====== R plots ====== ===== Box plots ===== An user guide to walk through data visualization by box plot in R.\\ Please see the tutorials for more details.\\ ==== One batch with replicates ==== === Single batch boxplot tutorials === * {{single-batch_boxplot_v01.pdf}} === Sample input === * {{sample1.xlsx}} === R script === # Aim: This R script is about making single batch boxplot. # Data Description: One batch with replicates # - sample1.xlsx # - Sheet1: Contains raw data across two columns: `Strain` and `Value`. Each strain includes 20 replicates. # - Sheet2: Contains the statistical test result (Tukey's HSD). # # Output (both pdf and png format): # - Version 1: Displays all data points using a single color. # - Version 2: Colors the data points by strain and adds statistical significance group labels at the top. #--------------------------------------------------------------------------------------------------------------- ### Step 0, Check packages # Required packages packages <- c("ggplot2", "readxl") # Check and install required packages for (pkg in packages) { if (!requireNamespace(pkg, quietly = TRUE)) { install.packages(pkg, repos = "https://cloud.r-project.org") } } #--------------------------------------------------------------------------------------------------------------- ### Step 1. Configuration ## Set up a dedicated working directory to store your raw files and output plots. # working directory working_path = "/scratch/hychang/R_wd/Rana001/Rana001.01/single-batch/" # input data located directory input_path = "/scratch/hychang/R_wd/Rana001/Rana001.01/single-batch/" # output folder for storing plots output_path = "/scratch/hychang/R_wd/Rana001/Rana001.01/single-batch/plots/" ## Set Input and output # File name of the raw data table, including the file extension input_file_raw = "sample1.xlsx" # Worksheet name of the raw data table input_sheet_raw = "Sheet1" # (Optional) File name of the statistical test result table, including the file extension input_file_stat = "sample1.xlsx" # (Optional) Worksheet name of the statistical test result table input_sheet_stat = "Sheet2" # Column names in the input data x_column = "Strain" y_column = "Value" # Column name of statistical significance group letters, skip if the statistical test not available stat_column = "groups" # Output file names of the figures, including the file extension output_file_1 = "single-batch_v1.pdf" output_file_2 = "single-batch_v2.pdf" output_file_3 = "single-batch_v1.png" output_file_4 = "single-batch_v2.png" # Set up a dedicated working directory to store your raw files and output plots. dir.create(working_path, recursive = TRUE, showWarnings = FALSE) dir.create(input_path, recursive = TRUE, showWarnings = FALSE) dir.create(output_path, recursive = TRUE, showWarnings = FALSE) ## Setting figure parameters # the names of x and y axis x_lab = "Strain" y_lab = "Value" # y axis range and breaks y_range = c(0, 7) y_breaks = c(0, 2, 4, 6) # font size of x and y axis title (axis label) axis_title_size = 16 # font size of x and y axis text (tick label) axis_text_size = 14 # sample ordering x_order = c("A", "B", "C") # size of data points point_size= 2 # box size box_size= 0.4 # use a single color for all data points color_single = c("#003175") # use multiple colors for the data points; the number of colors should match the number of samples color_multiple = c(c("#1B9E77", "#D95F02", "#7570B3")) ## Export setting pdf_width = 6 pdf_height = 4 png_width = 600 png_height = 400 #--------------------------------------------------------------------------------------------------------------- ### Step 2. Setup and Import data # Set working directory setwd(working_path) # load packages library(readxl) library(ggplot2) # Import raw data df1= read_xlsx(file.path(input_path, input_file_raw), sheet = input_sheet_raw) # (Optional) Import statistical test result. Skip this step if you don't have the results. df2= read_xlsx(file.path(input_path, input_file_stat), sheet = input_sheet_stat) # Sample ordering df1[[x_column]] = factor(df1[[x_column]], levels = x_order) df2[[x_column]] = factor(df2[[x_column]], levels = x_order) # skip this if df2 not available # Display the data frame and view its dimensions (number of rows and columns). df1; dim(df1) df2; dim(df2) # skip this if df2 not available # show sample counts and if there is NA value. table(df1[[x_column]], useNA = "always") # range of value range(df1[[y_column]]) #--------------------------------------------------------------------------------------------------------------- #### Step 3. Generate Plots # Version 1 # all data points using a single, uniform color. p1 = ggplot(data = df1, aes(x = .data[[x_column]], y = .data[[y_column]])) + geom_boxplot(width = box_size, outlier.shape = NA) + geom_point(color = color_single, size = point_size, position = position_jitter(width = 0.2, height = 0)) + scale_y_continuous(limits = y_range, breaks = y_breaks) + labs(x = x_lab, y = y_lab) + theme_classic() + theme(axis.title = element_text(size = axis_title_size), axis.text = element_text(size = axis_text_size)) # show plot p1 # Version 2 # colors data points by Strain and adds statistical significance group labels at the top. p2 = ggplot(data = df1, aes(x = .data[[x_column]], y = .data[[y_column]])) + geom_boxplot(width = box_size, outlier.shape = NA) + geom_point(aes(color = .data[[x_column]]), size = point_size, position = position_jitter(width = 0.2, height = 0)) + geom_text(data = df2, aes(x = .data[[x_column]], y = Inf, label = .data[[stat_column]]), vjust = "inward") + scale_color_manual(values = c(color_multiple)) + scale_y_continuous(limits = y_range, breaks = y_breaks) + labs(x = x_lab, y = y_lab) + theme_classic() + theme(axis.title = element_text(size = axis_title_size), axis.text = element_text(size = axis_text_size)) # show plot p2 #--------------------------------------------------------------------------------------------------------------- #### Step 4. Export Plots ## Export PDF # version 1 PDF export pdf(file.path(output_path, output_file_1), width = pdf_width, height = pdf_height) p1 dev.off() # version 2 PDF export pdf(file.path(output_path, output_file_2), width = pdf_width, height = pdf_height) p2 dev.off() ## Export PNG # Version 1 PNG export png(file.path(output_path, output_file_3), width = png_width, height = png_height) p1 dev.off() # Version 2 PNG export png(file.path(output_path, output_file_4), width = png_width, height = png_height) p2 dev.off() ### END this script === Sample output === Plots can be exported as **PDF** or **PNG** files. Here shows the **PNG** files.\\ * **Version 1**: Displays all data points using a single, uniform color. * PNG: setting width 600 and height 400. * {{ single-batch_v1.png }} * PDF: width 6 inch and height 4 inch. {{ single-batch_v1.pdf }} * **Version 2**: Colors data points by Strain and adds statistical significance group labels at the top. * PNG: setting width 600 and height 400. * {{ single-batch_v2.png }} * PDF: width 6 inch and height 4 inch. {{ single-batch_v2.pdf }} \\ ==== Multiple batches with replicates ==== === Multiple batches boxplot tutorials === * {{multiple-batches_boxplot_v01.pdf}} === Sample input === * {{sample2.xlsx}} === R script === # Aim: This R script is about generating multiple batches boxplot. # Data Description: Multiple batches with replicates # - sample1.xlsx # - Sheet1: Contains raw data across three columns: `Strain`, `Batch`, `Value`. Each strain has three batches. # - Sheet2: Contains the statistical test result (Tukey's HSD). # # Output (both pdf and png format): # - Version 1: Displays data with batches shown in distinct colors. # - Version 2: Displays batches in distinct colors and adds statistical significance group labels at the top. #--------------------------------------------------------------------------------------------------------------- ### Step 0, Check packages # Required packages packages <- c("ggplot2", "readxl") # Check and install required packages for (pkg in packages) { if (!requireNamespace(pkg, quietly = TRUE)) { install.packages(pkg, repos = "https://cloud.r-project.org") } } #--------------------------------------------------------------------------------------------------------------- ### Step 1. Configuration ## Set up a dedicated working directory to store your raw files and output plots. # working directory working_path = "/scratch/hychang/R_wd/Rana001/Rana001.01/multiple-batches/" # input data located directory input_path = "/scratch/hychang/R_wd/Rana001/Rana001.01/multiple-batches/" # output folder for storing plots output_path = "/scratch/hychang/R_wd/Rana001/Rana001.01/multiple-batches/plots/" ## Set Input and output # File name of the raw data table, including the file extension input_file_raw = "sample2.xlsx" # Worksheet name of the raw data table input_sheet_raw = "Sheet1" # File name of the statistical test result table, including the file extension input_file_stat = "sample2.xlsx" # Worksheet name of the statistical test result table input_sheet_stat = "Sheet2" # Column names in the input data x_column = "Strain" y_column = "Value" batch_column = "Batch" # Column name of statistical significance group letters, # skip if the statistical test not available stat_column = "groups" # Output file names of the figures, including the file extension output_file_1 = "mult-batches_v01.pdf" output_file_2 = "mult-batches_v02.pdf" output_file_3 = "mult-batches_v01.png" output_file_4 = "mult-batches_v02.png" # Set up a dedicated working directory to store your raw files and output plots. dir.create(working_path, recursive = TRUE, showWarnings = FALSE) dir.create(input_path, recursive = TRUE, showWarnings = FALSE) dir.create(output_path, recursive = TRUE, showWarnings = FALSE) ## Setting figure parameters # the names of x and y axis x_lab = "Strain" y_lab = "Value" # y axis range and breaks y_range = c(0, 6) y_breaks = c(0, 2, 4, 6) # font size of x and y axis title (axis label) axis_title_size = 16 # font size of x and y axis text (tick label) axis_text_size = 14 # sample ordering x_order = c("A", "B", "C") # size of data points point_size= 2 # box size box_size= 0.4 # space between batches (points) space_batch = 0.3 # use multiple colors for the data points; # the number of colors should match the number of batches color_multiple = c(c("#1B9E77", "#D95F02", "#7570B3")) ## Export setting pdf_width = 6 pdf_height = 4 png_width = 600 png_height = 400 #--------------------------------------------------------------------------------------------------------------- ### Step 2. Setup and Import data # Set working directory setwd(working_path) # load packages library(readxl) library(ggplot2) # Import raw data df1= read_xlsx(file.path(input_path, input_file_raw), sheet = input_sheet_raw) # (Optional) Import statistical test result. Skip this step if you don't have the results. df2= read_xlsx(file.path(input_path, input_file_stat), sheet = input_sheet_stat) # Sample ordering df1[[x_column]] = factor(df1[[x_column]], levels = x_order) df2[[x_column]] = factor(df2[[x_column]], levels = x_order) # skip this if df2 not available # Display the data frame and view its dimensions (number of rows and columns). df1; dim(df1) df2; dim(df2) # skip this if df2 not available # Show sample counts, including NA values table(df1[[x_column]], useNA = "always") # Show the range of values range(df1[[y_column]]) # check batch counts, including NA values table(df1[[batch_column]], useNA = "always") #--------------------------------------------------------------------------------------------------------------- #### Step 3. Generate Plots # Version 1 # all data points using a single, uniform color. p1= ggplot(data= df1, aes(x= .data[[x_column]], y= .data[[y_column]]))+ geom_boxplot(width = box_size, outlier.shape = NA)+ geom_point(data= df1, aes(fill= .data[[batch_column]]), shape = 21, size = point_size, position= position_dodge(width = space_batch))+ scale_fill_manual(values = color_multiple)+ scale_y_continuous(limits = y_range, breaks = y_breaks)+ labs(x= x_lab, y= y_lab)+ theme_classic()+ theme(axis.title = element_text(size = axis_title_size), axis.text = element_text(size = axis_text_size)) # show plot p1 # Version 2 # colors data points by Strain and adds statistical significance group labels at the top. p2= ggplot(data = df1, aes(x= .data[[x_column]], y= .data[[y_column]]))+ geom_boxplot(width = box_size, outlier.shape = NA)+ geom_point(aes(fill= .data[[batch_column]]), shape = 21, size = point_size, position= position_dodge(width = space_batch))+ geom_text(data = df2, aes(x= .data[[x_column]], y= Inf, label = .data[[stat_column]]), vjust= "inward")+ scale_fill_manual(values = color_multiple)+ scale_y_continuous(limits = y_range, breaks = y_breaks)+ labs(x= x_lab, y= y_lab)+ theme_classic()+ theme(axis.title = element_text(size = axis_title_size), axis.text = element_text(size = axis_text_size)) # show plot p2 #--------------------------------------------------------------------------------------------------------------- #### Step 4. Export Plots ## Export PDF # version 1 PDF export pdf(file.path(output_path, output_file_1), width = pdf_width, height = pdf_height) p1 dev.off() # version 2 PDF export pdf(file.path(output_path, output_file_2), width = pdf_width, height = pdf_height) p2 dev.off() ## Export PNG # Version 1 PNG export png(file.path(output_path, output_file_3), width = png_width, height = png_height) p1 dev.off() # Version 2 PNG export png(file.path(output_path, output_file_4), width = png_width, height = png_height) p2 dev.off() ### END this script === Sample output === Plots can be exported as **PDF** or **PNG** files. Here shows the **PNG** files.\\ * **Version 1**: Displays data with batches shown in distinct colors. * PNG: setting width 600 and height 400. * {{ mult-batches_v01.png }} * PDF: width 6 inch and height 4 inch. {{ mult-batches_v01.pdf }} * **Version 2**: Displays batches in distinct colors and adds statistical significance group labels at the top. * PNG: setting width 600 and height 400. * {{ mult-batches_v02.png }} * PDF: width 6 inch and height 4 inch. {{ mult-batches_v02.pdf }}