Error adding p-value (group comparisons) to ggboxplot

[jmicrobe]

Great package! I'm trying to apply it to a relatively small data set that I have and I'm getting an error when trying to add significance data to the plot.

Running

compare_means(per_bead ~ beads, protein)

yields a 6 x 8 tibble with some of the significance data I'm hoping to add to my plot:

# A tibble: 6 x 8
       .y. group1 group2          p     p.adj p.format p.signif   method
     <chr>  <chr>  <chr>      <dbl>     <dbl>    <chr>    <chr>    <chr>
1 per_bead     20     10 1.00000000 1.0000000    1.000       ns Wilcoxon
2 per_bead     20      8 0.30649217 0.9194765    0.306       ns Wilcoxon
3 per_bead     20      5 0.03038282 0.1822969    0.030        * Wilcoxon
4 per_bead     10      8 0.56136321 1.0000000    0.561       ns Wilcoxon
5 per_bead     10      5 0.06060197 0.3030098    0.061       ns Wilcoxon
6 per_bead      8      5 0.06060197 0.3030098    0.061       ns Wilcoxon

Following the example given in the docs for adding p-values comparing groups:

my_comparisons <- list(c("5, 20"), c("5, 10"), c("10, 20"))
ggboxplot(prot, x = "beads", y = "per_bead",
          color = "beads", palette = "jco") +
  stat_compare_means(comparisons = my_comparisons)

I get the following errors:

Warning messages:
1: Removed 16 rows containing non-finite values (stat_boxplot). 
2: Removed 16 rows containing non-finite values (stat_signif). 
3: Computation failed in `stat_signif()`:
missing value where TRUE/FALSE needed

I've tried several different arguments in stat_compare_means() without success. I'm hoping to achieve something similar to the example plot below (or just the output of "p.signif" instead of a value):

[kassambara]

Hi,

Use:

my_comparisons <- list(c("5", "20"), c("5", "10"), c("10", "20"))

instead of:

my_comparisons <- list(c("5, 20"), c("5, 10"), c("10, 20"))

[jmicrobe]

@kassambara of course it was a simple typo. That worked great, thank you. Can't wait to use this package more!

[elcansaito]

This is my data

ID	Period	Tra
2	O	33.81648475
4	O	46.0599032
7	O	39.24617467
15	O	34.30622148
21	O	42.71691764
23	O	38.99065985
24	O	48.87056621
51	O	41.14124291
55	O	39.86366882
57	O	34.41268599
72	I	40.86443519
93	I	40.01271913
94	I	38.52221601
97	I	40.60892037
128	I	34.41268599
129	I	36.41421875
141	I	39.73591141
148	I	37.64920704
151	I	34.90242273
166	I	39.13971016
179	M	45.46370195
180	M	40.43857716
181	M	49.55193907
183	M	44.05837044
184	M	47.89109274
185	M	46.37929672
188	M	51.10632089
189	M	36.52068326
198	M	45.37853034
200	M	41.52451514

I am trying to add the pvalues on a boxplot and I always get the same error. I have tried different methods but I always get the same error.

METHOD 1

my_comparisons <- list( c("O M"), c("O I"), c("I M") ) ggboxplot(newdata, x = "Period", y = "tra", color = "Period", palette = "jco",add="jitter")+ stat_compare_means(comparisons = my_comparisons,method = "anova")+ # Add global p-value stat_compare_means(aes(label = ..p.signif..), method = "t.test", ref.group = "0.5")

METHOD 2

ggboxplot(newdata, x = "Period", y = "tra", color = "Period", palette = "npg")+ stat_compare_means(method = "anova", label.y = 50)+ # Add global p-value stat_compare_means(aes(label = ..p.signif..), method = "t.test", ref.group = "0.5")

ALWAYS THE SAME ERROR and I don´t see the starts or the pvalues on the plot 2: Computation failed in stat_compare_means(): missing value where TRUE/FALSE needed

Any suggestions?

THANKS,

[bioemprendiendo]

How to add comparissons between treatments for heach gene expression? i mean, a want to add p-value and brackets for each gene (GATA3, PTEN, XBP1) but comparing BRCA to OV and BRCA to LUSC for each gene like this:

my_comparisons <- list(c("BRCA", "OV"), c("OV", "LUSC")) ggboxplot(expr, x = "dataset", y = "GATA3", title = "GATA3", ylab = "Expression", color = "dataset", palette = "jco")+ stat_compare_means(comparisons = my_comparisons)

but doing this:

ggboxplot(expr, x = "dataset", y = c("GATA3", "PTEN", "XBP1"), merge = "flip", ylab = "Expression", palette = "jco")

thank you

[micdonato]

Why is this closed? I was interested in the same, and I cannot get it to work

[NJU-Bio-Info]

same question......

[kassambara]

Are you looking for something like this:

library(ggpubr)

bxp <- ggboxplot(gene_expression, x = "dataset",
          y = c("GATA3", "PTEN", "XBP1"),
          merge = "flip",
          ylab = "Expression",
          palette = "jco")
bxp +
  geom_pwc(
    aes(group = dataset), tip.length = 0,
    method = "wilcox_test", label = "p.adj.format",
    bracket.nudge.y = -0.08
  ) +
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

The following docs might help:

• Add Pairwise Comparisons P-values to a GGPlot
• How to Add P-Values onto a Grouped GGPLOT using the GGPUBR R Package

Please, consider installing the latest dev version of ggpubr using devtools::install_github("kassambara/ggpubr")