CODEWITHSUNDEEP
bashawk
Callie
Callie

Reputation: 343

Replace headers of a multi fasta file awk

I am trying to change the headers from a multi fasta file using awk.

My file looks as follows:

>NS500106:288:HGHTHAFXY:1:11101:16650:2011:CTATAC;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
>NS500106:288:HGHTHAFXY:1:11101:1149:1903:CTATAC;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

I am using this command:

awk '/[^;]*/{print "Variant_" ++i; next} {print}' < input.fasta > output.fasta

Where, /[^;]*/ matches lines upto the first occurence of the semi colon and replaces matched part of the header with Variant_1, Variant_2 and so on.

The above command changes the entire header to Variant_1 and so on and also does not print the sequence lines.

How can I resolve this?

My desired output is as below:

Variant_1;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
Variant_2;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

Upvotes: 0

Views: 1358

Answers (3)

stack0114106
stack0114106

Reputation: 8711

Check this Perl-one liner 

$ cat callie.txt
>NS500106:288:HGHTHAFXY:1:11101:16650:2011:CTATAC;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
>NS500106:288:HGHTHAFXY:1:11101:1149:1903:CTATAC;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

$ perl -ne ' BEGIN {$x=1} { s/(.*?);/"Variant_${x};"/e and $x++ ; print }' callie.txt
Variant_1;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
Variant_2;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

$

Even shorter

$ perl -pe ' { s/(.*?);/sprintf("Variant_%d;",++$x)/e; }' callie.txt
Variant_1;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
Variant_2;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

$

Upvotes: 1

oguz ismail
oguz ismail

Reputation: 50750

Where, /[^;]*/ matches lines upto the first occurence of the semi colon and replaces matched part of the header with Variant_1, Variant_2 and so on.

No, this is wrong. It maches lines containing zero or more non-semicolon chars (e.g all lines) and prints Variant_n where n is line number. {print} part is never reached.

Here is my approach:

$ awk 'BEGIN{FS=OFS=";"}NF>1{$1="Variant_"++i}1' file
Variant_1;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
Variant_2;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

Upvotes: 1

Joaquin
Joaquin

Reputation: 2091

You could try with: 

>> awk '/>.*/{sub(/[^;]*/,"Variant_" ++i )}1' < input.fasta  

Variant_1;size=206083;
GTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG
Variant_2;size=19175;
CAGCGAGAGCGAGACGCAGCCGGTGAACCGCGTGGCGTACAACGTCGGCGGGCAGATGGCCACCAACAACCAGAGCTCCA
CCACTGCCCCCGCGACCGGCACGTACAACCTCCAGGAAATCGTGCCCGGCAGCGTGTGGATGGAGAGGGACGTG

What you tried deletes all the line, in this case, sub solves that problem

Upvotes: 1

Related Questions