How can I do this kind of filtering

[SDPA_Pet]

Hi I have a fasta file, sequence like this. Basically, it is an annotated files the sequences name include fuction name, and organism.

I want to do this kind of filtering.

1> extract the sequence name ("mgm4510423.3|contig02227|RefSeq|73954f841ecd7c512c5428ed1b1a747e accession=[NP_559375.1],function=[carbamate kinase],organism=[Pyrobaculum aerophilum str. IM2]")to a text file. I would be better to separate by comma. That is, make three columns. ID, function and organism.

2> After I create the upper text file. I can choose the organism that I want to keep. Filter the fasta files, so I will get all the sequences that I need for particular organisms.

Any software or Unix command like grep /awk can do this.

>mgm4510423.3|contig02227|RefSeq|73954f841ecd7c512c5428ed1b1a747e accession=[NP_559375.1],function=[carbamate kinase],organism=[Pyrobaculum aerophilum str. IM2]
AAGAAACGTCGACGTAGCCGCCAGAGTCGTGGCAGGGgTAATGCAGGGAGGCCACCAGGTGGTGGTGACGCACGGCAACGGGCCCCAGGTGGGCTACCTGGCGgAGTTGCaGAgaGACAACGGCACATTTCGGCTGGACGCCCTAAACGCCATGACGCaGGGgATGCTCGGCTACTTCCTTGTCTCTGCGCTTGATAAATACTTAGGCAGGGGGAGGGCCGCGGCTTTGGTGACCAGAGTCGAGGTGGACTGCGACGACCCGGCTTTTaaagaCCCGACcAAGTTCATAGGTCCCCTATACGGCAAGGAaCaGgCTGAGGCCCTCGCACAGAGGTACGGGTGGCAGTTTAGGCAAGACCCAAGAGGAGGCTGGCgtCGCGTCGTCGCGTCGCCTACGCCGCTCAGAAtcGTGGAGATAGAGGCCGTAAAGaGGTTGCTGgACGCGgGTTTCGTCGTTGTGGCGgCGGGCGGCGGCGGTaTACCGCTCTGCGGAGACAGAgaCGTAGAGGGGGTTATAGACAAGGACTTGGCCTCTTCTCTCCTCGCTGTGGAGCTCGGCGCGGACTTCTTCATGATGCTGACCGACATAGACGCCGTCTACCTAAACTACGGGAaGCCGAACCAGAGGAGGCTAGACAGCGTAGGGGCTGACGAGCTGGAGAGGTATTTCGCCGAtGGCcACTTCCCGCCGGGCTCCATGGGGCCGAAGGTGCAGGCCGCGATAAACTTCGTGAAacAAAcGGggaGAaGGGCGGCCATCGGGGCGCTGGAGGAGGGCTAtGACGtGTTCAGGGGAATAAAGGGGACCCAGGTgACGCCTTAGAGCTCGTTTATTGGCTTTTCGTATTCCTCCCTcTtCtGGAGGTCTCGgATCTTgACTACGCCGCGCTCCAGCTCTTTCTTGCCGATTATGATTAGGtACCGCGTGCCTATCTTCAAGATGTATTCAAAGGCCTcTTTtAGGCTTTTCTCGCCCAGCTCCACAGCCACGCTGAAGCCTGCGCTCCTCAGCTTCTtcGCAACTGCCACGGCCTGCGGGTACGCCTCgTCGTCGAAGATGTAGATGTAGTAGTCCAGCGGCTTCTCCACGTTGTGGAGCCCcACGgCCTCcATAAACcTCTCAACGCCGATGGCGAaCCCCAGCGCCGgCGtCtttACGCCGCTGTAGAGCT

[vivek_]

grep "^>" your_fasta > headers.txt

Gives you the headers

If your entire sequence is in one line, you can use

grep -A 1 "Pyrobaculum aerophilum str. IM2" > paerophilum.fasta

to select all sequences belonging to that organism.

[SDPA_Pet]

for the 2nd question. Is it possible to filter it by column? Let's say I have column list all the name of organism that I want to extract.

[vivek_]

You can try this

for organism in `cat your_organisms.txt`;do grep -A 1 -w $organism your_fasta.fa > $organism.fa;done