Introduces a package that adds a significant difference symbol to the plots of the “ggplot2” package. See the example output.
Package version is 0.6.3. Checked with R version 4.2.2.
Install Package
Run the following command.
#Install Package install.packages("ggsignif")
Example
See the command and package help for details.
#Loading the library library("ggsignif") #Install the tidyverse package if it is not already present if(!require("tidyverse", quietly = TRUE)){ install.packages("tidyverse");require("tidyverse") } ###Creating Data##### TestData <- data.frame("Group" = c(rep("Group1", 50), sample(paste0("Group", 2:3), 50, replace = TRUE)), "Data1" = c(sample(c(1:8), 10, replace = TRUE), sample(2:8, 40, replace = TRUE), sample(4:7, 50, replace = TRUE))) ######## ###Creating Plot Examples##### #Box Plot BoxPlot <- ggplot(TestData, aes(x = Group, y = Data1)) + geom_boxplot(fill = "lightblue") #Histogram HistPlot <- ggplot(TestData, aes(x = Group, y = Data1, group = Group)) + geom_histogram(aes(fill = Group), stat = "identity") ######## #Significant difference display on plot: geom_signif command #Specify variables: comparisons option; specify by list #Specify bivariate test methods: test option; #t.test, wilcox.test, etc. can be specified #Calculate with missing values removed: na.rm option #Display test results with asterisks: #map_signif_level option;"***"=0.001,"**"=0.01,"*"=0.05 ###Plot##### BoxPlot + geom_signif(comparisons = list(c("Group1", "Group2")), test = "t.test", na.rm = FALSE, map_signif_level = TRUE, col = "red") #Manual Significant Difference Display added #stat option:"identity" ###Display position is specified by data.frame;x,xend,y,yend,annotation#### PlotPoint <- data.frame(x = c(0.8, 1.8, 2.8, 0.8), y = c(8.2, 7.5, 2.5, 8.7), xend = c(1.2, 2.2, 3.2, 3.2), yend = c(8.2, 7.5, 2.5, 8.7), annotation = c("***", "0.05", "**", "NS")) ###Plot##### BoxPlot + geom_signif(data = PlotPoint, aes(x = x, xend = xend, y = y, yend = yend, annotation = annotation), stat = "identity", col = "red")
Output Example
・Significant difference display on plot: geom_signif command
・Manual Significant Difference Display added
I hope this makes your analysis a little easier !!