def test_generate_model(self):
""" Test that Monte Carlo modeling runs """
try:
import networkx
import subprocess
except ImportError as detail:
self.skipTest(str(detail))
domino_model = self.import_python_application('emagefit')
fn_config = self.get_input_file_name("config.py")
exp = utility.get_experiment_params(fn_config)
fn_database = "monte_carlo_output_database.db"
domino_model.generate_monte_carlo_model(exp,
fn_database,
seed=-1,
write_solution=True)
# test that the database and pdb files are generated and that they
# are not empty
self.assertTrue(os.path.exists(fn_database))
self.assertGreater(os.path.getsize(fn_database), 0)
fn_pdb = fn_database + ".pdb"
self.assertTrue(os.path.exists(fn_pdb))
self.assertGreater(os.path.getsize(fn_pdb), 0)
# check that there is one solution in the database
db = solutions_io.ResultsDB()
db.connect(fn_database)
data = db.get_solutions()
self.assertEqual(len(data), 1)
os.remove(fn_database)
os.remove(fn_pdb)
def test_generate_model(self):
""" Test that Monte Carlo modeling runs """
try:
import networkx
import subprocess
except ImportError as detail:
self.skipTest(str(detail))
domino_model = self.import_python_application('emagefit')
fn_config = self.get_input_file_name("config.py")
exp = utility.get_experiment_params(fn_config)
fn_database = "monte_carlo_output_database.db"
domino_model.generate_monte_carlo_model(exp, fn_database, seed=-1,
write_solution=True)
# test that the database and pdb files are generated and that they
# are not empty
self.assertTrue(os.path.exists(fn_database))
self.assertGreater(os.path.getsize(fn_database), 0)
fn_pdb = fn_database + ".pdb"
self.assertTrue(os.path.exists(fn_pdb))
self.assertGreater(os.path.getsize(fn_pdb), 0)
# check that there is one solution in the database
db = solutions_io.ResultsDB()
db.connect(fn_database)
data = db.get_solutions()
self.assertEqual(len(data), 1)
os.remove(fn_database)
os.remove(fn_pdb)
def test_generating_clusters(self):
"""
Test that the solutions generated by the clustering a the same
as those obtainedduring the benchmark for the paper
"""
emagefit_cluster = \
self.import_python_application('emagefit_cluster')
fn_config = self.get_input_file_name("config.py")
fn_database = self.get_input_file_name("domino_solutions.db")
fn_db_clusters = "clusters.db"
# modify the names of the PDB files to include the proper name for
# testing
exp = utility.get_experiment_params(fn_config)
for i in range(len(exp.fn_pdbs)):
exp.fn_pdbs[i] = self.get_input_file_name(exp.fn_pdbs[i])
n_solutions = 30
orderby = "em2d"
max_rmsd = 10
tc = emagefit_cluster.AlignmentClustering(exp)
tc.cluster(fn_database, n_solutions, orderby, max_rmsd)
tc.store_clusters(fn_db_clusters, "clusters")
# retrieve the largest cluster
solutions_stored = '9|10|11|12|13|14|15|16|17|18|19|20|21|22|23|24|25|26|27|28|29'
solutions_stored = map(int, solutions_stored.split("|"))
db_clusters = solutions_io.ResultsDB()
db_clusters.connect(fn_db_clusters)
cl_record = db_clusters.get_nth_largest_cluster(1)
elements = map(int, cl_record.elements.split("|"))
for i, j in zip(solutions_stored, elements):
self.assertEqual(i, j)
db_clusters.close()
os.remove(fn_db_clusters)
def test_generate_model(self):
""" Test that the DOMINO modeling runs """
try:
import networkx
import subprocess
except ImportError as detail:
self.skipTest(str(detail))
domino_model = self.import_python_application('emagefit')
IMP.set_log_level(IMP.SILENT)
fn = self.get_input_file_name("config.py")
exp = utility.get_experiment_params(fn)
fn_output_db = "domino_solutions_temp.db"
domino_model.generate_domino_model(exp, fn_output_db)
# assert that a database of results is created
self.assertTrue(os.path.exists(fn_output_db))
self.assertGreater(os.path.getsize(fn_output_db), 0)
# check that there are solutions in the database
db = solutions_io.ResultsDB()
db.connect(fn_output_db)
data = db.get_solutions()
self.assertGreater(len(data), 0)
columns = db.get_table_column_names("results")
self.assertTrue("em2d" in columns)
os.remove(fn_output_db)