#!/usr/bin/perl

# Very simple format converter: Convert blast hits to a BED file that can be loaded as a custom track
# in the UCSC browser.
# Jason de Koning, 2010.  http://jasondk.org/Teaching.html

# Filename for your blast output (in tabular format; specify with '-m 8')
$file = shift;

# Need your template information for here:
# (If you obtained your data following the example in class,
#  this information is on the first line of your template sequence file).
$templateChromosome = "chr1";
$startPositionOnChr = 131970000;
$endPositionOnChr   = 131995000;


# Open input and output files
open(IN, $file);

# The output file has '.bed' appended to the end
open(OUT, ">" . $file . ".bed");

# This information will tell the genome browser to center its display on your template region
print OUT "browser position " . $templateChromosome . ":" . $startPositionOnChr . "-" . $endPositionOnChr . "\n";
print OUT "browser hide all\n";
print OUT "track name=hits description=\"Hits from de novo assembly experiment\" visibility=2\n";

# Now lets create a track for each blast hit 
while ($line = <IN>) {
	chomp($line);
	@tokens = split('\t', $line);

	# BLAST will give both forward- and reverse-strand hits - we will annotate these with different colors
	# (the last field in the line specifies color.  See: http://genome.ucsc.edu/FAQ/FAQformat.html#format1 )
 
	# if the end coordinate of a hit is less than the start coord, it is a reverse strand hit
	if ($tokens[8] < $tokens[9]) { 
		# Forward strand hit	
		print OUT $templateChromosome . "\t" . ( $tokens[8] + $startPositionOnChr ) . "\t" . ( $tokens[9] + $startPositionOnChr ) . "\t$tokens[0]\t1000\n";
	} else {
		# Reverse strand hit - we need to give the coords to UCSC in the opposite orientation	
		print OUT $templateChromosome . "\t" . ( $tokens[9] + $startPositionOnChr ) . "\t" . ( $tokens[8] + $startPositionOnChr ) . "\t$tokens[0]\t500\n";
	}
}

close OUT;
close IN;