The command to run the regression tests is

python regression_test.py <test case directory> <location of tophat executable> <location of samtools executable>

e.g.,

python regression_test.py test_cases ~/SVN/trunk/bin/tophat `which samtools`

"regression_test.py" will look in the <test case directory> for any test cases to execute. A test case is any sub-directory that begins with the word "test". Within that subdirectory, the user should place:
	1. A file "command.txt" specifying a particular TopHat command
	2. Any input files necessary to run the TopHat command (including a FASTA version of the genome)
	3. Any output files generated from the TopHat command

The output files should be exactly as though "command.txt" was executed within the directory. For example,

test_sample/
	command.txt
	input/
		test_reads.fq
		genome/
			genome.1.ebwt
			...
			genome.fa
	tophat_out/
		accepted_hits.bam
		insertions.bed
		deletions.bed
		junctions.bed

and the contents of "command.txt" would be:

tophat input/genome/genome input/test_reads.fq

Please note that the input genome directory must also contain a FASTA version of the genome. This is necessary to perform certain validations of the BAM file generated by TopHat. 

The script will re-run the command and perform several validations:
	1. Verify that TopHat can be executed successfully executed on the input data
	2. Compare the output files against the gold-standard provided by the test case
	3. Perform additional sanity checks on the output
		a. Can the BAM file be converted successfully to SAM?
		b. Does "samtools calmd" perform any tag modifications? 

