strings in rows of a data frame column

[eriklarsen4]

Hi! Hoping to solve this.. but need help

I'm trying to plot something in ggplot that is a bit tricky (to me). I've got a subsetted data frame and I'm characterizing each row in it by one of 10 "classes" (we'll call them identities). Each row can have any combination of the 10 identities. I want to display a value of each row (another column in the data frame) but by identity. So it would be something like a clustered bar graph, and each cluster/identity has a different, often redundant, number of rows within its cluster.. I tried creating a list of strings and putting them in their corresponding rows within a column at the end to then use as my mapping and filling variable. But R is only recognizing the first string in the list of the column.

# New column with each row containing a list of strings describing the row (Gene)
CellType$subtype = character(length = length(CellType$GeneID))


CellType$subtype[which(CellType$GeneID=="Gfra2")]= c("Tyrosine Hydroxylase", "Non-peptidergic 2")
CellType$subtype[which(CellType$GeneID=="Mrgpra3")]= c("Non-peptidergic 2")
CellType$subtype[which(CellType$GeneID=="Mrgprd")]= c("Non-peptidergic 1")
CellType$subtype[which(CellType$GeneID=="Sst")]= c("Non-peptidergic 3")
CellType$subtype[which(CellType$GeneID=="Piezo2")]= c("Tyrosine Hydroxylase")
CellType$subtype[which(CellType$GeneID=="Ldhb")]= c("Neurofilament 1", "Neurofilament 2", "Neurofilament 3", "Neurofilament 4", "Neurofilament 5")
CellType$subtype[which(CellType$GeneID=="Cacna1h")]= c("Neurofilament 1", "Neurofilament 2")
CellType$subtype[which(CellType$GeneID=="Necab2")]= c("Neurofilament 2")
CellType$subtype[which(CellType$GeneID=="Fam19a1")]= c("Neurofilament 3", "Peptidergic 2")

Sub_pop_cluster = ggplot(CellType, aes(x = CellType$subtype, y = CellType$log2.FC.)) + geom_bar(aes(fill = CellType$GeneID), position = "dodge", stat = "identity")

So when R plots this, it is recognizing, from the first example, Gene Gfra2 as "Tyrosine hydroxylase", instead of both "Tyrosine hydroxylase" and "Non-peptidergic 2".

Is there a way to fix this?

Thanks!

[andresrcs]

Hi Erik, welcome!
It would be easier to help if you provide some sample data, so could you please turn this into a self-contained REPRoducible EXample (reprex)? A reprex makes it much easier for others to understand your issue and figure out how to help.

If you've never heard of a reprex before, you might want to start by reading this FAQ:

FAQ: How to do a minimal reproducible example ( reprex ) for beginners

A minimal reproducible example consists of the following items: A minimal dataset, necessary to reproduce the issue The minimal runnable code necessary to reproduce the issue, which can be run on the given dataset, and including the necessary information on the used packages. Let's quickly go over each one of these with examples: Minimal Dataset (Sample Data) You need to provide a data frame that is small enough to be (reasonably) pasted on a post, but big enough to reproduce your issue. Let's say, as an example, that you are working with the iris data frame head(iris) #> Sepal.Length Sepal.Width Petal.Length Petal.Width Species #> 1 5.1 3.5 1.4 0.…

[nathania]

Hi! If I'm understanding you correctly, reshaping your data as follows might be what you're looking for.

The second left_join() isn't strictly necessary, but without NAs the columns in your plot might have different widths.

library(tidyverse)
set.seed(123)

gene_to_subtype <- list(Gfra2= c("Tyrosine Hydroxylase", "Non-peptidergic 2"),
                        Mrgpra3 = c("Non-peptidergic 2"),
                        Mrgprd = c("Non-peptidergic 1"),
                        Sst = c("Non-peptidergic 3"),
                        Piezo2 = c("Tyrosine Hydroxylase"),
                        Cacna1h = c("Neurofilament 1", "Neurofilament 2"),
                        Necab2 = c("Neurofilament 2"),
                        Fam19a1 = c("Neurofilament 3", "Peptidergic 2"),
                        Ldhb = c("Neurofilament 1", "Neurofilament 2", "Neurofilament 3", "Neurofilament 4", "Neurofilament 5")) %>% 
  enframe(name = 'gene_id', value = 'subtype') %>% 
  unnest()

dummy_data <- distinct(gene_to_subtype, gene_id) %>% 
  mutate(log2.FC. = rnorm(n = length(gene_id), 1))

(df <- left_join(dummy_data, gene_to_subtype) %>% 
  left_join(expand(gene_to_subtype, gene_id, subtype), .))  # explicit NAs 
#> Joining, by = "gene_id"
#> Joining, by = c("gene_id", "subtype")
#> # A tibble: 90 x 3
#>    gene_id subtype              log2.FC.
#>    <chr>   <chr>                   <dbl>
#>  1 Cacna1h Neurofilament 1          2.72
#>  2 Cacna1h Neurofilament 2          2.72
#>  3 Cacna1h Neurofilament 3         NA   
#>  4 Cacna1h Neurofilament 4         NA   
#>  5 Cacna1h Neurofilament 5         NA   
#>  6 Cacna1h Non-peptidergic 1       NA   
#>  7 Cacna1h Non-peptidergic 2       NA   
#>  8 Cacna1h Non-peptidergic 3       NA   
#>  9 Cacna1h Peptidergic 2           NA   
#> 10 Cacna1h Tyrosine Hydroxylase    NA   
#> # … with 80 more rows

# ggplot(df, aes(x = gene_id, y = log2.FC.)) +
#   geom_col(aes(fill = subtype), position = 'dodge')

^{Created on 2019-03-13 by the reprex package (v0.2.1)}

[eriklarsen4]

Thanks so much. I think I see. When I tried to reproduce the code, the "enframe" function wasn't being recognized despite having tidyverse or tibble package installed.

[mara]

Could you please post a reprex of your code? I'm able to run @nathania's code successfully.

library(tidyverse)
set.seed(123)

gene_to_subtype <- list(Gfra2= c("Tyrosine Hydroxylase", "Non-peptidergic 2"),
                        Mrgpra3 = c("Non-peptidergic 2"),
                        Mrgprd = c("Non-peptidergic 1"),
                        Sst = c("Non-peptidergic 3"),
                        Piezo2 = c("Tyrosine Hydroxylase"),
                        Cacna1h = c("Neurofilament 1", "Neurofilament 2"),
                        Necab2 = c("Neurofilament 2"),
                        Fam19a1 = c("Neurofilament 3", "Peptidergic 2"),
                        Ldhb = c("Neurofilament 1", "Neurofilament 2", "Neurofilament 3", "Neurofilament 4", "Neurofilament 5")) %>% 
  enframe(name = 'gene_id', value = 'subtype') %>% 
  unnest()

dummy_data <- distinct(gene_to_subtype, gene_id) %>% 
  mutate(log2.FC. = rnorm(n = length(gene_id), 1))

(df <- left_join(dummy_data, gene_to_subtype) %>% 
    left_join(expand(gene_to_subtype, gene_id, subtype), .))
#> Joining, by = "gene_id"
#> Joining, by = c("gene_id", "subtype")
#> # A tibble: 90 x 3
#>    gene_id subtype              log2.FC.
#>    <chr>   <chr>                   <dbl>
#>  1 Cacna1h Neurofilament 1          2.72
#>  2 Cacna1h Neurofilament 2          2.72
#>  3 Cacna1h Neurofilament 3         NA   
#>  4 Cacna1h Neurofilament 4         NA   
#>  5 Cacna1h Neurofilament 5         NA   
#>  6 Cacna1h Non-peptidergic 1       NA   
#>  7 Cacna1h Non-peptidergic 2       NA   
#>  8 Cacna1h Non-peptidergic 3       NA   
#>  9 Cacna1h Peptidergic 2           NA   
#> 10 Cacna1h Tyrosine Hydroxylase    NA   
#> # … with 80 more rows

^{Created on 2019-03-15 by the reprex package (v0.2.1.9000)}

@eriklarsen4

despite having tidyverse or tibble package installed.

Note also that you must loat/attach the library in each session, so, if you haven't run library(tidyverse) (as in the reprex), the function enframe() will not be found.