def test(self):
fasta_filename = os.path.join(self.temp_dir, 'transcripts.fasta.gz')
fastq_filenames = [
os.path.join(self.temp_dir, 'reads_1.fastq.gz'),
os.path.join(self.temp_dir, 'reads_2.fastq.gz'),
]
# download files
urllib.request.urlretrieve('https://github.com/pachterlab/kallisto/raw/master/test/transcripts.fasta.gz',
fasta_filename)
urllib.request.urlretrieve('https://github.com/pachterlab/kallisto/raw/master/test/reads_1.fastq.gz',
fastq_filenames[0])
urllib.request.urlretrieve('https://github.com/pachterlab/kallisto/raw/master/test/reads_2.fastq.gz',
fastq_filenames[1])
# run kallisto on test files
index_filename = os.path.join(self.temp_dir, 'index.idx')
rna_seq_util.Kallisto().index([fasta_filename], index_filename=index_filename)
self.assertTrue(os.path.isfile(index_filename))
output_dirname = os.path.join(self.temp_dir, 'out')
rna_seq_util.Kallisto().quant(fastq_filenames, index_filename=index_filename, output_dirname=output_dirname)
self.assertTrue(os.path.isdir(output_dirname))
self.assertTrue(os.path.isfile(os.path.join(output_dirname, 'abundance.tsv')))
self.assertTrue(os.path.isfile(os.path.join(output_dirname, 'abundance.h5')))
self.assertTrue(os.path.isfile(os.path.join(output_dirname, 'run_info.json')))
def test_error(self):
with capturer.CaptureOutput(merged=False, relay=False) as captured:
with self.assertRaises(Exception):
rna_seq_util.Kallisto()._run('index', ['__undefined__.fasta'], verbose=True)
self.assertNotEqual(captured.stdout.get_text(), '')
self.assertNotEqual(captured.stderr.get_text(), '')
with capturer.CaptureOutput(merged=False, relay=False) as captured:
with self.assertRaises(Exception):
rna_seq_util.Kallisto()._run('index', ['__undefined__.fasta'], verbose=False)
self.assertEqual(captured.stdout.get_text(), '')
self.assertEqual(captured.stderr.get_text(), '')
def process_fastq(experiment_name, sample_name, strain_name, num_fastq_files,
read_type, output_directory, temp_directory):
exp_dirname = "{}/{}".format(output_directory, experiment_name)
if not os.path.isdir(exp_dirname):
os.makedirs(exp_dirname)
sample_dirname = "{}/{}".format(exp_dirname, sample_name)
if not os.path.isdir(sample_dirname):
os.makedirs(sample_dirname)
fastq_filenames = []
for num in range(num_fastq_files):
fastq_filenames.append("{}/FASTQ_Files/{}__{}__{}.fastq.gz".format(
temp_directory, experiment_name, sample_name, num))
index_filename = '{}/kallisto_index_files/{}.idx'.format(
output_directory, strain_name)
output_dirname = '{}/output'.format(sample_dirname)
if read_type == "single":
rna_seq_util.Kallisto().quant(fastq_filenames,
index_filename=index_filename,
output_dirname=output_dirname,
single_end_reads=True,
fragment_length=180,
fragment_length_std=20)
elif read_type == "paired":
rna_seq_util.Kallisto().quant(fastq_filenames,
index_filename=index_filename,
output_dirname=output_dirname)
new_pandas = pd.read_csv('{}/output/abundance.tsv'.format(sample_dirname),
sep='\t').set_index("target_id")
total_tpm = 0
for num in new_pandas['tpm']:
total_tpm = total_tpm + num
new_column = [num / total_tpm for num in new_pandas['tpm']]
new_pandas['target_id'] = new_column
new_pandas.to_pickle("{}/{}_abundances_binary".format(
sample_dirname, sample_name))