Manipulating a dataframe (with R)

Question

I have a difficult aim to reach in order to facilitate my analyses; to the best of my knowledge there are no similar questions. I have a very long dataframe in Excel, which I reproduce here - in a simpler form - in R environment:

A1 <- cbind("sp1","sp2","sp3", "sp4", "sp7", "sp8") 
A2 <- cbind("sp1","sp3", "sp4", "sp7", "sp9") 
A3 <- cbind("sp5","sp6","sp7", "sp10") 
A4 <- cbind("sp1","sp2","sp7", "sp9", "sp10") 
A5 <- cbind("sp3","sp4") 

max_row <- 6

A1 <- c(A1, rep(NA, max_row - length(A1)))
A2 <- c(A2, rep(NA, max_row - length(A2))) 
A3 <- c(A3, rep(NA, max_row - length(A3))) 
A4 <- c(A4, rep(NA, max_row - length(A4))) 
A5 <- c(A5, rep(NA, max_row - length(A5))) 
df <-cbind(A1,A2, A3, A4, A5)
df <- as.data.frame(df)
df <- data.frame(lapply(df, as.character), stringsAsFactors=FALSE)

To better understand the context in which I work, 'sp' are species, and A* are the sites where I detected a given species.

I want to convert this dataframe to another one structured as follow:

The first column contains the names of the sites, and the following ones are all the species names (obviously, repeated only one time). Then, I need to assign '1' for the presence, and '0' for the absence in a given site.

I spent many many hours to try to reach my aim, but it is a problem too complex for my R syntax capacities.

Anyone could kindly help me?

Answer 1

You can use gather and spread from tidyverse:

library(tidyverse)

df %>%
  gather(A, sp) %>%
  filter(!is.na(sp)) %>%
  group_by(A, sp) %>%
  count() %>%
  spread(sp, n) %>%
  replace(., is.na(.), 0)

  # A tibble: 5 x 11
# Groups:   A [5]
  A       sp1  sp10   sp2   sp3   sp4   sp5   sp6   sp7   sp8   sp9
* <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 A1       1.    0.    1.    1.    1.    0.    0.    1.    1.    0.
2 A2       1.    0.    0.    1.    1.    0.    0.    1.    0.    1.
3 A3       0.    1.    0.    0.    0.    1.    1.    1.    0.    0.
4 A4       1.    1.    1.    0.    0.    0.    0.    1.    0.    1.
5 A5       0.    0.    0.    1.    1.    0.    0.    0.    0.    0.

Answer 2

You can gather your data in long format to process and add column showing presence of species on a site. Afterwards use reshape2::dcast to spread the data in wide format as:

library(tidyverse)
library(reshape2)

df %>% gather(Site, Species) %>%
  filter(!is.na(Species)) %>%
  mutate(value = 1) %>%      #Species are present on a site
  dcast(Site~Species, value.var = "value", fill = 0)

#   Site sp1 sp10 sp2 sp3 sp4 sp5 sp6 sp7 sp8 sp9
# 1   A1   1    0   1   1   1   0   0   1   1   0
# 2   A2   1    0   0   1   1   0   0   1   0   1
# 3   A3   0    1   0   0   0   1   1   1   0   0
# 4   A4   1    1   1   0   0   0   0   1   0   1
# 5   A5   0    0   0   1   1   0   0   0   0   0