How to arrange a dataframe on a chosen column, while keeping two specific rows at the top?-CodePudding

I have a table with many samples. I would like to have Positive & Negative controls at the top of the table, but need help figuring out the most straighforward way to do this.

Here is a minimal example dataframe:

df = data.frame(structure(list(Well = c("A1", "A2", "B1", "B2"), 
                               Sample = c("Asample 2", "Positive control", "Asample 1", "Negative control"))))

The task:

Positive & negative controls at the top of the table.
The rest of the table arranged by well and NOT the sample name.
Keep in mind that positive & negative controls are always named like this, but could have any random well id in real life situation, whereas sample names and well ids can vary.

The desired output:

  well  sample          
  <chr> <chr>           
1 A2    Positive control
2 B2    Negative control
3 A1    Asample 2       
4 B1    Asample 1

CodePudding user response：

Another solution:

df %>% arrange(match(Sample,c("Positive control", "Negative control")),Well)

  Well           Sample
1   A2 Positive control
2   B2 Negative control
3   A1        Asample 1
4   B1        Asample 2

CodePudding user response：

One solution would be to make sample a factor() using well levels, and then to relevel sample appropriately.

df %>% 
    arrange(Well) %>% 
    mutate(
        Sample = fct_relevel(
            fct_inorder(factor(Sample)), 
            c('Positive control', 'Negative control')
        )
    ) %>% 
    arrange(Sample)

  Well           Sample
1   A2 Positive control
2   B2 Negative control
3   A1        Asample 2
4   B1        Asample 1