This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Remove text and add a string in odd rows in a fasta file using awk

I have a file organized as such:

>Prevalence_Sequence_ID:13|ARO_Name:AxyX|ARO:3004143|Detection_Model:Protein Homolog Model
ATGAAGCAAAGAGTCCCTCTACGCACGTTCGTCCTATCTGCCGTATTAATTCTTATTACTGGTTGCTCGAAACCGGAAACCCAACCAGCCG
>Prevalence_Sequence_ID:14|ARO_Name:adeF|ARO:3004143|Detection_Model:Protein Homolog Model
ATGAATATCTCGAAATTCTTCATCGACCGGCCGATCTTCGCCGGCGTGCTTTCGATCCTGGTGTTGCTGGCGGGCATACTGGCCATGTTCC

For every odd row, i need to keep only the first and third column and add the text "|kraken:taxid|32630" at the end. Example below

>Prevalence_Sequence_ID:13|ARO:3004143|kraken:taxid|32630
ATGAAGCAAAGAGTCCCTCTACGCACGTTCGTCCTATCTGCCGTATTAATTCTTATTACTGGTTGCTCGAAACCGGAAACCCAACCAGCCG
>Prevalence_Sequence_ID:14|ARO:3004143|kraken:taxid|32630
ATGAATATCTCGAAATTCTTCATCGACCGGCCGATCTTCGCCGGCGTGCTTTCGATCCTGGTGTTGCTGGCGGGCATACTGGCCATGTTCC

Is there a simple awk script that i can use? In alternative i could also keep only the first column if it's easier

awk
• 1,405 views • link
updated by cpad0112 cpad0112 • written by Mgs3 • Mgs3 
$ awk -F '|' '/^>/ {print fields, "text";next}1' input.fa
• 0 views • link
written by cpad0112 cpad0112

2 answers

I guess something like

awk -F "|" '{if($1 ~/^>/) {print $1"|"$3"|kraken:taxid|32630"}else{print $0}}' test

Should work (assuming you have a fastq file and you are aiming to change the header, not just the odd rows)

• 0 views • link
written by Sam Sam

Is there a simple awk script that i can use?

yes: awk -F '|' '/^>/ {printf(TODO);next;} {print;}'

II leave the TODO part as an exercice.

• 0 views • link
written by Pierre Lindenbaum Pierre Lindenbaum

Log in to answer this question.

More posts like this