Extracting Sequences After "Motif" & Between Motifs In Multifasta File

Question

[Raghul]

Hi I want to extract sequences after a motif say "TTTTTAAAAA" from a multifasta file. I do not want the nucleotides before this keyword. Is it possible to extract nucleotides between 2 motifs with grep? eg. nucleotides between TTTTTAAAA & AAAATTTT. I tried with grep but I need the fasta headers also. Can anybody suggest a solution in grep (if possible) or perl or python.

thanx raghul

[Manu Prestat]

You can get a case with the motif found several times within a same sequence. How do you want to deal with that?

[etarisal]

Hello!, I would like to do something similar...did you find a way to complete your task?

Answer 1

[Ying W]

I don't think it would be possible with grep but this can be done w/a regex in perl. Something along the lines of:

$line = "";
foreach(<FILE>) { #for every line of the file
  chomp;
  if($_[0] == ">") { #if line starts with >, it is a header so process the previous sequence
    if($line =~ /[TTTTTAAAAA([ACTGN]+)AAAATTTT/g) { #regex to match motif
      print "$1\n" #print sequence in between motif
    }
   $line = ""
    print "$_"; #print header
  }
  else {
    $line = $line.$_ #append sequence
  }
}
if($line =~ /[ACTGN]*TTTTTAAAAA([ACTGN]+)AAAATTTT/g) {
  print "$1\n"
}

or something like that, (warning above code is untested and should be treated as pseudocode)

[Neilfws]

Some (many?) versions of grep, such as the "standard" version included in Linux distributions, take the option "-P" meaning "interpret regex as a Perl regex". So if Perl can do it, so can grep.

Answer 2

[PoGibas]

grep way

  echo NNNTTTTTAAAACCCAAAATTTTNNN > sequence
  grep -o TTTTTAAAA[A-Z]*AAAATTTT sequence 
  TTTTTAAAACCCAAAATTTT