lapply on a list of dataframe to create plots - r

Question

I have created a list of dataframe df_list, each of which is an 18 x 4 dataframe.
The first columns of the dataframe is 18-times-repeated gene name, and the rest three columns are the gene's information. Each dataframe describes different gene.
Now I'd like to iterate the list of dataframe (i.e, a list of gene and their respective information) over the boxplot, to get plots on each gene; however, I am not sure how to deal with the ggtitle below:
Here is my simplified boxplot function:
box <- function(df){
df %>%
ggplot(df, aes(x = df[,4], y = df[,2])) +
geom_boxplot() +
ggtitle(g)
}
g is the gene name in each dataframe in the df_list
and when I run lapply(df_list,box),
I got Error: Mapping should be created with aes()oraes_().
Does anyone know how to fix this? Thank you.

Answer 1

As you used dplyr pipe, the first argument of ggplot() is already filled, leading df to be understood as aes argument.
box <- function(df){
df %>%
ggplot(aes(x = df[,4], y = df[,2])) +
geom_boxplot() +
ggtitle(g)
}

Answer 2

Instead of using df[,4] or df[,2] use column names in aes. Assuming the column names on x-axis is col1 and that on y-axis is col2 try -
box <- function(df, g){
df %>%
ggplot(aes(x = col1, y = col2)) +
geom_boxplot() +
ggtitle(df$g[1])
}
lapply(df_list,box)
g is the column name in the dataframe, so we can take the first value from it in title.

Answer 3

This is how to do it with another example dataset (Species instead of genes):
library(tidyverse)
plots <-
iris %>%
pivot_longer(-Species) %>%
nest(-Species) %>%
mutate(
plt = data %>% map2(Species, ~ {
.x %>%
ggplot(aes(name, value)) +
geom_boxplot() +
labs(title = .y)
})
) %>%
pull(plt)
#> Warning: All elements of `...` must be named.
#> Did you want `data = c(name, value)`?
plots[[1]]
Created on 2021-09-09 by the reprex package (v2.0.1)