trim fasta header, remove after hash (#)

Question

[mthm]

this is my fasta file

>BEL-3_Dvir#LINE/BelPao
GTAATTCTTGTGGTAGTTTTATGTTCTATGCAGGTTCAATATCGTACGCGCTCTGTGATG
CCCGCGCAGTACGCGTTTGTTTCCGATAGTTGATAACAAGTAATCGGTACAAATCGATAT
>P-2_Dmel#LINE/P
AATCTCTATATATAAAACTGTTTGTCCTGACTGACTGACTGACTGACTGACTGACTGACT
GACTGATTGGTGATCAACGCACAGCCCAAACCGTAAGAGCTAGGAAGCTGAAATTTTCAC
TGTAGCTACCTTATGTGATGTAGGTGCACGTTAAGACGGGGTTTCGGGAAATTCCACCCG

I want to remove the header after the hash:

>BEL-3_Dvi
GTAATTCTTGTGGTAGTTTTATGTTCTATGCAGGTTCAATATCGTACGCGCTCTGTGATG
CCCGCGCAGTACGCGTTTGTTTCCGATAGTTGATAACAAGTAATCGGTACAAATCGATAT
>P-2_Dme
AATCTCTATATATAAAACTGTTTGTCCTGACTGACTGACTGACTGACTGACTGACTGACT
GACTGATTGGTGATCAACGCACAGCCCAAACCGTAAGAGCTAGGAAGCTGAAATTTTCAC
TGTAGCTACCTTATGTGATGTAGGTGCACGTTAAGACGGGGTTTCGGGAAATTCCACCCG

I tried this code

sed -r -i 's/^>/.*#/\1/' file.fasta

the error is

 -e expression #1, char 10: unknown option to `s'

how to fix this?

[lieven.sterck]

also, be extremely careful when using the -i option ( the in-file modification parameter) , if something goes awol you lost your input file. Always work on a copy of the original file or send the sed output to a new file.

Answer 1

[cpad0112]

$ awk '/^>/ {sub(/#.*$/,"",$0)}1' test.fa
$ awk -F "#" '{print $1}' test.fa
$ sed -r '/^>/ s/#.*//' test.fa
$ cut -d '#' -f 1 test.fa

Code you pasted has several issues. Never use -i if you are not sure of code.