From: "Jesús Gabriel y Galán" <jgabrielygalan@...>
Date: 2010-10-28T01:07:11+09:00
Subject: Re: Extraction of single subarrays from multidimensional array

On Wed, Oct 27, 2010 at 2:05 PM, Maurizio Cirilli <mauricirl@gmail.com> wrote:
> Dear Robert,
>
>   this is I want to do:
>
> (1) extract subarrays from a md-arrays (but the number of subarrays
>     can vary from case to case)
> (2) use separated subarrays to make all possible nucleotide sequences
>     is possible to build from them permutating the codons at each
> position
>     (row) i.e. all the 6*2*6 sequences 9 nucleotide long in my
> example
> (3) convert them to string and put in a single array ( this is
> required for
>     compatibility with BioRuby classes and methods that deal withDNA
>     sequences as strings)
> (4) make further analysis on these sequences by BioRuby.

Looking around for permutations in Ruby, I came across
http://snippets.dzone.com/posts/show/3332
and adapting it a little bit for your use case, I came up with:

class Array
  def sequence(i = 0, *a)
    return a.join if i == size
    self[i].map {|x| sequence(i+1, *(a + [x]))}
  end
end

ss = [["tcg", "agt", "tct", "agc", "tca", "tcc"],
	 ["aaa", "aag"],
	 ["ctg", "tta", "ctt", "cta", "ctc", "ttg"]]

p ss.sequence.flatten

$ ruby permutations.rb
["tcgaaactg", "tcgaaatta", "tcgaaactt", "tcgaaacta", "tcgaaactc",
"tcgaaattg", "tcgaagctg", "tcgaagtta", "tcgaagctt", "tcgaagcta",
"tcgaagctc", "tcgaagttg", "agtaaactg", "agtaaatta", "agtaaactt",
"agtaaacta", "agtaaactc", "agtaaattg", "agtaagctg", "agtaagtta",
"agtaagctt", "agtaagcta", "agtaagctc", "agtaagttg", "tctaaactg",
"tctaaatta", "tctaaactt", "tctaaacta", "tctaaactc", "tctaaattg",
"tctaagctg", "tctaagtta", "tctaagctt", "tctaagcta", "tctaagctc",
"tctaagttg", "agcaaactg", "agcaaatta", "agcaaactt", "agcaaacta",
"agcaaactc", "agcaaattg", "agcaagctg", "agcaagtta", "agcaagctt",
"agcaagcta", "agcaagctc", "agcaagttg", "tcaaaactg", "tcaaaatta",
"tcaaaactt", "tcaaaacta", "tcaaaactc", "tcaaaattg", "tcaaagctg",
"tcaaagtta", "tcaaagctt", "tcaaagcta", "tcaaagctc", "tcaaagttg",
"tccaaactg", "tccaaatta", "tccaaactt", "tccaaacta", "tccaaactc",
"tccaaattg", "tccaagctg", "tccaagtta", "tccaagctt", "tccaagcta",
"tccaagctc", "tccaagttg"]

Which if I understood correctly solves your steps 1 to 3.

Hope this helps,

Jesus.