Looping dplyr and creating multiple dataframe - r

Question

I have a large dataset in which I would like to use dplyr and filter and select the data to create 12 separate dataframes.
Essentially, I am using only two columns of data from a larger dataset. The first column is "plot", where I filter by "plot" number and another condition in another 3rd column ("pos_ID"). I want to create a loop that filters by plot number (I tried plot==[i]) and the 3rd condition, and then creates a new dataframe. The loop would repeat 12 times (because plot spans from 1-12).
Here is the code that I used without a loop (based on sample data)
p1_Germ <- data %>% #p1 stands for plot 1
filter(plot==1, pos_ID<21) %>%
select(germ_bin)
Here is the code that I tried to incorporate a loop (based on sample data)
for(i in seq_along(plot)) {
data %>%
group_by(plot[[i]], pos_ID<21) %>%
select(germ_bin)
}
Here is some sample data
plot <- c(1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12)
germ_bin <- c(0,0,1,0,1,0,0,1,1,0,1,1,0,1,0,1,0,1,1,0,1,0,1,0)
pos_ID <- c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24)
dataset <- data.frame(plot, germ_bin, pos_ID)
dataset
My guess is to use a list, but I'm not familiar with loops and list and could not find a solution online. I need to create 12 dataframes because I'm trying to convert them each into a matrix after for another function. Any helpful would be much appreciated!

Answer 1

We can use group_split and map to filter based on criteria to get list of dataframes.
library(dplyr)
library(purrr)
dataset %>%
group_split(plot) %>%
map(. %>% filter(pos_ID < 21) %>% select(germ_bin))
#[[1]]
# A tibble: 2 x 1
# germ_bin
# <dbl>
#1 0
#2 0
#[[2]]
# A tibble: 2 x 1
# germ_bin
# <dbl>
#1 1
#2 0
#[[3]]
# A tibble: 2 x 1
# germ_bin
# <dbl>
#1 1
#2 0
#....
For the shared example, if you want to drop empty groups you can filter first
dataset %>%
filter(pos_ID < 21) %>%
group_split(plot) %>%
map(. %>% select(germ_bin))
As far your attempt with for loop is concerned, you can correct that by doing
unique_plot <- unique(dataset$plot)
plot_list <- list(length = length(unique_plot))
for(i in seq_along(unique_plot)) {
plot_list[[i]] <- dataset %>%
filter(plot == unique_plot[i], pos_ID<21) %>%
select(germ_bin)
}
Or keeping it completely in base R
lapply(split(dataset, dataset$plot), function(x)
subset(x, pos_ID < 21, select = germ_bin, drop = FALSE))