
- "plots for: ",
- pid,
- ".",
- isolate_specific_drug,
- ".",
- colnames(.data[i]), " saved"
- ))
+ # Store plot if requested
+ if (save_list_of_plots) {
+ plot_name <- paste(pid, i, sep = ".")
+ if (!is.null(isolate_specific_drug)) {
+ plot_name <- paste(isolate_specific_drug, plot_name, sep = ".")
+ }
+ list_plottos[[plot_name]] <- p
}
}
}
+
+ # Return results
+ if (save_list_of_plots) {
+ return(list_plottos)
+ } else {
+ invisible(NULL)
+ }
}
diff --git a/R/make_count_dataframe.R b/R/make_count_dataframe.R
index 2bc835856376b17266125e93ca74bfa9a46cbf24..edd3564065b8ba7f72578b59816426a740215f14 100644
--- a/R/make_count_dataframe.R
+++ b/R/make_count_dataframe.R
@@ -18,7 +18,7 @@
#' plotting_ready_dataframe <- change_data_format_to_longer(
#' counts_dataframe
#' )
-##' make_count_dataframe(
+# make_count_dataframe(
#' data,
#' name_of_the_markers_column = "Name",
#' unique_name_row_identifier = "filter_image"
diff --git a/R/make_run_config.R b/R/make_run_config.R
index ab009a74c835fe90931b0d7eb36972c3eb8dcd61..3ae38e3ec352238a6f1b08a540094de5905c0497 100644
--- a/R/make_run_config.R
+++ b/R/make_run_config.R
@@ -14,7 +14,8 @@
#' make_run_config()
#' }
make_run_config <- function(overwrite_config = FALSE, forcePath = NULL) {
- currentPath <- if (is.null(forcePath)) getwd() else forcePath
+
+ currentPath <- if (is.null(forcePath)) tempdir() else forcePath
config_file <- file.path(currentPath, "config_drugsens.txt")
if (file.exists(config_file)) {
diff --git a/R/parsers.R b/R/parsers.R
index 034cac0ec7c921aa9c13bb3032273755da56eb10..72c6055d08cc7824d21d5659a875acde753b1d1b 100644
--- a/R/parsers.R
+++ b/R/parsers.R
@@ -1,4 +1,4 @@
-#' Main parsing function
+#' @title Parse image filenames to extract metadata
#' @description
#' This function will parse the data from the Image name and will return the metadata there contained
#' The metadata will be then associated to the count file as well
diff --git a/man/generate_qupath_script.Rd b/man/generate_qupath_script.Rd
index 89b18160e079a24ebf48d517e7b513df0efa10ce..2f7f74d87817211de857e2d02c38cb8ee670165c 100644
--- a/man/generate_qupath_script.Rd
+++ b/man/generate_qupath_script.Rd
@@ -1,13 +1,16 @@
% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/generate_qu_path_script.R
+% Please edit documentation in R/generate_qupath_script.R
\name{generate_qupath_script}
\alias{generate_qupath_script}
\title{Generate the groovy script used for the analysis}
\usage{
-generate_qupath_script()
+generate_qupath_script(output_dir = NULL)
+}
+\arguments{
+\item{output_dir}{Directory where the script should be saved. If NULL, uses tempdir()}
}
\value{
-\code{script_for_qupath.txt} in local working directory.
+Invisibly returns the path to the generated script file.
}
\description{
Generate a useful script to consistently save the output data from QuPath in .csv format following the naming conventions
@@ -15,6 +18,11 @@ followed during the package development.
}
\examples{
\dontrun{
+ # Generate script in a temporary directory
generate_qupath_script()
+
+ # Generate script in a specific directory
+ output_dir <- tempdir()
+ generate_qupath_script(output_dir = output_dir)
}
}
diff --git a/man/get_QC_plots.Rd b/man/get_QC_plots.Rd
index 0b4825e4aded1c593f206b2172c7f3b463ef1b1d..14cc7ab84cde079df6b6a4234d5bd0d8a6cf3e5f 100644
--- a/man/get_QC_plots.Rd
+++ b/man/get_QC_plots.Rd
@@ -9,35 +9,57 @@ get_QC_plots(
patient_column_name = "PID",
colors = c("darkgreen", "red", "orange", "pink"),
save_plots = FALSE,
- folder_name = "figures",
+ folder_name = NULL,
x_plot_var = "Treatment_complete",
isolate_a_specific_patient = NULL
)
}
\arguments{
-\item{.data}{The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer()) merged data.frame that should be visualized}
+\item{.data}{The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer())
+merged data.frame that should be visualized}
\item{patient_column_name}{The PID's column name in the merged data.frame (defaults to "PID")}
-\item{colors}{A list of colors to supply to personalize the plot, as default 4 colors c("dark green", "red", "orange", "pink")}
+\item{colors}{A list of colors to supply to personalize the plot, defaults to c("darkgreen", "red", "orange", "pink")}
-\item{save_plots}{A Boolean value indicating if the plots should be saved or not, TRUE for saving in the current working directory, FALSE to not. Default is FALSE}
+\item{save_plots}{A Boolean value indicating if the plots should be saved or not (default is FALSE)}
-\item{folder_name}{A string indicating the name of the folder where to save the plots in case that save_plots = TRUE}
+\item{folder_name}{A string indicating the name of the folder where to save the plots if save_plots is TRUE}
\item{x_plot_var}{A string indicating the treatment's full name for the QC plots (default is "Treatment_complete")}
\item{isolate_a_specific_patient}{A string indicating the patient name to isolate for single plot case (default is NULL)}
}
\value{
-A \code{dataframe}/\code{tibble}.
+Invisibly returns NULL, but saves plots to disk if save_plots is TRUE
}
\description{
-Plot data to visualize immediate trends
+Plot data to visualize immediate trends. This function expects data that has been processed
+through make_count_dataframe() and change_data_format_to_longer() to ensure the correct
+data structure for plotting.
}
\examples{
\dontrun{
- get_QC_plots(longer_format_dataframe, patient_column_name = "PID",
- save_plots = TRUE, folder_name = "figures")
+# First process example data
+example_path <- system.file("extdata/to_merge/", package = "drugsens")
+raw_data <- data_binding(path_to_the_projects_folder = example_path)
+count_data <- make_count_dataframe(raw_data)
+processed_data <- change_data_format_to_longer(count_data)
+
+# Create and save plots to temporary directory
+temp_dir <- file.path(tempdir(), "qc_plots")
+get_QC_plots(
+ processed_data,
+ save_plots = TRUE,
+ folder_name = temp_dir
+)
+
+# Create plots for a specific patient
+get_QC_plots(
+ processed_data,
+ isolate_a_specific_patient = "B39",
+ save_plots = TRUE,
+ folder_name = temp_dir
+)
}
}
diff --git a/man/get_QC_plots_parsed_merged_data.Rd b/man/get_QC_plots_parsed_merged_data.Rd
index de2b2e59456f854836d974661dd327f540b712f0..084f98acc5d53208636c47c22dfffccb29ceb34c 100644
--- a/man/get_QC_plots_parsed_merged_data.Rd
+++ b/man/get_QC_plots_parsed_merged_data.Rd
@@ -2,13 +2,13 @@
% Please edit documentation in R/get_QC_plots_and_stats.R
\name{get_QC_plots_parsed_merged_data}
\alias{get_QC_plots_parsed_merged_data}
-\title{Plot some QC plots for the bound data}
+\title{Plot QC plots and calculate statistics for bound data}
\usage{
get_QC_plots_parsed_merged_data(
.data,
list_of_columns_to_plot = NULL,
save_plots = FALSE,
- saving_plots_folder = "figures",
+ saving_plots_folder = NULL,
save_plots_in_patient_specific_subfolders = TRUE,
fill_color_variable = NULL,
PID_column_name = "PID",
@@ -17,45 +17,75 @@ get_QC_plots_parsed_merged_data(
drug_column_name = "Treatment",
save_list_of_plots = TRUE,
p_height = 10,
- p_width = 10
+ p_width = 10,
+ verbose = TRUE
)
}
\arguments{
-\item{.data}{The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer()) merged data.frame that should be visualized}
+\item{.data}{The preprocessed data frame to analyze}
-\item{list_of_columns_to_plot}{The preprocessed data (after running make_count_dataframe() and change_data_format_to_longer()) merged data.frame that should be visualized}
+\item{list_of_columns_to_plot}{Columns to include in plots. If NULL, all numeric columns are used.}
-\item{save_plots}{Boolean, TRUE if plots should be saved (default is FALSE)}
+\item{save_plots}{Logical, whether to save plots to files. Defaults to FALSE.}
-\item{saving_plots_folder}{String indicating the folder where the plots should be stored (default is "figures")}
+\item{saving_plots_folder}{Directory for saving plots. If NULL and save_plots=TRUE, uses a subdirectory of tempdir().}
-\item{save_plots_in_patient_specific_subfolders}{Boolean, if TRUE the plots will be saved (if \code{save_plots} TRUE) in sample specific folders (default is TRUE)}
+\item{save_plots_in_patient_specific_subfolders}{Logical, whether to create patient subdirectories. Defaults to TRUE.}
-\item{fill_color_variable}{Boolean, String, indicating the name of the variable (discrete) to use for the plot's filling}
+\item{fill_color_variable}{Variable name for plot color filling}
-\item{PID_column_name}{String, indicating the name of the sample to subset (default is "Treatment")}
+\item{PID_column_name}{Column name for patient IDs. Defaults to "PID".}
-\item{isolate_specific_drug}{String, indicating if there should be a Treatment specific data subset (default is NULL)}
+\item{isolate_specific_drug}{Drug name to subset data}
-\item{isolate_specific_patient}{String, indicating a spacific sample to plot only (default is NULL)}
+\item{isolate_specific_patient}{Patient ID to subset data}
-\item{drug_column_name}{String, indicate the column indicating the Drug/Treament (default is "Treatment")}
+\item{drug_column_name}{Column name for drug information. Defaults to "Treatment".}
-\item{save_list_of_plots}{Boolean, if TRUE returns a named list of all the plots ran (default is TRUE), this can be usefult to isolate specific plots}
+\item{save_list_of_plots}{Logical, whether to return list of plot objects. Defaults to TRUE.}
-\item{p_height}{Integer, indicate the plot's height (default is 10 inches)}
+\item{p_height}{Plot height in inches. Defaults to 10.}
-\item{p_width}{Integer, indicate the plot's width (default is 10 inches)}
+\item{p_width}{Plot width in inches. Defaults to 10.}
+
+\item{verbose}{Logical, whether to show progress messages. Defaults to TRUE.}
}
\value{
-A \code{list}/\code{NULL}.
+If save_list_of_plots=TRUE, returns a named list of ggplot objects. Otherwise returns invisible(NULL).
}
\description{
-This plot can show trends within the dataset and run some basic statistics.
+This function creates quality control plots and calculates basic statistics for microscopy data.
+The plots provide visual insights into marker expression patterns and data quality.
}
\examples{
\dontrun{
- qc <- get_QC_plots_parsed_merged_data(bind_data, save_plots = TRUE,
- save_list_of_plots = FALSE)
+# First load and process example data
+example_path <- system.file("extdata/to_merge/", package = "drugsens")
+raw_data <- data_binding(path_to_the_projects_folder = example_path)
+count_data <- make_count_dataframe(raw_data)
+processed_data <- change_data_format_to_longer(count_data)
+
+# Basic usage - create plots for all patients
+plots <- get_QC_plots_parsed_merged_data(processed_data)
+
+# Save plots to a temporary directory
+temp_dir <- file.path(tempdir(), "qc_plots")
+plots <- get_QC_plots_parsed_merged_data(
+ processed_data,
+ save_plots = TRUE,
+ saving_plots_folder = temp_dir
+)
+
+# Focus on a specific patient
+plots <- get_QC_plots_parsed_merged_data(
+ processed_data,
+ isolate_specific_patient = "B39"
+)
+
+# Color plots by tissue type
+plots <- get_QC_plots_parsed_merged_data(
+ processed_data,
+ fill_color_variable = "Tissue"
+)
}
}
diff --git a/man/make_count_dataframe.Rd b/man/make_count_dataframe.Rd
index 256353e95a1e776504baacad5edf2d833426549a..f1aa2bfd7012df9f380397ce7ad2a4393130c4b7 100644
--- a/man/make_count_dataframe.Rd
+++ b/man/make_count_dataframe.Rd
@@ -34,7 +34,6 @@ counts_dataframe <- make_count_dataframe(bind_data)
plotting_ready_dataframe <- change_data_format_to_longer(
counts_dataframe
)
- make_count_dataframe(
data,
name_of_the_markers_column = "Name",
unique_name_row_identifier = "filter_image"
diff --git a/man/string_parsing.Rd b/man/string_parsing.Rd
index 1317c7378cc70d3dc95cac42faf674ef9f6793f8..31fee436dc02935cdb6ff0337c41585ec8beed6a 100644
--- a/man/string_parsing.Rd
+++ b/man/string_parsing.Rd
@@ -2,7 +2,7 @@
% Please edit documentation in R/parsers.R
\name{string_parsing}
\alias{string_parsing}
-\title{Main parsing function}
+\title{Parse image filenames to extract metadata}
\usage{
string_parsing(.data)
}
diff --git a/tests/testthat/test_cases.R b/tests/testthat/test_cases.R
index fbd7cf5187f8e75e49b7beddbac83d685dcd01d9..3aae63c0874ac2163b0a74ec141465e595c5e326 100644
--- a/tests/testthat/test_cases.R
+++ b/tests/testthat/test_cases.R
@@ -1,110 +1,107 @@
test_that("Verify that the internal file in to_merge examples in exdata are available", {
list_files_exdata <- system.file("extdata/to_merge/", package = "drugsens") |> list.files()
- expect_true(length(list_files_exdata) > 3 )
+ expect_true(length(list_files_exdata) > 3)
})
test_that("Verify that the internal file examples in exdata merged are available", {
list_files_exdata <- system.file("extdata/merged/", package = "drugsens") |> list.files()
- expect_true(length(list_files_exdata) >= 1 )
+ expect_true(length(list_files_exdata) >= 1)
})
test_that("list_all_files returns correct file paths", {
# Setup: Create temporary files and directory
- temp_dir <- tempdir()
+ temp_dir <- file.path(tempdir(), "drugsens_test")
+ dir.create(temp_dir, recursive = TRUE, showWarnings = FALSE)
+ on.exit(unlink(temp_dir, recursive = TRUE))
+
file.create(file.path(temp_dir, "file1.csv"))
file.create(file.path(temp_dir, "file2.csv"))
file.create(file.path(temp_dir, "file3.txt"))
file.create(file.path(temp_dir, "file4.tsv"))
- # Test 1
+ # Test CSV files
files_list <- list_all_files(define_path = temp_dir, extension = "\\.csv$",
- recursive_search = T)
+ recursive_search = TRUE)
expect_length(files_list, 2)
expect_true(all(grepl("file1.csv|file2.csv", files_list)))
- # Test 2
+ # Test TXT files
files_list <- list_all_files(define_path = temp_dir, extension = "\\.txt$",
- recursive_search = T)
+ recursive_search = TRUE)
expect_length(files_list, 1)
expect_true(all(grepl("file3.txt", files_list)))
- # Test 3
+ # Test TSV files
files_list <- list_all_files(define_path = temp_dir, extension = "\\.tsv$",
- recursive_search = T)
+ recursive_search = TRUE)
expect_length(files_list, 1)
expect_true(all(grepl("file4.tsv", files_list)))
-
- # remove the dir
- unlink(temp_dir, recursive = TRUE)
-
})
test_that("Config creation and reading works", {
- temp_dir <- tempdir()
- on.exit(unlink(file.path(temp_dir, "config_drugsens.txt")))
+ temp_dir <- file.path(tempdir(), "drugsens_config_test")
+ dir.create(temp_dir, recursive = TRUE, showWarnings = FALSE)
+ on.exit(unlink(temp_dir, recursive = TRUE))
- make_run_config(forcePath = temp_dir)
make_run_config(forcePath = temp_dir)
expect_true(file.exists(file.path(temp_dir, "config_drugsens.txt")))
+
+ # Test re-running doesn't error
+ expect_silent(make_run_config(forcePath = temp_dir))
expect_true(exists("list_of_relabeling"))
})
-test_that("Check that the example file can be read correctly", {
- datas <- drugsens::data_binding(path_to_the_projects_folder = system.file("extdata/to_merge/", package = "drugsens"))
+test_that("Example file can be read correctly", {
+ datas <- drugsens::data_binding(
+ path_to_the_projects_folder = system.file("extdata/to_merge/", package = "drugsens")
+ )
expect_true(exists("datas"))
expect_equal(ncol(datas), expected = 28)
})
-test_that("Check that the drugs combination have two unit and two concentration and control none", {
- datas <- drugsens::data_binding(path_to_the_projects_folder = system.file("extdata/to_merge/", package = "drugsens"))
- expect_true(datas[datas$Treatment == "GentamicinePaclitaxel", "Treatment_complete"][1] == "GentamicinePaclitaxel100uM-10uM" || datas[datas$Treatment == "GentamicinePaclitaxel", "Treatment_complete"][1] == "gentamicinePaclitaxel100uM-10uM")
- expect_true(datas[datas$Treatment == "Control", "Treatment_complete"][1] == "Control" || datas[datas$Treatment == "Control", "Treatment_complete"][1] == "control")
-})
-
-test_that("Config file was there and removed correctly", {
- expect_silent( file.remove(path.expand(paste0(getwd(), "/config_drugsens.txt"))) )
-})
+test_that("Drug combinations have correct units and concentrations", {
+ datas <- drugsens::data_binding(
+ path_to_the_projects_folder = system.file("extdata/to_merge/", package = "drugsens")
+ )
+ # Test drug combination formatting
+ combo_row <- datas[datas$Treatment == "GentamicinePaclitaxel", "Treatment_complete"][1]
+ expect_true(combo_row == "GentamicinePaclitaxel100uM-10uM" ||
+ combo_row == "gentamicinePaclitaxel100uM-10uM")
-test_that("The parsing is working", {
- input_data <- data.frame(Image = "PID1_Tissue1_2024-02-13_DOC2024.02.13_TreatmentRana_10_uM_15_nm_Replica_(series.10)")
- expected_output <- data.frame(
- Image = "PID1_Tissue1_2024-02-13_DOC2024.02.13_TreatmentRana_10_uM_15_nm_Replica_(series.10)",
- Image_number = "series.10",
- PID = "PID1",
- Tissue = "Tissue1",
- Date1 = "2024-02-13",
- DOC = "2024.02.13",
- ReplicaOrNot = "Replica",
- Treatment = "TreatmentRana",
- Concentration1 = "10",
- Concentration2 = "15",
- ConcentrationUnits1 = "uM",
- ConcentrationUnits2 = "nm",
- Treatment_complete = "TreatmentRana10uM-15nm")
- expect_equal(drugsens::string_parsing(input_data), expected = expected_output)
+ # Test control formatting
+ control_row <- datas[datas$Treatment == "Control", "Treatment_complete"][1]
+ expect_true(control_row == "Control" || control_row == "control")
})
-test_that("Another parsing test", {
- input_data <- data.frame(Image = "B516_Ascites_2023-11-25_DOC2020-12-14_dmso_rep_Ecad_cCasp3_(series 01).tif")
- expected_output <- data.frame(
- Image = "B516_Ascites_2023-11-25_DOC2020-12-14_dmso_rep_Ecad_cCasp3_(series 01).tif",
- Image_number = "series 01",
- PID = "B516",
- Tissue = "Ascites",
- Date1 = "2023-11-25",
- DOC = "2020-12-14",
- ReplicaOrNot = "Replica",
- Treatment = "dmso",
- Concentration1 = NA_character_, #WIP
- Concentration2 = NA_integer_,
- ConcentrationUnits1 = NA_character_,
- ConcentrationUnits2 = NA_character_,
- Treatment_complete = "dmso")
- expect_equal(drugsens::string_parsing(input_data), expected = expected_output)
- # Image1 <- "B516_Ascites_2023-11-25_DOC2020-12-14_CarboplatinPaclitaxel_100_uM_10_nM_Ecad_cCasp3_(series 01).tif"
- # Image2 <- "A8759_Spleen_2020.11.10_DOC2001.10.05_compoundX34542_1000_uM_EpCAM_Ecad_cCasp3_(series 01).tif"
- # Image3 <- "A8759_Spleen_2020.11.10_DOC2001.10.05_compoundX34542_1000_uM_EpCAM_Ecad_cCasp3_(series 01).tif"
- # Image4 <- "B38_Eye_2023.11.10_DOC2023.10.05_GentamicinePaclitaxel_100_uM_10_nM_EpCAM_Ecad_cCasp3_(series 01).tif"
+test_that("String parsing works correctly for single drug", {
+ input_data <- data.frame(
+ Image = "PID1_Tissue1_2024-02-13_DOC2024.02.13_TreatmentRana_10_uM_15_nm_Replica_(series.10)"
+ )
+ result <- drugsens::string_parsing(input_data)
+
+ expect_equal(result$PID, "PID1")
+ expect_equal(result$Tissue, "Tissue1")
+ expect_equal(result$Date1, "2024-02-13")
+ expect_equal(result$DOC, "2024.02.13")
+ expect_equal(result$Treatment, "TreatmentRana")
+ expect_equal(result$Concentration1, "10")
+ expect_equal(result$ConcentrationUnits1, "uM")
+ expect_equal(result$Treatment_complete, "TreatmentRana10uM-15nm")
})
+test_that("String parsing works correctly for DMSO control", {
+ input_data <- data.frame(
+ Image = "B516_Ascites_2023-11-25_DOC2020-12-14_dmso_rep_Ecad_cCasp3_(series 01).tif"
+ )
+ result <- drugsens::string_parsing(input_data)
+
+ expect_equal(result$PID, "B516")
+ expect_equal(result$Tissue, "Ascites")
+ expect_equal(result$Date1, "2023-11-25")
+ expect_equal(result$DOC, "2020-12-14")
+ expect_equal(result$Treatment, "dmso")
+ expect_true(is.na(result$Concentration1))
+ expect_true(is.na(result$ConcentrationUnits1))
+ expect_equal(result$Treatment_complete, "dmso")
+})