Extract the last word between | |

Question

I have the following dataset

> head(names$SAMPLE_ID)
[1] "Bacteria|Proteobacteria|Gammaproteobacteria|Pseudomonadales|Moraxellaceae|Acinetobacter|"
[2] "Bacteria|Firmicutes|Bacilli|Bacillales|Bacillaceae|Bacillus|"                            
[3] "Bacteria|Proteobacteria|Gammaproteobacteria|Pasteurellales|Pasteurellaceae|Haemophilus|" 
[4] "Bacteria|Firmicutes|Bacilli|Lactobacillales|Streptococcaceae|Streptococcus|"             
[5] "Bacteria|Firmicutes|Bacilli|Lactobacillales|Streptococcaceae|Streptococcus|"             
[6] "Bacteria|Firmicutes|Bacilli|Lactobacillales|Streptococcaceae|Streptococcus|"

I want to extract the last word between || as a new variable i.e.

Acinetobacter
Bacillus
Haemophilus

I have tried using

library(stringr)
names$sample2 <-   str_match(names$SAMPLE_ID, "|.*?|")

The easy route: `vapply(strsplit(names$SAMPLE_ID, "|", fixed = TRUE), tail, "", 1)` — Rich Scriven, Dec 17 '15 at 19:21
Or of you don't like typing (or efficiency) then `sapply(strsplit(x, "\\|"), tail, 1)` — Rich Scriven, Dec 17 '15 at 19:31

score 5 · Accepted Answer · answered Dec 17 '15 at 19:10

5

We can use

library(stringi)
stri_extract_last_regex(v1, '\\w+')
#[1] "Acinetobacter"

data

v1 <- "Bacteria|Proteobacteria|Gammaproteobacteria|Pseudomonadales|Moraxellaceae|Acinetobacter|"

answered Dec 17 '15 at 19:10

akrun

874,273
37
540
662

1

seems that `stringi` package is better than `stringr` in some aspects. – AnilGoyal Apr 20 '21 at 06:05

score 5 · Answer 2 · answered Dec 17 '15 at 19:12

5

Using just base R:

myvar <- gsub("^..*\\|(\\w+)\\|$", "\\1", names$SAMPLE_ID)

answered Dec 17 '15 at 19:12

Zelazny7

39,946
18
70
84

score 3 · Answer 3 · edited Dec 17 '15 at 19:30

^.*\\|\\K.*?(?=\\|)

Use \K to remove rest from the final matche.See demo.Also use perl=T

https://regex101.com/r/fM9lY3/45

x <- c("Bacteria|Firmicutes|Bacilli|Lactobacillales|Streptococcaceae|Streptococcus|",
       "Bacteria|Firmicutes|Bacilli|Lactobacillales|Streptococcaceae|Streptococcus|" )

unlist(regmatches(x, gregexpr('^.*\\|\\K.*?(?=\\|)', x, perl = TRUE)))
# [1] "Streptococcus" "Streptococcus"

score 3 · Answer 4 · 2015-12-17T19:37:32.490

3

The ending is all you need [^|]+(?=\|$)

Per @RichardScriven :

Which in R would be regmatches(x, regexpr("[^|]+(?=\\|$)", x, perl = TRUE)

edited Dec 17 '15 at 19:37

answered Dec 17 '15 at 19:34

Frankie · Answer 5 · 2018-04-12T08:07:47.067

1

You can use package "stringr" as well in this case. Here is the code:

v1<- str_replace_all(v, "\\|", " ")

word(v1,-2)

Here I used v as the string. The basic theory is to replace all the | with spaces, and then get the last word in the string by using function word().

edited Apr 12 '18 at 08:07

answered Feb 06 '17 at 22:11

Frankie

61
3

Extract the last word between | |

5 Answers5

data

Linked

Related