Replace starting from the second column

Question

Well, I have the following file:

12721   2 2 2 1 1 2 1 2 1 1 1 2 2 1 1 2 1 2 1 1 2
12722   2 2 2 1 1 2 1 2 1 1 1 2 2 1 1 2 1 2 1 1 2
12734   2 2 2 1 1 2 1 2 1 1 1 2 2 1 1 2 1 2 1 1 2
12753   2 2 2 1 1 2 1 2 1 1 1 2 2 1 1 2 1 2 1 1 2
12756   2 2 2 1 1 2 1 2 1 1 1 2 2 1 1 2 1 2 1 1 2

I need to remove the spaces starting from the second column, so that my file looks like this:

12721 222112121112211212112
12722 222112121112211212112
12734 222112121112211212112
12753 222112121112211212112
12756 222112121112211212112

I tried this command to replace:

sed '1,$s/ //g' snpdata > snpdata1

it did not work and I got it:

12721222112121112211212112
12722222112121112211212112
12734222112121112211212112
12753222112121112211212112
12756222112121112211212112

any suggestions to replace starting from the second column?

Note: my original dataset has thousands of columns and rows.

you seems to be changed you expected output now :) let me tweak my code too now. — RavinderSingh13, Jun 06 '18 at 13:27
If you have access to a GNU sed, you may use `sed 's/[[:space:]]//2g'` (or `sed -r 's/[[:space:]]+//2g'`) — Wiktor Stribiżew, Jun 06 '18 at 13:30
Possible duplicate of [Have sed make substitute on string but SKIP first occurrence](https://stackoverflow.com/questions/2166007/have-sed-make-substitute-on-string-but-skip-first-occurrence) — Sundeep, Jun 06 '18 at 14:20

RavinderSingh13 · Accepted Answer · 2018-06-06T13:29:42.253

1

EDIT: Since OP changed expected output now, so tweaked code a bit this should help OP for latest expected output.

awk '{val=$1;$1="";gsub(/[[:space:]]+/,"");print val,$0}'  Input_file

Following awk may help you here.

awk '{val=substr($0,5);gsub(/ +/,"",val);print substr($0,1,4), val;val=""}' Input_file

edited Jun 06 '18 at 13:29

answered Jun 06 '18 at 13:18

RavinderSingh13

130,504
14
57
93

oliv · Answer 2 · 2018-06-06T13:52:49.040

1

Using sed:

 sed 's/ //g;s/\([0-9]\{5\}\)\([0-9]\+\)/\1 \2/' file

where the first command removes all spaces and the second command groups the digits

edited Jun 06 '18 at 13:52

answered Jun 06 '18 at 13:38

oliv

12,690
25
45

Sorry, I edit my question. Please, Can you get your program right? – Greg Rov Jun 06 '18 at 13:50

score 1 · Answer 3 · answered Jun 06 '18 at 14:21

1

This might work for you (GNU sed):

sed 's/\([^ ]\)  */\1\n/1;s/ //g;s/\n/ /' file

Replace the first set of spaces following a column, by a newline. Remove all other spaces. Replace the newline by a space.

answered Jun 06 '18 at 14:21

potong

55,640
6
51
83

Replace starting from the second column

3 Answers3