I'd like to perform a t_test on each group within my data. I'd like to be able to plot the results in ggplot, with each group and t_test in separate facets of facet_grid.
Here's some example data
data <- structure(list(line = c("low", "low", "low", "mid", "mid", "mid",
"high", "high", "high", "high", "high", "high", "low", "low",
"low", "mid", "mid", "mid", "high", "high", "high", "high", "high",
"high"), dose = structure(c(1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L,
1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L), levels = c("0",
"5000", "1e+05"), class = "factor"), expression = c(1, 1.07876018361211,
2.74797781407644, 1.06055856449722, 1.26985386969035, 2.24216268515872,
1.20035861056899, 1.29539908920738, 3.76428705070618, 1.31537212284813,
1.43373360739086, 3.31565979589036, 1.2, 1.29451222033453, 3.29757337689173,
1.27267027739667, 1.52382464362842, 2.69059522219046, 1.44043033268279,
1.55447890704886, 4.51714446084742, 1.57844654741776, 1.72048032886903,
3.97879175506843), time = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2)), row.names = c(NA, -24L
), class = c("tbl_df", "tbl", "data.frame"))
Which looks like this
> head(data)
# A tibble: 6 × 4
line dose expression time
<chr> <fct> <dbl> <dbl>
1 low 0 1 1
2 low 5000 1.08 1
3 low 1e+05 2.75 1
4 mid 0 1.06 1
5 mid 5000 1.27 1
6 mid 1e+05 2.24 1
If this weren't grouped data, I could do the t_test like so
ttest <-
data %>%
t_test(expression ~ line) %>%
add_y_position()
ttest
> ttest
# A tibble: 3 × 12
.y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif y.position groups
<chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr> <dbl> <named list>
1 expression high low 12 6 0.906 12.5 0.382 0.764 ns 4.74 <chr [2]>
2 expression high mid 12 6 1.31 15.9 0.209 0.627 ns 5.06 <chr [2]>
3 expression low mid 6 6 0.193 8.59 0.851 0.851 ns 5.39 <chr [2]>
And the plot the results like this
ggplot(data, aes(x = line, y = expression)) +
geom_boxplot() +
stat_pvalue_manual(ttest,
label = ifelse("p.adj.signif" %in% names(ttest), "p.adj.signif", "p"),
tip.length = 0.01, hide.ns = F)
However, if I add my grouping variables in, I can't get t_test to work
ttest <-
data %>%
group_by(dose, time) %>%
t_test(expression ~ line) %>%
add_y_position()
Error in `mutate()`:
ℹ In argument: `data = map(.data$data, .f, ...)`.
Caused by error in `map()`:
ℹ In index: 1.
Caused by error in `map()`:
ℹ In index: 1.
Caused by error in `t.test.default()`:
! not enough 'y' observations
Run `rlang::last_error()` to see where the error occurred.
What I want is to plot it facetted like this, but with the t_test results in each facet
ggplot(data, aes(x = line, y = expression)) +
facet_grid(time ~ dose) +
geom_boxplot()
Any ideas? I would also like to be able to export a table of t_test results.
