How to Add P-values to GGPLOT Facets

Home FAQ How to Add P-values to GGPLOT Facets
062-add-p-values-to-ggplot-facets-multipanel-box-plots-facet-wrap-1.png


29 May

How to Add P-values to GGPLOT Facets

Alboukadel

|

|

FAQ

|

7

    1      
1
Share

This article describes how to compute and automatically add p-values onto ggplot facets using the ggpubr and the rstatix R packages.

You will learn how to:

  • Add p-values to a multipanel facet plots containing two or more groups by panel. Examples are shown for box plots and bar plots.
  • Add p-values to a faceted grouped plots (box plots and bar plots). Examples, containing three groups by x position, are shown.
  • Display the p-values combined with the significance levels onto the plots

We will follow the steps below for adding significance levels onto a ggplot:

  1. Compute easily statistical tests (t_test() or wilcox_test()) using the rstatix package
  2. Auto-compute p-value label positions using the function add_xy_position() [in rstatix package].
  3. Add the p-values to the plot using the function stat_pvalue_manual() [in ggpubr package]. The following key options are illustrated in some of the examples:
    • The option bracket.nudge.y is used to move up or to move down the brackets.
    • The option step.increase is used to add more space between brackets.
    • The option vjust is used to vertically adjust the position of the p-values labels

Note that, in some situations, the p-value labels are partially hidden by the plot top border. In these cases, the ggplot2 function scale_y_continuous(expand = expansion(mult = c(0, 0.1))) can be used to add more spaces between labels and the plot top border. The option mult = c(0, 0.1) indicates that 0% and 10% spaces are respectively added at the bottom and the top of the plot.



Contents:

Prerequisites

Make sure you have installed the following R packages:

  • tidyverse for data manipulation and visualization
  • ggpubr for creating easily publication ready plots
  • rstatix provides pipe-friendly R functions for easy statistical analyses.

Start by loading the following required packages:

library(ggpubr)
library(rstatix)

Data preparation

# Transform `dose` into factor variable
df <- ToothGrowth
df$dose <- as.factor(df$dose)
# Add a random grouping variable
df$group <- factor(rep(c("grp1", "grp2"), 30))
head(df, 3)
##    len supp dose group
## 1  4.2   VC  0.5  grp1
## 2 11.5   VC  0.5  grp2
## 3  7.3   VC  0.5  grp1

Multipanel plots containing two groups by panel

Facet wrap by one variable

Statistical tests

Facet by the dose variable and compare the levels of the supp variable on the x-axis.

stat.test <- df %>%
  group_by(dose) %>%
  t_test(len ~ supp) %>%
  adjust_pvalue(method = "bonferroni") %>%
  add_significance()
stat.test 
## # A tibble: 3 x 11
##   dose  .y.   group1 group2    n1    n2 statistic    df       p   p.adj p.adj.signif
##   <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>   <dbl>   <dbl> <chr>       
## 1 0.5   len   OJ     VC        10    10    3.17    15.0 0.00636 0.0191  *           
## 2 1     len   OJ     VC        10    10    4.03    15.4 0.00104 0.00312 **          
## 3 2     len   OJ     VC        10    10   -0.0461  14.0 0.964   1       ns

Box plots

# Create a box plot
bxp <- ggboxplot(
  df, x = "supp", y = "len", fill = "#00AFBB", 
  facet.by = "dose"
  )

# Make facet and add p-values
stat.test <- stat.test %>% add_xy_position(x = "supp")
bxp + stat_pvalue_manual(stat.test)

# Make the facet scale free and add jitter points
# Move down the bracket using `bracket.nudge.y`
# Hide ns (non-significant)
# Show adjusted p-values and significance levels
# Add 10% spaces between the p-value labels and the plot border
bxp <- ggboxplot(
  df, x = "supp", y = "len", fill = "#00AFBB", 
  facet.by = "dose", scales = "free", add = "jitter"
  )
bxp +  
  stat_pvalue_manual(
    stat.test, bracket.nudge.y = -2, hide.ns = TRUE,
    label = "{p.adj}{p.adj.signif}"
    ) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.1)))

Bar plots

The option add = "mean_sd" is specified in the bar plot function for creating bar plot with error bars (mean +/- SD). You need to specify the same summary statistics function to auto-compute p-value labels positions in add_xy_position() using the option fun.

# Create a bar plot with error bars (mean +/- sd)
bp <- ggbarplot(
  df, x = "supp", y = "len", add = "mean_sd", 
  fill = "#00AFBB", facet.by = "dose"
  )

# Add p-values onto the bar plots
stat.test <- stat.test %>% add_xy_position(fun = "mean_sd", x = "supp")
bp + stat_pvalue_manual(stat.test)

Bar plots with jitter points

You need to specify fun = "max" when computing the p-value labels positions, so that the first bracket will start at the maximum of the data points. This avoids the overlap between data points and brackets.

# Create a bar plot with error bars (mean +/- sd)
bp <- ggbarplot(
  df, x = "supp", y = "len", add = c("mean_sd", "jitter"), 
  fill = "#00AFBB", facet.by = "dose"
  )

# Add p-values onto the bar plots
stat.test <- stat.test %>% add_xy_position(fun = "max", x = "supp")
bp + stat_pvalue_manual(stat.test)

Facet grid by two variables

Statistical test

Facet by the dose and group variables, and compare the levels of the supp variable on the x-axis.

stat.test <- df %>%
  group_by(group, dose) %>%
  t_test(len ~ supp) %>%
  adjust_pvalue(method = "bonferroni") %>%
  add_significance()
stat.test 
## # A tibble: 6 x 12
##   dose  group .y.   group1 group2    n1    n2 statistic    df       p   p.adj p.adj.signif
##   <fct> <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>   <dbl>   <dbl> <chr>       
## 1 0.5   grp1  len   OJ     VC         5     5     3.71   7.35 0.00697 0.0418  *           
## 2 1     grp1  len   OJ     VC         5     5     1.37   7.65 0.208   1       ns          
## 3 2     grp1  len   OJ     VC         5     5    -0.485  6.52 0.643   1       ns          
## 4 0.5   grp2  len   OJ     VC         5     5     1.13   5.79 0.304   1       ns          
## 5 1     grp2  len   OJ     VC         5     5     6.91   5.84 0.00051 0.00306 **          
## 6 2     grp2  len   OJ     VC         5     5     0.319  6.07 0.76    1       ns

Box plots

# Create box plots with significance levels
# Hide ns (non-significant)
stat.test <- stat.test %>% add_xy_position(x = "supp")
ggboxplot(
  df, x = "supp", y = "len", fill = "#E7B800",
  facet = c("group", "dose")
  ) +
  stat_pvalue_manual(stat.test, hide.ns = TRUE)

Bar plots

# Create bar plots with significance levels
# Hide ns (non-significant)
stat.test <- stat.test %>% add_xy_position(x = "supp", fun = "mean_sd")
ggbarplot(
  df, x = "supp", y = "len", fill = "#E7B800",
  add = c("mean_sd", "jitter"), facet = c("group", "dose")
  ) +
  stat_pvalue_manual(stat.test, hide.ns = TRUE)

Multipanel plots containing three or more groups by panel

Facet wrap by one variable

Perform all pairwise comparisons

Group by the supp variable and then perform pairwise comparisons between the levels of dose variable.

Statistical test:

stat.test <- df %>%
  group_by(supp) %>%
  t_test(len ~ dose)
stat.test
## # A tibble: 6 x 11
##   supp  .y.   group1 group2    n1    n2 statistic    df            p      p.adj p.adj.signif
## * <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>        <dbl>      <dbl> <chr>       
## 1 OJ    len   0.5    1         10    10     -5.05  17.7 0.0000878    0.000176   ***         
## 2 OJ    len   0.5    2         10    10     -7.82  14.7 0.00000132   0.00000396 ****        
## 3 OJ    len   1      2         10    10     -2.25  15.8 0.039        0.039      *           
## 4 VC    len   0.5    1         10    10     -7.46  17.9 0.000000681  0.00000136 ****        
## 5 VC    len   0.5    2         10    10    -10.4   14.3 0.0000000468 0.00000014 ****        
## 6 VC    len   1      2         10    10     -5.47  13.6 0.0000916    0.0000916  ****

Add the p-values onto the plots. The ggplot2 function scale_y_continuous(expand = expansion(mult = c(0, 0.1))) is used to add more spaces between labels and the plot top border

# Box plots with p-values
stat.test <- stat.test %>% add_y_position()
ggboxplot(df, x = "dose", y = "len", fill = "#FC4E07", facet.by = "supp") +
  stat_pvalue_manual(stat.test, label = "p.adj.signif", tip.length = 0.01) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.1)))

# Bar plot with p-values
# Add 10% space on the y-axis above the box plots
stat.test <- stat.test %>% add_y_position(fun = "mean_sd")
ggbarplot(
  df, x = "dose", y = "len", fill = "#FC4E07",
  add = "mean_sd", facet.by = "supp"
  ) + 
  stat_pvalue_manual(stat.test,  label = "p.adj.signif", tip.length = 0.01) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.1)))

Pairwise comparisons against a reference group

Statistical test:

stat.test <- df %>%
  group_by(supp) %>%
  t_test(len ~ dose, ref.group = "0.5")
stat.test
## # A tibble: 4 x 11
##   supp  .y.   group1 group2    n1    n2 statistic    df            p        p.adj p.adj.signif
## * <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>        <dbl>        <dbl> <chr>       
## 1 OJ    len   0.5    1         10    10     -5.05  17.7 0.0000878    0.0000878    ****        
## 2 OJ    len   0.5    2         10    10     -7.82  14.7 0.00000132   0.00000264   ****        
## 3 VC    len   0.5    1         10    10     -7.46  17.9 0.000000681  0.000000681  ****        
## 4 VC    len   0.5    2         10    10    -10.4   14.3 0.0000000468 0.0000000936 ****
# Box plots with p-values
stat.test <- stat.test %>% add_y_position()
ggboxplot(df, x = "dose", y = "len", fill = "#FC4E07", facet.by = "supp") +
  stat_pvalue_manual(stat.test, label = "p.adj.signif", tip.length = 0.01) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.1)))

# Show only significance levels at x = group2
# Move down significance symbols using vjust
stat.test <- stat.test %>% add_y_position()
ggboxplot(df, x = "dose", y = "len", fill = "#FC4E07", facet.by = "supp") +
  stat_pvalue_manual(stat.test, label = "p.adj.signif", x = "group2", vjust = 2) 

# Bar plot with p-values
# Add 10% space on the y-axis above the box plots
stat.test <- stat.test %>% add_y_position(fun = "mean_sd")
ggbarplot(
  df, x = "dose", y = "len", fill = "#FC4E07",
  add = c("mean_sd", "jitter"), facet.by = "supp"
  ) + 
  stat_pvalue_manual(stat.test,  label = "p.adj.signif", tip.length = 0.01) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.1)))

Facet grid by two variables

Statistical test

Facet by the supp and group variables, and compare the levels of the dose variable on the x-axis.

stat.test <- df %>%
  group_by(group, supp) %>%
  t_test(len ~ dose) %>%
  adjust_pvalue(method = "bonferroni") %>%
  add_significance()
stat.test 
## # A tibble: 12 x 12
##   supp  group .y.   group1 group2    n1    n2 statistic    df         p    p.adj p.adj.signif
##   <fct> <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>     <dbl>    <dbl> <chr>       
## 1 OJ    grp1  len   0.5    1          5     5     -2.50  7.80 0.038     0.456    ns          
## 2 OJ    grp1  len   0.5    2          5     5     -5.66  7.13 0.000717  0.00860  **          
## 3 OJ    grp1  len   1      2          5     5     -2.22  6.49 0.064     0.768    ns          
## 4 VC    grp1  len   0.5    1          5     5     -5.33  7.54 0.000855  0.0103   *           
## 5 VC    grp1  len   0.5    2          5     5     -8.81  6.75 0.0000606 0.000727 ***         
## 6 VC    grp1  len   1      2          5     5     -3.84  7.67 0.005     0.06     ns          
## # … with 6 more rows

Box plots

# Create box plots with significance levels
# Hide ns (non-significant)
# Add 15% space between labels and the plot top border
stat.test <- stat.test %>% add_xy_position(x = "dose")
ggboxplot(
  df, x = "dose", y = "len", fill = "#FC4E07",
  facet = c("group", "supp")
  ) +
  stat_pvalue_manual(stat.test, hide.ns = TRUE) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.15)))

Bar plots

# Create bar plots with significance levels
# Hide ns (non-significant)
# Add 15% space between labels and the plot top border
stat.test <- stat.test %>% add_xy_position(x = "dose", fun = "mean_sd")
ggbarplot(
  df, x = "dose", y = "len", fill = "#FC4E07",
  add = c("mean_sd", "jitter"), facet = c("group", "supp")
  ) +
  stat_pvalue_manual(stat.test, hide.ns = TRUE) +
  scale_y_continuous(expand = expansion(mult = c(0.05, 0.15)))

Faceted grouped plots

Simple plots

# Box plots
bxp <- ggboxplot(
  df, x = "supp", y = "len", color = "dose", 
   palette = "jco", facet.by = "group"
  )
bxp 

# Bar plots
bp <- ggbarplot(
  df, x = "supp", y = "len", color = "dose",
  palette = "jco", add = "mean_sd",
  position = position_dodge(0.8),
  facet.by = "group"
  )
bp

Perform all pairwise comparisons

Statistical test:

stat.test <- df %>%
  group_by(supp, group) %>%
  t_test(len ~ dose) 
stat.test
## # A tibble: 12 x 12
##   supp  group .y.   group1 group2    n1    n2 statistic    df        p p.adj p.adj.signif
## * <fct> <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>    <dbl> <dbl> <chr>       
## 1 OJ    grp1  len   0.5    1          5     5     -2.50  7.80 0.038    0.076 ns          
## 2 OJ    grp1  len   0.5    2          5     5     -5.66  7.13 0.000717 0.002 **          
## 3 OJ    grp1  len   1      2          5     5     -2.22  6.49 0.064    0.076 ns          
## 4 OJ    grp2  len   0.5    1          5     5     -4.85  5.65 0.003    0.007 **          
## 5 OJ    grp2  len   0.5    2          5     5     -5.28  6.35 0.002    0.005 **          
## 6 OJ    grp2  len   1      2          5     5     -1.00  7.69 0.347    0.347 ns          
## # … with 6 more rows

Add p-values onto the plots:

# Box plots with p-values
# Hide ns (non-significant)
stat.test <- stat.test %>%
  add_xy_position(x = "supp", dodge = 0.8)
bxp + 
  stat_pvalue_manual(
    stat.test, label = "p.adj.signif", tip.length = 0.01,
    hide.ns = TRUE
    ) +
  scale_y_continuous(expand = expansion(mult = c(0.01, 0.1)))

# Bar plots with p-values
stat.test <- stat.test %>%
  add_xy_position(x = "supp", fun = "mean_sd", dodge = 0.8)
bp + 
  stat_pvalue_manual(
    stat.test, label = "p.adj.signif", tip.length = 0.01,
    hide.ns = TRUE
    ) +
  scale_y_continuous(expand = expansion(mult = c(0.01, 0.1)))

Pairwise comparisons against a reference group

Statistical test:

stat.test <- df %>%
  group_by(supp, group) %>%
  t_test(len ~ dose, ref.group = "0.5")
stat.test
## # A tibble: 8 x 12
##   supp  group .y.   group1 group2    n1    n2 statistic    df         p    p.adj p.adj.signif
## * <fct> <fct> <chr> <chr>  <chr>  <int> <int>     <dbl> <dbl>     <dbl>    <dbl> <chr>       
## 1 OJ    grp1  len   0.5    1          5     5     -2.50  7.80 0.038     0.038    *           
## 2 OJ    grp1  len   0.5    2          5     5     -5.66  7.13 0.000717  0.001    **          
## 3 OJ    grp2  len   0.5    1          5     5     -4.85  5.65 0.003     0.003    **          
## 4 OJ    grp2  len   0.5    2          5     5     -5.28  6.35 0.002     0.003    **          
## 5 VC    grp1  len   0.5    1          5     5     -5.33  7.54 0.000855  0.000855 ***         
## 6 VC    grp1  len   0.5    2          5     5     -8.81  6.75 0.0000606 0.000121 ***         
## # … with 2 more rows

Add p-values onto the plots:

# Box plots with p-values
stat.test <- stat.test %>%
  add_xy_position(x = "supp", dodge = 0.8)
bxp + 
  stat_pvalue_manual(
    stat.test, label = "p.adj.signif", tip.length = 0.01
    ) +
  scale_y_continuous(expand = expansion(mult = c(0.01, 0.1)))

# Bar plots with p-values
stat.test <- stat.test %>%
  add_xy_position(x = "supp", fun = "mean_sd", dodge = 0.8)
bp + 
  stat_pvalue_manual(
    stat.test, label = "p.adj.signif", tip.length = 0.01
    ) +
  scale_y_continuous(expand = expansion(mult = c(0.01, 0.1)))

Conclusion

This article describes how to add p-values onto ggplot facets, such as box plots and bar plots. See other related frequently questions: ggpubr FAQ.



 
1
Share
    1     

Version: Français





Comments ( 7 )

  • Vatsay
    01 Jul 2020

    Hi, thanks for the nice feature, chow can we do that in plot_ly ?, any insights are appreciated

    Reply
  • Steph
    02 Mar 2021

    Hi this is great- is stat.test a function? it keeps coming back unrecognised for me

    Reply
    • Iordanis
      07 Apr 2021

      stat.test is used as a variable in these examples. It is used to store the results of the respective statistical tests performed [notice the rstatix package and the accompanying pipes (%>%) that perform the actual tests].

      Reply
  • Alicia
    10 Jun 2022

    This was soooo helpful! Thank you!

    Reply
  • Phan
    19 Mar 2023

    Hi, Should we use anova-post hoc test for the part “Multipanel plots containing three or more groups by panel”? If yes, could you show me how to do it? Thank you very much!

    Reply
  • Alec
    26 Nov 2023

    This article may very well have saved my life. Very well written, had every use case imaginable. Exactly what I needed after hours of looking

    Reply
  • Laura
    14 Dec 2023

    Hi, can you help me? I have a facet_wrap with 5 facets, boxplot with different numbers of boxes per facet. I wanna have aaall the p-values of pairwise comparisons.
    I put the following, but receive error messages:
    stat.test % group_by(method, year) %>% t_test(H~crop) %>%
    ggplot(aes(crop, H, fill=crop))+
    geom_boxplot()+
    facet_wrap(~method, scales=”free_x”)+
    stat_pvalue_manual(stat.test, label = “p.adj.signif”, tip.length = 0.01)

    what do I need to change?
    Thanks a lot!!!

    Reply

Give a comment

Want to post an issue with R? If yes, please make sure you have read this: How to Include Reproducible R Script Examples in Datanovia Comments