How to change sequence description automatically in an MSA

Question

[jbt38]

I have a MSA (fasta format) with hundreds of sequences , and the descriptions are this format:

>gi|AY015275.1|taxonid|154401|organism|Leuenbergeria guamacho|seqid|AY015275.1|description|Pereskia guamacho tRNA-Lys (trnK) gene partial sequence; and maturase K (matK) gene complete cds; chloroplast genes for chloroplast products

How can I change the description of each entry to look like this?

>Leuenbergeria_guamacho

Edited to add an underscore between genus and species.

[cpad0112]

Assuming that scientific name is always sandwiched between organism and seqid:

$ seqkit seq test.fa -w 0  -i --id-regexp ".*organism\|(.*)\|seqid.*" | seqkit replace -p " " -r "_"

with sed:

$ sed -r '/^>/ s/(>).*organism\|(.*)\s(.*)\|seqid.*/\1\2_\3/' input.fa

Answer 1

[cschu181]

awk -F '|' '/^>/ { print ">"$6; next; } { print $0; }' fasta_file

You also might want to replace the space with a _

awk -F '|' '/^>/ { print ">"$6; next; } { print $0; }' fasta_file | tr " " "_"

[cpad0112]

$ awk -F '|' '/^>/ {gsub(/ /,"_",$6);print ">"$6;next}1' fasta_file may suffice to print scientific name separated by _

[cschu181]

fair enough, I struggle with awk internals..

[cpad0112]

@cshu181 My awk fundamentals came from Biostars like Pierre, Kevin and my struggle is as good as yours, if not worse.