how can I take some info out of text

Question

I have a text files like this

sp|O15304|SIVA_HUMAN MPKRSCPFADVAPLQLKVRVSQRELSRGVCAERYSQEVFEKTKRLLFLGAQAYLDHVWDEGCAVVHLPESPKPGPTGAPRAARGQMLIGPDGRLIRSLGQASEADPSGVASIACSSCVRAVDGKAVCGQCERALCGQCVRTCWGCGSVACTLCGLVDCSDMYEKVLCTSCAMFET IGPDGR
tr|A0A1B1L9R9|A0A1B1L9R9_BACTU MNKQLFLASLKETQKSILSYACGAALYLWLLIWIFPSMVSAKGLNELIAAMPDSVKKIVGMESPIQNVMDFLAGEYYSLLFIIILTIFCVTVATHLIARHVDKGAMAYLLATPVSRVQIAITQATVLILGLLIIVSVTYVAGLVGAEWFLQDNNLNKELFLKINIVGGLIFLVVSAYSFFFSCICNDERKALSYSASLTILFFVLDMVGKLSDKLEWMKNLSLFTLFRPKEIAEGAYNIWPVSIGLIAGALCIFIVAIVVFKKRDLPL NKELFLKINIVGGLIFLVVSAYSFFFSCICNDERKALSYSASLTILFFVLDMVGKLSDKLEWM

I am trying to lower case the third part of the two set. I tried the following but does not work

awk '{ gsub($3, tolower($3)); print $1"\t"$2}'

I have a Mac, is there any other way to do that ?

Answer 1

You're splitting on the default awk delimiter to get $1 and $2. Then you need to split $1 on "|" and lowercase the 3rd part of $1?

$awk '{split($1,a,"|") ; print a[1] "|" a[2] "|" tolower(a[3]) "\t" $2 "\t" $3}' test.txt

sp|O15304|siva_human    MPKRSCPFADVAPLQLKVRVSQRELSRGVCAERYSQEVFEKTKRLLFLGAQAYLDHVWDEGCAVVHLPESPKPGPTGAPRAARGQMLIGPDGRLIRSLGQASEADPSGVASIACSSCVRAVDGKAVCGQCERALCGQCVRTCWGCGSVACTLCGLVDCSDMYEKVLCTSCAMFET
tr|A0A1B1L9R9|a0a1b1l9r9_bactu  MNKQLFLASLKETQKSILSYACGAALYLWLLIWIFPSMVSAKGLNELIAAMPDSVKKIVGMESPIQNVMDFLAGEYYSLLFIIILTIFCVTVATHLIARHVDKGAMAYLLATPVSRVQIAITQATVLILGLLIIVSVTYVAGLVGAEWFLQDNNLNKELFLKINIVGGLIFLVVSAYSFFFSCICNDERKALSYSASLTILFFVLDMVGKLSDKLEWMKNLSLFTLFRPKEIAEGAYNIWPVSIGLIAGALCIFIVAIVVFKKRDLPL

Answer 2

Use a read into a variable declared as lowercase.

In all these examples I am printing the sections wrapped in square brackets ([]) so you can see how it's parsing, and I'm just putting spaces between. You can edit all that. The important part is to understand what defines the separations and to get the right part into the variable that will lowercase it.

declare -l three
while IFS='|' read -r one two three
do echo "[$one] [$two] [$three]"
done < infile
[sp] [O15304] [siva_human mpkrscpfadvaplqlkvrvsqrelsrgvcaerysqevfektkrllflgaqayldhvwdegcavvhlpespkpgptgapraargqmligpdgrlirslgqaseadpsgvasiacsscvravdgkavcgqceralcgqcvrtcwgcgsvactlcglvdcsdmyekvlctscamfet igpdgr]
[tr] [A0A1B1L9R9] [a0a1b1l9r9_bactu mnkqlflaslketqksilsyacgaalylwlliwifpsmvsakglneliaampdsvkkivgmespiqnvmdflageyysllfiiiltifcvtvathliarhvdkgamayllatpvsrvqiaitqatvlilglliivsvtyvaglvgaewflqdnnlnkelflkinivggliflvvsaysfffscicnderkalsysasltilffvldmvgklsdklewmknlslftlfrpkeiaegayniwpvsigliagalcifivaivvfkkrdlpl nkelflkinivggliflvvsaysfffscicnderkalsysasltilffvldmvgklsdklewm]

If you only want the part after the pipe, but before the space - and if the format is consistent -

declare -l three
while IFS='| ' read -r one two three four
do echo "[$one] [$two] [$three] [$four]"
done < infile
[sp] [O15304] [siva_human] [MPKRSCPFADVAPLQLKVRVSQRELSRGVCAERYSQEVFEKTKRLLFLGAQAYLDHVWDEGCAVVHLPESPKPGPTGAPRAARGQMLIGPDGRLIRSLGQASEADPSGVASIACSSCVRAVDGKAVCGQCERALCGQCVRTCWGCGSVACTLCGLVDCSDMYEKVLCTSCAMFET IGPDGR]
[tr] [A0A1B1L9R9] [a0a1b1l9r9_bactu] [MNKQLFLASLKETQKSILSYACGAALYLWLLIWIFPSMVSAKGLNELIAAMPDSVKKIVGMESPIQNVMDFLAGEYYSLLFIIILTIFCVTVATHLIARHVDKGAMAYLLATPVSRVQIAITQATVLILGLLIIVSVTYVAGLVGAEWFLQDNNLNKELFLKINIVGGLIFLVVSAYSFFFSCICNDERKALSYSASLTILFFVLDMVGKLSDKLEWMKNLSLFTLFRPKEIAEGAYNIWPVSIGLIAGALCIFIVAIVVFKKRDLPL NKELFLKINIVGGLIFLVVSAYSFFFSCICNDERKALSYSASLTILFFVLDMVGKLSDKLEWM]

If all you want is that LAST bit after the spaces downcased, then the default delimiter is fine.

declare -l three
while read -r one two three
do echo "[$one] [$two] [$three]"
done < infile
[sp|O15304|SIVA_HUMAN] [MPKRSCPFADVAPLQLKVRVSQRELSRGVCAERYSQEVFEKTKRLLFLGAQAYLDHVWDEGCAVVHLPESPKPGPTGAPRAARGQMLIGPDGRLIRSLGQASEADPSGVASIACSSCVRAVDGKAVCGQCERALCGQCVRTCWGCGSVACTLCGLVDCSDMYEKVLCTSCAMFET] [igpdgr]
[tr|A0A1B1L9R9|A0A1B1L9R9_BACTU] [MNKQLFLASLKETQKSILSYACGAALYLWLLIWIFPSMVSAKGLNELIAAMPDSVKKIVGMESPIQNVMDFLAGEYYSLLFIIILTIFCVTVATHLIARHVDKGAMAYLLATPVSRVQIAITQATVLILGLLIIVSVTYVAGLVGAEWFLQDNNLNKELFLKINIVGGLIFLVVSAYSFFFSCICNDERKALSYSASLTILFFVLDMVGKLSDKLEWMKNLSLFTLFRPKEIAEGAYNIWPVSIGLIAGALCIFIVAIVVFKKRDLPL] [nkelflkinivggliflvvsaysfffscicnderkalsysasltilffvldmvgklsdklewm]

Answer 3

So the question is how to correctly use the 3rd field as a pattern to do a sub in the rest of the string, and also how to send the output of the join to the awk command. Note that the gsub should have a target, in case field 3 is e.g. a single character, and that would also match and replace anything in $1.

join df1.txt df2.txt | awk '{gsub($3, tolower($3), $2) ; print $1 "\t" $2}'

To show an example, with and without the target:

ian@orca:~/tmp$ cat t
sp|O15304|SIVA_HUMAN FALALALALA A

ian@orca:~/tmp$ awk '{gsub($3, tolower($3)) ; print $1 "\t" $2}' t
sp|O15304|SIVa_HUMaN    FaLaLaLaLa

ian@orca:~/tmp$ awk '{gsub($3, tolower($3), $2) ; print $1 "\t" $2}' t
sp|O15304|SIVA_HUMAN    FaLaLaLaLa

Answer 4

 sed -rn 's/(.*\s.*\s)(.*)$/\1 \L\2 /p' tmp.txt

Sources:

• upper- to lower-case using sed
• How to output only captured groups with sed?
• https://superuser.com/questions/112834/how-to-match-whitespace-in-sed

Explanation:

I do not know awk well and it likely is possible to do this with awk as well. sed takes each line on its own and:

's/    substitutes
(      a group
  .*     containing any characters of any amount
  \s     a whitespace
  .*     again some characters
  \s     again a whitespace
)      and stores that group as \1
(.*)   and puts all the remaining characters in group \2
$      until the end of the line
/      Substitute all of this with:
\1     The first group
       a space (you might not want that. then remove it.
\L\2   The second group in lowercase
/p     and print that

The flag -r is neccessary to enable the capture of groups. The -n flag tells sed not to pring every line by itself already.

Tested on cygwin. Perhaps you need the -e flag on your OS. Perhaps you need to use the POSIX compliant [[:space:]] instead of \s for whitespace.

Answer 5

Try something like this:

cat text.txt | cut -d"|" -f3