Wildcards for filter function in dplyr

Question

I am using dplyr and I would like to filter my dataframe (biotypes) according to sample IDs which are the first column of the data frame, e.g. they look like this:

ID
chrX.tRNA494-SerAGA 
chrX.tRNA636-AlaCGC
mmu_piR_000007
...

I want to filter IDs starting with "chr" from IDs starting with "mmu":

biotype<- biotype %>% 
  filter( str_detect (biotype, "^chr") ==TRUE )
biotype

Can anyone help please? I am just looking for something like * that allows me to filter all rows that have a string starting with these particular characters ...

So your desired output would include `chrX.tRNA494-SerAGA` and `chrX.tRNA636-AlaCGC` but not include `mmu_piR_000007`? Is that correct? (It's always helpful to include example input and output in your question.) — Loren, Dec 02 '17 at 15:29
Yes exactly. Sorry you are right I should have been more precise — Anna, Dec 02 '17 at 16:31

score 6 · Answer 1 · answered Dec 04 '17 at 04:42

6

I think you were very close already.

library(stringr)
biotype %>% filter(str_detect(ID,"^chr"))

(you need to specify the column name, and == TRUE is superfluous).

answered Dec 04 '17 at 04:42

Ben Bolker

211,554
25
370
453

score 3 · Answer 2 · edited Dec 04 '17 at 05:54

3

What about grepl?

biotype <- biotype %>%
    filter(grepl('^chr', ID))

edited Dec 04 '17 at 05:54

smci

32,567
20
113
146

answered Dec 02 '17 at 22:13

Birger

1,111
7
17

Wildcards for filter function in dplyr

2 Answers2

Linked