'Select columns based on string match - dplyr::select

I have a data frame ("data") with lots and lots of columns. Some of the columns contain a certain string ("search_string").

How can I use dplyr::select() to give me a subset including only the columns that contain the string?

I tried:

# columns as boolean vector
select(data, grepl("search_string",colnames(data)))

# columns as vector of column names names 
select(data, colnames(data)[grepl("search_string",colnames(data))])

Neither of them work.

I know that select() accepts numeric vectors as substitute for columns e.g.:

select(data,5,7,9:20)

But I don't know how to get a numeric vector of columns IDs from my grepl() expression.

rregexdplyr


Solution 1:[1]

You can try:

select(data, matches("search_string"))

It is more general than contains - you can use regex (e.g. "one_string|or_the_other").

For more examples, see: http://rpackages.ianhowson.com/cran/dplyr/man/select.html.

Solution 2:[2]

No need to use select just use [ instead

data[,grepl("search_string", colnames(data))]

Let's try with iris dataset

>iris[,grepl("Sepal", colnames(iris))]
  Sepal.Length Sepal.Width
1          5.1         3.5
2          4.9         3.0
3          4.7         3.2
4          4.6         3.1
5          5.0         3.6
6          5.4         3.9

Solution 3:[3]

Based on Piotr Migdals response I want to give an alternate solution enabling the possibility for a vector of strings:

myVectorOfStrings <- c("foo", "bar")
matchExpression <- paste(myVectorOfStrings, collapse = "|")
# [1] "foo|bar"
df %>% select(matches(matchExpression))

Making use of the regex OR operator (|)

ATTENTION: If you really have a plain vector of column names (and do not need the power of RegExpression), please see the comment below this answer (since it's the cleaner solution).

Solution 4:[4]

Alternatively using a DataFrame of 22 columns:

library(plyr) # for baseball dataset.
library(dplyr)

baseball %>% colnames() %>% length()
[1] 22

baseball %>% colnames()
 [1] "id"    "year"  "stint" "team"  "lg"    "g"     "ab"    "r"     "h"     "X2b"   "X3b"   "hr"    "rbi"  
[14] "sb"    "cs"    "bb"    "so"    "ibb"   "hbp"   "sh"    "sf"    "gidp"

You can use starts_with("s") and ends_with("b"):

> baseball %>% select(starts_with("s")) %>% head(5)
    stint sb so sh sf
4       1  6  1 NA NA # players.columns.str.startswith('p')
44      1  8  0 NA NA
68      1  2  0 NA NA 
99      1  4  0 NA NA
102     1  3  0 NA NA

> baseball %>% select(ends_with("b")) %>% head(5)
     ab X2b X3b sb bb ibb
4   120  11   3  6  2  NA
44  162   9   4  8  4  NA
68   89   3   1  2  2  NA
99  161   5   1  4  3  NA
102 128   3   7  3  1  NA

# contains("g") matches names that contain “g”.
> baseball %>% select(contains("g")) %>% head(5)
    lg  g gidp
4      25   NA
44     32   NA
68     19   NA
99     33   NA
102    29   NA

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1
Solution 2 Jilber Urbina
Solution 3
Solution 4 rubengavidia0x

Related Questions