How to import a .tsv file

Question

I need to read a table that is a .tsv file in R.

test <- read.table(file='drug_info.tsv')
# Error in scan(file, what, nmax, sep, dec, quote, skip, nlines, na.strings,  : 
#   line 1 did not have 10 elements
test <- read.table(file='drug_info.tsv', )
# Error in scan(file, what, nmax, sep, dec, quote, skip, nlines, na.strings,  : 
#   line 1 did not have 10 elements
scan("drug_info.tsv")
# Error in scan(file, what, nmax, sep, dec, quote, skip, nlines, na.strings,  : 
#   scan() expected 'a real', got 'ChallengeName'
scan(file = "drug_info.tsv")
# Error in scan(file, what, nmax, sep, dec, quote, skip, nlines, na.strings,  : 
#   scan() expected 'a real', got 'ChallengeName'

How should I read it?

Please copy/paste the first 5 rows of the file into your question and remove the picture. — Rich Scriven, Oct 24 '15 at 19:53
`read.table` default to using a whitespace delimited (meaning space or tab generally). If you have spaces, you can explicitly set the delimiter as tab with `sep="\t"`. `read.table` works with valid input files, so if there is a problem importing your data, it's with the file, and not the function. So in order to help you, we'd need you to share a sample of the file you are actually trying to import, not a picture of the data in some other program. — MrFlick, Oct 24 '15 at 20:23

score 44 · Answer 1 · answered Oct 24 '15 at 19:26

44

This should do it:

read.table(file = 'drug_info.tsv', sep = '\t', header = TRUE)

answered Oct 24 '15 at 19:26

Robert

2,111
4
18
32

3

That should give the same error as reported, line 1 does not have sufficient elements – Robert Hijmans Oct 24 '15 at 19:48
1

I think the down-vote came a bit prematurely here, as we don't have any actual data to test with any method yet. – Rich Scriven Oct 24 '15 at 20:34

score 15 · Answer 2 · edited Mar 12 '18 at 06:24

15

Using fread from the package data.table will read the data and will skip the error you are getting using read.table.

require(data.table)

data<-as.data.frame(fread("drug_info.tsv"))

edited Mar 12 '18 at 06:24

Pang

9,564
146
81
122

answered Mar 12 '18 at 06:05

TBhavnani

721
7
12

Thumbs up for this solution as it can handle large data table avoiding session time out on the ShinyUI webpage – Stone Jan 29 '19 at 18:53
this is such a cool answer! thanks! – stats_noob Mar 25 '23 at 03:59

score 13 · Answer 3 · answered Feb 21 '19 at 22:39

13

You can treat the data like a csv, and specify tab delimination.

read.csv("drug_info.tsv", sep = "\t")

answered Feb 21 '19 at 22:39

Sam Old

142
1
7

score 5 · Answer 4 · answered Oct 24 '15 at 19:52

Assuming that only the first line does not have the right number of elements, and that this is the column names line. Skip the first line:

 d <- read.table('drug_info.tsv', skip=1)

Now read it

 first <- readLines('drug_info.tsv', n=1)

Inspect it, fix it such that its number of elements matches d and then

 colnames(d) <- first

If that does not work, you can do

 x <- readLines('drug_info.tsv')

and diagnostics like this:

 sapply(x, length)

score 5 · Answer 5 · answered Nov 15 '19 at 20:58

5

You need to include fill = TRUE.

test <- read.table(file='drug_info.tsv', sep = '\t', header = TRUE, fill = TRUE)

answered Nov 15 '19 at 20:58

woutcault

51
1
2

score 4 · Answer 6 · answered Jan 26 '19 at 15:43

utils::read.delim() is most commonly used in such case if you don't want to install other library. The sample code could be something like:

test <- read.delim(file='drug_info.tsv')

or much more friendly io functions could be available from readr library, where a read_tsv named function is available directly:

test <- readr::read_tsv('drug_info.tsv')

How to import a .tsv file

6 Answers6

Related