This article describes how to add p-values generated elsewhere to a ggplot<\/strong> using the ggpubr package. The following key ggpubr functions will be used:<\/p>\n You will learn how to:<\/p>\n Contents:<\/p>\n Make sure you have the following R packages:<\/p>\n Load required R packages:<\/p>\n Key R function: Create a simple box plot<\/strong>:<\/p>\n Faceted plots<\/strong>: Comparisons between two groups<\/p>\n Faceted plots<\/strong>: Pairwise comparisons between multiple groups<\/p>\n Grouped plots<\/strong><\/p>\n This section describes the function This article describes how to add p-values generated elsewhere to a ggplot<\/strong> using the ggpubr package.<\/p>\n<\/div>\n<\/div>\n <\/p>\n","protected":false},"excerpt":{"rendered":" This article describes how to add p-values generated elsewhere to a ggplot using the ggpubr package. The following key ggpubr functions will be used: stat_pvalue_manual(): Add manually p-values to […]<\/p>\n","protected":false},"author":1,"featured_media":16063,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"rating_form_position":"","rating_results_position":"","mr_structured_data_type":"","footnotes":""},"categories":[341],"tags":[343],"class_list":["post-16062","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-faq","tag-ggpubr"],"yoast_head":"\n\n
stat_pvalue_manual()<\/code>: Add manually p-values to a ggplot, such as box blots, dot plots and stripcharts.<\/li>\ngeom_bracket()<\/code>: Add brackets with label annotation to a ggplot. Helpers for adding p-value or significance levels to a plot.<\/li>\n<\/ul>\n\n
\n
\n
\n
Prerequisites<\/h2>\n
\n
tidyverse<\/code> for data manipulation and visualization<\/li>\nggpubr<\/code> for creating easily publication ready plots<\/li>\nrstatix<\/code> provides pipe-friendly R functions for easy statistical analyses<\/li>\n<\/ul>\nlibrary(tidyverse)\r\nlibrary(ggpubr)\r\nlibrary(rstatix)<\/code><\/pre>\n<\/div>\nData preparation<\/h2>\n
# Convert `dose` variable into factor\r\ndf <- ToothGrowth\r\ndf$dose <- factor(df$dose)\r\n\r\n# Inspect the data \r\n# Display two random rows by groups\r\nset.seed(123)\r\ndf %>% sample_n_by(dose, size = 2)<\/code><\/pre>\n## # A tibble: 6 x 3\r\n## len supp dose \r\n## <dbl> <fct> <fct>\r\n## 1 10 VC 0.5 \r\n## 2 14.5 OJ 0.5 \r\n## 3 18.8 VC 1 \r\n## 4 25.8 OJ 1 \r\n## 5 29.4 OJ 2 \r\n## 6 23.6 VC 2<\/code><\/pre>\n<\/div>\nAdd p-values computed from elsewhere<\/h2>\n
stat_pvalue_manual()<\/code> [in ggpubr package]<\/p>\nstat_pvalue_manual(data, label = NULL)<\/code><\/pre>\n\n
data<\/code>: a data frame containing statistical test results. The expected default format should contain the following columns: group1 | group2 | p | y.position | etc<\/code>. group1<\/code> and group2<\/code> are the groups that have been compared. p<\/code> is the resulting p-value. y.position<\/code> is the y coordinates of the p-values in the plot.<\/li>\nlabel<\/code>: the column containing the label (e.g.: label = \"p\"<\/code> or label = \"p.adj\"<\/code>), where p is the p-value. Can be also an expression that can be formatted by the glue<\/code> package. For example, when specifying label = \"t-test, p = {p}\"<\/code>, the expression {p}<\/code> will be replaced by its value.<\/li>\n<\/ul>\n# p-values\r\nstat.test <- tibble::tribble(\r\n ~group1, ~group2, ~p.adj,\r\n \"0.5\", \"1\", 2.54e-07,\r\n \"0.5\", \"2\", 1.32e-13,\r\n \"1\", \"2\", 1.91e-05\r\n )\r\nstat.test<\/code><\/pre>\n## # A tibble: 3 x 3\r\n## group1 group2 p.adj\r\n## <chr> <chr> <dbl>\r\n## 1 0.5 1 2.54e- 7\r\n## 2 0.5 2 1.32e-13\r\n## 3 1 2 1.91e- 5<\/code><\/pre>\n# Box plots + p-values\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n stat_pvalue_manual(\r\n stat.test, \r\n y.position = 35, step.increase = 0.1,\r\n label = \"p.adj\"\r\n )<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-custom-p-values-1.png\")
<\/p>\n<\/div>\nAdd p-values obtained from the rstatix package<\/h2>\n
Create a simple box plot<\/h3>\n
bxp <- ggboxplot(df, x = \"dose\", y = \"len\")\r\nbxp<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-box-plot-1.png\")
<\/p>\n<\/div>\nPairwise comparisons<\/h3>\n
# Statistical test\r\nstat.test <- df %>% t_test(len ~ dose)\r\nstat.test<\/code><\/pre>\n## # A tibble: 3 x 10\r\n## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif\r\n## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr> \r\n## 1 len 0.5 1 20 20 -6.48 38.0 1.27e- 7 2.54e- 7 **** \r\n## 2 len 0.5 2 20 20 -11.8 36.9 4.40e-14 1.32e-13 **** \r\n## 3 len 1 2 20 20 -4.90 37.1 1.91e- 5 1.91e- 5 ****<\/code><\/pre>\n# Box plot\r\nstat.test <- stat.test %>% add_xy_position(x = \"dose\")\r\nbxp + stat_pvalue_manual(stat.test, label = \"p.adj.signif\", tip.length = 0.01)<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-simple-pairwise-comparisons-1.png\")
<\/p>\n<\/div>\nFaceted plots<\/h3>\n
# Statistical tests\r\nstat.test <- df %>%\r\n group_by(supp) %>%\r\n t_test(len ~ dose, ref.group = \"0.5\")\r\nstat.test<\/code><\/pre>\n## # A tibble: 4 x 11\r\n## supp .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif\r\n## * <fct> <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr> \r\n## 1 OJ len 0.5 1 10 10 -5.05 17.7 0.0000878 0.0000878 **** \r\n## 2 OJ len 0.5 2 10 10 -7.82 14.7 0.00000132 0.00000264 **** \r\n## 3 VC len 0.5 1 10 10 -7.46 17.9 0.000000681 0.000000681 **** \r\n## 4 VC len 0.5 2 10 10 -10.4 14.3 0.0000000468 0.0000000936 ****<\/code><\/pre>\n# Box plots\r\nstat.test <- stat.test %>% add_xy_position(x = \"dose\")\r\nggboxplot(df, x = \"dose\", y = \"len\", facet.by = \"supp\") +\r\n stat_pvalue_manual(stat.test, label = \"p.adj.signif\", tip.length = 0.01)<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-faceted-plots-1.png\")
<\/p>\n<\/div>\nGrouped plots<\/h3>\n
# Box plot: comparison against reference\r\nstat.test <- df %>%\r\n group_by(supp) %>%\r\n t_test(len ~ dose, ref.group = \"0.5\") \r\n\r\n# Box plots\r\nstat.test <- stat.test %>% \r\n add_xy_position(x = \"supp\", dodge = 0.8)\r\nbxp <- ggboxplot(df, x = \"supp\", y = \"len\", color = \"dose\")\r\nbxp + stat_pvalue_manual(stat.test, label = \"p.adj\", tip.length = 0.01)<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-grouped-plots-three-groups-1.png\")
<\/p>\n<\/div>\nSpecify manually the y position of the p-values<\/h3>\n
# Pairwise t-test between groups\r\nstat.test <- ToothGrowth %>%\r\n t_test(len ~ dose) %>%\r\n mutate(y.position = c(29, 35, 39))\r\nstat.test<\/code><\/pre>\n## # A tibble: 3 x 11\r\n## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif y.position\r\n## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr> <dbl>\r\n## 1 len 0.5 1 20 20 -6.48 38.0 1.27e- 7 2.54e- 7 **** 29\r\n## 2 len 0.5 2 20 20 -11.8 36.9 4.40e-14 1.32e-13 **** 35\r\n## 3 len 1 2 20 20 -4.90 37.1 1.91e- 5 1.91e- 5 **** 39<\/code><\/pre>\n# Create a box plot and add the p-value\r\nggboxplot(ToothGrowth, x = \"dose\", y = \"len\") +\r\n stat_pvalue_manual(stat.test, label = \"p.adj\")<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-add-custom-pvalues-1.png\")
<\/p>\n# Pairwise t-test between groups\r\nstat.test <- df %>%\r\n group_by(dose) %>%\r\n t_test(len ~ supp) %>%\r\n adjust_pvalue() %>%\r\n mutate(y.position = 35)\r\nstat.test<\/code><\/pre>\n## # A tibble: 3 x 11\r\n## dose .y. group1 group2 n1 n2 statistic df p p.adj y.position\r\n## * <fct> <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <dbl>\r\n## 1 0.5 len OJ VC 10 10 3.17 15.0 0.00636 0.0127 35\r\n## 2 1 len OJ VC 10 10 4.03 15.4 0.00104 0.00312 35\r\n## 3 2 len OJ VC 10 10 -0.0461 14.0 0.964 0.964 35<\/code><\/pre>\n# Create a box plot and add the p-value\r\np <- ggboxplot(\r\n df, x = \"supp\", y = \"len\",\r\n color = \"supp\", palette = \"jco\",\r\n facet.by = \"dose\", ylim = c(0, 40)\r\n )\r\n\r\np + stat_pvalue_manual(stat.test, label = \"p.adj\")<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-facet-comparisons-between-two-groups-1.png\")
<\/p>\n# Pairwise t-test between groups\r\nstat.test <- df %>%\r\n group_by(supp) %>%\r\n t_test(len ~ dose) %>%\r\n adjust_pvalue() %>%\r\n mutate(y.position = rep(c(29, 35, 39), 2))\r\nstat.test<\/code><\/pre>\n## # A tibble: 6 x 12\r\n## supp .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif y.position\r\n## * <fct> <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr> <dbl>\r\n## 1 OJ len 0.5 1 10 10 -5.05 17.7 0.0000878 0.000263 *** 29\r\n## 2 OJ len 0.5 2 10 10 -7.82 14.7 0.00000132 0.00000528 **** 35\r\n## 3 OJ len 1 2 10 10 -2.25 15.8 0.039 0.039 * 39\r\n## 4 VC len 0.5 1 10 10 -7.46 17.9 0.000000681 0.00000340 **** 29\r\n## 5 VC len 0.5 2 10 10 -10.4 14.3 0.0000000468 0.000000281 **** 35\r\n## 6 VC len 1 2 10 10 -5.47 13.6 0.0000916 0.000263 **** 39<\/code><\/pre>\n# Create a box plot and add the p-value\r\np <- ggboxplot(\r\n df, x = \"dose\", y = \"len\",\r\n color = \"supp\", palette = \"jco\",\r\n facet.by = \"supp\", ylim = c(0, 40)\r\n )\r\n\r\np + stat_pvalue_manual(stat.test)<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-facet-pairwise-comparisons-between-mumtiple-groups-1.png\")
<\/p>\n# Pairwise t-test between groups\r\nstat.test <- df %>%\r\n group_by(dose) %>%\r\n t_test(len ~ supp) %>%\r\n adjust_pvalue() %>%\r\n mutate(y.position = 35)\r\nstat.test<\/code><\/pre>\n## # A tibble: 3 x 11\r\n## dose .y. group1 group2 n1 n2 statistic df p p.adj y.position\r\n## * <fct> <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <dbl>\r\n## 1 0.5 len OJ VC 10 10 3.17 15.0 0.00636 0.0127 35\r\n## 2 1 len OJ VC 10 10 4.03 15.4 0.00104 0.00312 35\r\n## 3 2 len OJ VC 10 10 -0.0461 14.0 0.964 0.964 35<\/code><\/pre>\n# Create a box plot and add the p-value\r\np <- ggboxplot(\r\n ToothGrowth, x = \"dose\", y = \"len\",\r\n color = \"supp\", palette = \"jco\",\r\n ylim = c(0, 40)\r\n )\r\n\r\np + stat_pvalue_manual(stat.test, xmin = \"dose\", xmax = NULL)<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-grouped-boxplot-1.png\")
<\/p>\n<\/div>\n<\/div>\nAdd brackets with custom p-value labels to a ggplot<\/h2>\n
geom_bracket()<\/code> [in ggpubr package] for adding brackets with label annotation to a ggplot. It makes it easy to add p-value or significance levels created elsewhere to a plot.<\/p>\nBasic brackets with labels<\/h3>\n
# Add bracket with labels\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n xmin = \"0.5\", xmax = \"1\", y.position = 30,\r\n label = \"t-test, p < 0.05\"\r\n )\r\n\r\n# Customize bracket tip.length tip.length\r\n# Move up the label using vjust\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n xmin = \"0.5\", xmax = \"1\", y.position = 30,\r\n label = \"t-test, p < 0.05\", \r\n tip.length = c(0.2, 0.02), vjust = -1\r\n )<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-add-brackets-with-labels-to-a-ggplot-1.png\")
![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-add-brackets-with-labels-to-a-ggplot-2.png\")
<\/p>\n<\/div>\nUsing plotmath expression and specifying multiple brackets manually<\/h3>\n
#Using plotmath expression\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n xmin = \"0.5\", xmax = \"1\", y.position = 30,\r\n label = \"list(~italic(p)<=0.001)\", type = \"expression\",\r\n tip.length = c(0.2, 0.02)\r\n )\r\n\r\n\r\n# Specify multiple brackets manually\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n xmin = c(\"0.5\", \"1\"), xmax = c(\"1\", \"2\"),\r\n y.position = c(30, 35), label = c(\"***\", \"**\"),\r\n tip.length = 0.01\r\n )<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-using-plotmath-expression-and-multiple-brackets-1.png\")
![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-using-plotmath-expression-and-multiple-brackets-2.png\")
<\/p>\n<\/div>\nCompute statistical tests and add p-values<\/h3>\n
# Compute statistical tests and add p-values\r\nstat.test <- compare_means(len ~ dose, ToothGrowth, method = \"t.test\")\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n aes(xmin = group1, xmax = group2, label = signif(p, 2)),\r\n data = stat.test, y.position = 35\r\n )<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-compute-statistical-testand-add-p-values-1.png\")
<\/p>\n# Increase step length between brackets\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n aes(xmin = group1, xmax = group2, label = signif(p, 2)),\r\n data = stat.test, y.position = 35, step.increase = 0.1\r\n )<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-compute-statistical-testand-add-p-values-2.png\")
<\/p>\n# Or specify the positions of each comparison\r\nggboxplot(df, x = \"dose\", y = \"len\") +\r\n geom_bracket(\r\n aes(xmin = group1, xmax = group2, label = signif(p, 2)),\r\n data = stat.test, y.position = c(32, 35, 38)\r\n )<\/code><\/pre>\n![](\"https:\/\/www.datanovia.com\/en\/wp-content\/uploads\/dn-tutorials\/ggpubr\/figures\/054-add-p-values-generated-elsewhere-compute-statistical-testand-add-p-values-3.png\")
<\/p>\n<\/div>\n<\/div>\nConclusions<\/h2>\n