Bioinformatics coding question

[charm]

Perlmonks, This is the question I am having trouble answering

Use random DNA sequence generator as many times as you need to get the protein-coding region(s) (nucleotide triplets). Minimum sequence length is 500bp.

• Find all possible protein-coding regions in microbes (between start and stop codons, ATG – TAG|TGA|TAA).

• Using Standard Genetic Code table (Wikipedia or any other sources), create hash table for all amino acids ($amino{TTT} => "F", ...).

• Write found protein sequences into file with the next format:
   Position of 1st start codon: protein sequence [length of protein sequence]
   Position of 2nd start codon: ….

For example: 45: FLPQWCV [7]

I am sort of clueless and not sure where to start. To find the coding sequence I have generated a 5000bp random nucleotide but everytime i use the code below to find a coding region it returns nothing. Can anyone tell me what i am doing wrong?

@nucs=("A","C","G","T");
$size=5000;
for ($i=0; $i<$size; $i++) { 
    $seqR .= $nucs[int(rand(4))];
}
print "Seq($size): $seqR\n";

if (/ATG([ACGT][ACGT][ACGT]){3,5000}(TAA|TAG|TGA)/) {
        print "This seq. might contain a coding region\n"
} else{ 
    print "This sequence most liklely does not contin a coding region\
+n"
}

[talexb]

I echo brother kscwab's comments .. this sounds an awful lot like homework.

At the very least, break down what kind of pattern you are looking for. The regexp you've listed appears to look for

• ATG;
• ACGT repeated three times, which you capture into $1, then expect to repeat from three to five thousand times; and
• One of TAA, TAG and TGA.
It's a bit hard to figure out exactly what your code is, because you haven't mastered putting the code between code tags. If you can explain in more complete sentences what you are looking for, I'm sure we can guide you into completing the Perl that you need to get the job done.

For example, the triplet you are looking for at the end .. the 'T' appears at the beginning of each pattern, so you are really looking for

• T; followed by
• AA, AG or GA.

[kschwab]

This sounds suspiciously like homework...see this faq.

Edit: Maybe break the problem down a bit, and ask a more focused question, versus just dumping the assignment almost verbatim?

[BillKSmith]

The more urgent your problem, the more important it is that you ask the right question. Properly formed questions always receive timely replies. Even the right answer to the wrong question will not do you much good. You can increase the number and quality of the replies even more if you ask the question in the vocabulary of perl rather than biology. (more of your readers will understand it)

[GotToBTru]

First thing needed: a meaningful title!

[shawnhcorey]

The triple-argument for loop can be replaced with a more-readable foreach loop.

for ( 1 .. $size ) {
    $seqR .= $nucsint(rand(4));
}

[Lotus1]

This perlmonks node has an example of using a hash for a codon table.

[charm]

Thank you all, It was homework but I figured it out

[AnomalousMonk]

Please try to master the use of  <code> ... </code> tags before your next post. Please see Markup in the Monastery and Writeup Formatting Tips.