def test_cmscan_task_multithreaded(self, tmpdir, datadir):
with tmpdir.as_cwd():
transcript = datadir('rnaseP-bsu.fa')
cm = datadir('rnaseP-eubact.c.cm')
out_single = str(tmpdir.join('single'))
out_multi = str(tmpdir.join('multi'))
for n_threads in (2, 3, 4, 5):
db_task = CMPressTask().task(cm, params=self.cmpress_cfg)
aln_tasks_single = CMScanTask().task(transcript,
out_single,
cm,
cutoff=1.0,
n_threads=1)
aln_tasks_multi = CMScanTask().task(transcript,
out_multi,
cm,
cutoff=1.0,
n_threads=n_threads)
run_tasks([db_task, aln_tasks_single], ['run'])
run_task(aln_tasks_multi)
alns_single = pd.concat(InfernalParser(out_single))
alns_multi = pd.concat(InfernalParser(out_multi))
assert all(alns_single['e_value'].sort_values() == \
alns_multi['e_value'].sort_values())
def test_hmmscan_task_multithreaded(self, tmpdir, datadir):
with tmpdir.as_cwd():
prot = datadir('20aa-alitest.fa')
hmm = datadir('20aa.hmm')
out_single = str(tmpdir.join('out-single'))
out_multi = str(tmpdir.join('out-multi'))
for n_threads in (2, 3, 4, 5):
db_task = HMMPressTask().task(hmm, params=self.hmmpress_cfg)
aln_task_single = HMMScanTask().task(prot,
out_single,
hmm,
cutoff=1.0,
n_threads=1)
aln_task_multi = HMMScanTask().task(prot,
out_multi,
hmm,
cutoff=1.0,
n_threads=n_threads)
run_tasks([db_task, aln_task_single], ['run'])
run_tasks([aln_task_multi], ['run'])
print(os.listdir(), file=sys.stderr)
print(open(out_single).read())
alns_single = pd.concat(HMMerParser(out_single))
alns_multi = pd.concat(HMMerParser(out_multi))
assert all(alns_single['domain_i_evalue'].sort_values().reset_index(drop=True) == \
alns_multi['domain_i_evalue'].sort_values().reset_index(drop=True))
def test_annotate_threaded(self, tmpdir, datadir):
'''Test the --n_threads argument.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
args = ['annotate', transcripts, '--n_threads', '2']
status, out, err = run(args)
def test_longorfs_task(self, tmpdir, datadir):
with tmpdir.as_cwd():
transcript = datadir('test-transcript.fa')
exp_orf = datadir('test-transcript-orf.pep')
task = TransDecoderLongOrfsTask().task(transcript,
params=self.longorfs_cfg)
run_tasks([task], ['run'])
output_dir = transcript + '.transdecoder_dir'
exp_pep = open(exp_orf).read()
pep_fn = os.path.join(output_dir, 'longest_orfs.pep')
assert os.path.isfile(pep_fn)
pep = open(pep_fn).read()
assert exp_pep in pep
def test_hmmscan_task(self, tmpdir, datadir):
with tmpdir.as_cwd():
prot = datadir('test-protein.fa')
hmm = datadir('test-profile.hmm')
out = str(tmpdir.join('test.out'))
db_task = HMMPressTask().task(hmm, params=self.hmmpress_cfg)
aln_task = HMMScanTask().task(prot, out, hmm,
cutoff=1.0, n_threads=1)
run_tasks([db_task, aln_task], ['run'])
print(os.listdir(), file=sys.stderr)
aln = open(out).read()
print(aln)
assert aln.count('accession') == 2
assert 'i-Evalue' in aln
def test_annotate_full(self, tmpdir, datadir):
'''Run a full annotation and verify the results.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
exp_gff3 = datadir('pom.50.fa.dammit.gff3.full')
exp_fasta = datadir('pom.50.fa.dammit.fasta.full')
args = ['annotate', transcripts, '--full']
status, out, err = run(args)
outdir = '{0}.dammit'.format(transcripts)
gff3_fn = os.path.join(outdir, 'pom.50.fa.dammit.gff3')
fasta_fn = os.path.join(outdir, 'pom.50.fa.dammit.fasta')
assert compare_gff(gff3_fn, exp_gff3)
assert open(fasta_fn).read() == open(exp_fasta).read()
def test_annotate_evalue(self, tmpdir, datadir):
'''Test the --evalue argument and verify the results.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
exp_gff3 = datadir('pom.50.fa.dammit.gff3.evalue10')
exp_fasta = datadir('pom.50.fa.dammit.fasta.evalue10')
args = ['annotate', transcripts, '--evalue', '10.0']
status, out, err = run(args)
outdir = '{0}.dammit'.format(transcripts)
gff3_fn = os.path.join(outdir, 'pom.50.fa.dammit.gff3')
fasta_fn = os.path.join(outdir, 'pom.50.fa.dammit.fasta')
assert compare_gff(gff3_fn, exp_gff3)
assert open(fasta_fn).read() == open(exp_fasta).read()
def test_annotate_multiple_user_databases(self, tmpdir, datadir):
'''Test that multiple user databases work.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
pep = datadir('pep.fa')
pep2 = datadir('odb_subset.fa')
exp_gff3 = datadir('pom.50.fa.dammit.gff3.udb')
exp_fasta = datadir('pom.50.fa.dammit.fasta.udb')
args = ['annotate', '--quick',
transcripts, '--user-databases', pep, pep2,
'--verbosity', '2']
status, out, err = run(args)
outdir = '{0}.dammit'.format(transcripts)
assert status == 0
def test_annotate_dbdir(self, tmpdir, datadir):
'''Test that --database-dir works.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
db_dir = os.environ['DAMMIT_DB_DIR']
args = ['annotate', '--quick', transcripts, '--database-dir', db_dir]
status, out, err = run(args)
def test_annotate_dbdir_fail(self, tmpdir, datadir):
'''Test annotation with a faulty database directory.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
args = ['annotate', transcripts, '--database-dir', '.']
status, out, err = run(args, fail_ok=True)
assert 'install databases to continue' in out
assert status == 2
def test_annotate_outdir(self, tmpdir, datadir):
'''Test that the --output-dir argument works.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
outdir = 'test_out'
args = ['annotate', '--quick', transcripts, '-o', outdir]
status, out, err = run(args)
fn = os.path.join(outdir, os.path.basename(transcripts))
assert os.path.isfile(fn)
def test_annotate_basic(self, tmpdir, datadir):
'''Run a basic annotation and verify the results.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.single.fa')
exp_gff3= datadir('pom.single.fa.dammit.gff3')
exp_fasta = datadir('pom.single.fa.dammit.fasta')
args = ['annotate', transcripts]
status, out, err = run(args)
outdir = '{0}.dammit'.format(transcripts)
gff3_fn = os.path.join(outdir, 'pom.single.fa.dammit.gff3')
fasta_fn = os.path.join(outdir, 'pom.single.fa.dammit.fasta')
print(os.listdir(outdir))
print(gff3_fn, fasta_fn)
assert compare_gff(gff3_fn, exp_gff3)
assert open(fasta_fn).read() == open(exp_fasta).read()
def test_cmscan_task(self, tmpdir, datadir):
with tmpdir.as_cwd():
transcript = datadir('test-transcript.fa')
cm = datadir('test-covariance-model.cm')
out = str(tmpdir.join('test.out'))
db_task = CMPressTask().task(cm, params=self.cmpress_cfg)
aln_task = CMScanTask().task(transcript,
out,
cm,
cutoff=1.0,
n_threads=1)
run_tasks([db_task, aln_task], ['run'])
print(os.listdir(), file=sys.stderr)
aln = ''.join(open(out).readlines())
print(aln, file=sys.stderr)
# TODO: better correctness check
assert aln.count('accession') == 2
assert 'E-value' in aln
def test_annotate_no_rename(self, tmpdir, datadir):
'''Test the --no-rename argument.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.50.fa')
exp_gff3 = datadir('pom.50.fa.dammit.gff3.norename')
exp_fasta = datadir('pom.50.fa.dammit.fasta.norename')
args = ['annotate', transcripts, '--no-rename']
status, out, err = run(args)
assert status == 0
outdir = '{0}.dammit'.format(transcripts)
fn = os.path.join(outdir, os.path.basename(transcripts))
assert os.path.isfile(fn)
contents = open(fn).read()
assert 'SPAC212' in contents
gff3_fn = os.path.join(outdir, 'pom.50.fa.dammit.gff3')
assert compare_gff(gff3_fn, exp_gff3)
def test_annotate_user_databases(self, tmpdir, datadir):
'''Test that a user database works.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.single.fa')
pep = datadir('pep.fa')
exp_gff3 = datadir('pom.single.fa.dammit.gff3.udb')
exp_fasta = datadir('pom.single.fa.dammit.fasta.udb')
args = ['annotate', transcripts, '--user-databases', pep,
'--verbosity', '2']
status, out, err = run(args)
outdir = '{0}.dammit'.format(transcripts)
gff3_fn = os.path.join(outdir, 'pom.single.fa.dammit.gff3')
fasta_fn = os.path.join(outdir, 'pom.single.fa.dammit.fasta')
assert status == 0
assert compare_gff(gff3_fn, exp_gff3)
assert open(fasta_fn).read() == open(exp_fasta).read()
def test_cmpress_task(self, tmpdir, datadir):
with tmpdir.as_cwd():
tf = datadir('test-covariance-model.cm')
task = CMPressTask().task(tf, params=self.cmpress_cfg)
run_tasks([task], ['run'])
status = check_status(task)
print(os.listdir(), file=sys.stderr)
for ext in self.extensions:
assert os.path.isfile(tf + ext)
assert status.status == 'up-to-date'
def test_hmmpress_task_existing(self, tmpdir, datadir):
with tmpdir.as_cwd():
tf = datadir('test-profile.hmm')
for ext in self.extensions:
touch(tf + ext)
task = HMMPressTask().task(tf)
run_tasks([task], ['run'])
print(os.listdir(), file=sys.stderr)
print(task, file=sys.stderr)
status = check_status(task)
assert status.status == 'up-to-date'
def test_predict_task(self, tmpdir, datadir):
with tmpdir.as_cwd():
transcript = datadir('pom.50.fa')
pfam = datadir('test-protein-x-pfam-a.tbl')
orf_task = TransDecoderLongOrfsTask().task(
transcript, params=self.longorfs_cfg)
pred_task = TransDecoderPredictTask().task(transcript,
pfam,
params=self.predict_cfg)
run_tasks([orf_task, pred_task], ['run'])
pprint(tmpdir.listdir())
for ext in self.extensions:
fn = transcript + '.transdecoder' + ext
assert os.path.isfile(fn)
contents = open(fn).read()
if ext == '.gff3':
assert 'mRNA' in contents
assert 'gene' in contents
assert 'CDS' in contents
assert 'three_prime_UTR' in contents
assert 'exon' in contents
def test_annotate_no_rename(self, tmpdir, datadir):
'''Test the --no_rename argument.
'''
with tmpdir.as_cwd():
transcripts = datadir('pom.single.fa')
args = ['annotate', transcripts, '--no_rename']
status, out, err = run(args)
outdir = '{0}.dammit'.format(transcripts)
fn = os.path.join(outdir, os.path.basename(transcripts))
assert os.path.isfile(fn)
contents = open(fn).read()
assert 'SPAC212' in contents
assert status == 0
