##Homework 6
This assignment was adapted from Rosalind http://rosalind.info/. Rosalind is a great training tool. It describes Python code but you can do each exercise in any language and then test it using their datasets.
In DNA strings, symbols 'A' and 'T' are complements of each other, as are 'C' and 'G'.
The reverse complement of a DNA string s is the string sc formed by reversing the symbols of s, then taking the complement of each symbol (e.g., the reverse complement of "GTCA" is "TGAC").
##For this assignment:
Create and edit a file named rev_comp.pl.
####Step 1: Reverse complement subroutine
Create a subroutine called reverse_comp.
Within the subroutine:
Create a hash where the keys are the uppercase nucleotide and lowercase nucleotides and the values are the corresponding upper or lowercase complements.
The subroutine should take a string of DNA sequence as the argument $_[0]. Within the subroutine:
-- Define $contig as $_[0].
-- Create a new variable called $new_contig.
-- Reverse $contig and add its bases to an array using the split then reverse functions in Perl to create an array of reversed bases called @seq (google "perl split" or "perl reverse array" or look again at your reading in the Perl text for a refresher)
-- Loop through @seq using your hash to take the current base as the key and concatenate it's complement to $new_contig (if the base is ambiguous (not upper or lowercase A, C, G or T concatenate it as-is without complementing)
When the loop is finished return $new_contig
####Step 2: Open your input file
Open the fasta file passed as the first argument to your perl script for reading. Name the file $file when you read $ARGV[0];
####Step 3: Create your output file
Parse the filename of the fasta file using the fileparse function from the Perl module File::Basename. fileparse is similar to basename in Bash. It takes a filename and an extension as arguments but it outputs an array with the basename, directory, and extension as array elements.
The following two lines can allow you to create variables from the parts of an input filename. This assumes you named $ARGV[0] $file. Otherwise change $file to the correct variable for your input filename. You can use these to create output filenames based on the input filename:
use File::Basename; # enable maipulating of the full path
my ($basename, $directories, $suffix) = fileparse($file,'\..*'); # break appart filenames
Now you can create a new output filename from the original filename:
my $outfile = "${directories}${basename}_revcomp${suffix}";
To redirect this output to your working directory output you would write:
my $outfile = "output/${basename}_revcomp${suffix}";
Create an output filename with that begins with the directory output/ and has _revcomp added to the basename using fileparse.
####Step 4: Open your output file
Open your output file with write permissions ( using > instead of <)
####Step 5: Define a new variable
Declare the variable $current_contig but do not define it
####Step 6: Parse your input file
While there are lines to read in your input fasta file:
-- If a line is a header and $current_contig is defined (remember for the first line $current_contig is undefined for the first header) then run &reverse_comp with the argument $current_contig. Print the return from &reverse_comp and a new line character to the output file. Then print $_. Whether or not the header is the first print the header $_ to the output file.
-- If the line is not a header (i.e. it is sequence data) remove newline characters and concatinate the line to $current_contig
-- If the line is the end of the file run &reverse_comp with the argument $current_contig. Print the return from &reverse_comp and a new line character to the output file.
When you are done your script should run using the command:
scripts/rev_comp.pl /homes/sheltonj/solutions/mini.fasta
Upload your finished script to KSOL.
##Contents of /homes/sheltonj/solutions/mini.fasta
>seq_1
TNNnActg
>seq_2
aatNNnAc
CCCGgggt
act
##Correct contents of your output file
>seq_1
cagTnNNA
>seq_2
agtacccCGGGgTnNNatt
####An extra problem
If you solved this problem and want more than you should try to wrap the sequence in your output fasta file for extra credit.