shuffleseq from EMBOSS shuffles a set of sequences maintaining composition.

here's a function in python that "mutates" the original sequence, maintaining gc, at content.

import random

def seq_shuffler(original_seq="ACCAACXTGGGGTTTCCGGGGCCCCC"):
    original_seq = list(original_seq)
    while True:
        random.shuffle(original_seq)
        yield "".join(original_seq)

random_seq_gen = seq_shuffler()
print random_seq_gen.next()
print random_seq_gen.next()
print random_seq_gen.next()
print random_seq_gen.next()

# or loop.
for k in random_seq_gen:
    print k

OK I got a bit obsessed with doing this in R because I thought you could do it in one line, which you can !! (not including the input that is)

#input string of choice
a <- 'agcactacgactacgacagcata';

#shuffle it
paste(sample(unlist(strsplit(a,split=''))),collapse='');

and to do this 100 times and print out the answer:-

for(i in 1:100){
    print(paste(sample(unlist(strsplit(a,split=''))),collapse=''),q=F);
}

While the EMBOSS solution is probably the best, if it needs to be incorporated into a script, the Bioruby library gives you the very convenient 'randomize' method:

require 'bio'
s = Bio::Sequence::NA.new("ACCAACXTGGGGTTTCCGGGGCCCCC")
s.randomize         # ==> "tagccggcctxgatcactgcgcgccg"

What language are you using? Here's something in Ruby:

class Array
  #fisher-yates/knuth shuffle
  def shuffle!
    n = length
    for i in 0...n
      r = Kernel.rand(n-i)+i
      self[r], self[i] = self[i], self[r]
    end
    self
  end

  # Return a shuffled copy of the array
  def shuffle
    dup.shuffle!
  end
end

string = "AAATTTGGGCCC"
string.split(//).shuffle.join("")

> "AACGCTTTCAGG"

As always, there may be a more concise way to do this, but this will get the job done.

Hi Jess,

You can use Sean Eddy's Squid lib from Sean Eddy to do this. It's will generate a set of command-line application able to shuffle you sequence in several ways. Additionally, you can use uShuffle which will do a similar job. Both can shuffle preserving the base counts and preserving n-base (dibase, tribase, etc.) counts too.

also in perl without modules

my $seq = "AAAAAGTATACAACATCA"; #input seq
my @seqarray = split(//,$seq); #put seq in array
my @randarray = sort {rand() <=> rand()} @seqarray; #suffle indexes
my $outseq = join("",@randarray); #join shuffled sequence
print "$outseq\n"; #output

note that the perl sort function compares 2 numbers by <=> and returns -1, 0 or 1 depending which one is larger. if you sort by rand()<=>rand() then the sorting is random.

If you fancy doing it in Perl there are three different ways you can try listed here

There are a whole set of web-based tools available for this at http://www.bioinformatics.org/sms2/ This would be fine for the one-off or small set of sequences or for one who does not run perl or have access to tools found at a large institution. Nonetheless, the code examples above are also a way to learn...