def test_data_source_creation(self):
# Test creation
source_1 = DataSource("lol.pdb")
self.assertDictEqual({"source": "lol.pdb"}, source_1.source)
source_2 = DataSource({"source": "lol.pdb", "another_kw": 3})
self.assertDictEqual({
"source": "lol.pdb",
"another_kw": 3
}, source_2.source)
def test_data_source_comparison(self):
m_source = DataSource("mini")
n_source = DataSource("normal")
another_n_source = DataSource("normal")
b_source = DataSource("zobig")
# Test lexicographical comparison
self.assertTrue(m_source < n_source)
self.assertTrue(n_source == another_n_source)
self.assertTrue(b_source > n_source)
def test_load(self):
loader = FeatureArrayDataLoader(None)
source = DataSource(
os.path.join(TestFeatArrDataLoader.data_path,
"numerical_features_1.txt"))
self.compare_loaded_results(data.expected_1,
loader.load_data_from_source(source))
source.add_info("labels", ["uno", "dos", "tres", "cuatro", "cinco"])
self.compare_loaded_results(data.expected_2,
loader.load_data_from_source(source))
def setup_loader(self):
"""
Builds a loader with 2 preloaded pdbs from the data folder.
"""
loader = ProteinEnsembleDataLoader(ProtocolParameters({"matrix":{}}))
source1 = DataSource(os.path.join(test_data.__path__[0], "pdb1.pdb"))
loader.load(source1)
source2 = DataSource({
"source":os.path.join(test_data.__path__[0], "pdb2.dcd"),
"atoms_source": os.path.join(test_data.__path__[0], "pdb2.pdb")
})
loader.load(source2)
return loader
def test_load(self):
"""
If we already loaded 2 pdbs of 2 conformations, the elements corresponfding to
the next loaded pdb of 2 confs, 2 atoms will be [4,5] (elements start in 0).
It indirectly tests the dcd loading (which would be a responsabilit of the
data test indeed).
"""
loader = self.setup_loader()
source3 = DataSource(os.path.join(test_data.__path__[0], "pdb3.pdb"))
final_range = loader.load(source3)
self.assertEqual([4,5], list(final_range))
self.assertEqual(source3.source['number_of_conformations'], 2)
self.assertEqual(source3.source['number_of_atoms'], 2)
def test_load_with_base_selection(self):
"""
Fine grain case of test_load using base_selection in the file descriptor.
"""
loader = self.setup_loader()
source3 = DataSource({
"source": os.path.join(test_data.__path__[0], "pdb3.pdb"),
"base_selection": "resnum 3"
})
final_range = loader.load(source3)
self.assertSequenceEqual([4,5], list(final_range))
self.assertEqual(source3.source['number_of_conformations'], 2)
self.assertEqual(source3.source['number_of_atoms'], 1)
self.assertEqual(loader.loaded_data[-1].getResnames(), ['ILE'])
# Closing this loader must raise a prody exception (different number of atoms
# for some conformations)
self.assertRaises(ValueError, loader.close)
def test_close_with_model_remark(self):
"""
Tests that model number and remark info was acquired
"""
loader = self.setup_loader()
source3 = DataSource(os.path.join(test_data.__path__[0], "pdb3.pdb"))
loader.load(source3)
loader.close()
expected_models = [1,2,1,2,8,9] #[1,2,5,6,8,9] -> this if whe have a pdb
expected_remarks = [
["REMARK 400 HELLO\n"],
["REMARK 400 WORLD\n"],
[],
[],
[],
["REMARKS NINETH MODEL\n"]
]
self.assertSequenceEqual(expected_remarks, loader.model_remarks)
self.assertSequenceEqual(expected_models, loader.model_numbers)
def test_load_one_row(self):
loader = FeatureArrayDataLoader(None)
source = DataSource(os.path.join(TestFeatArrDataLoader.data_path, "numerical_features_1.txt"))
source.add_info("labels", ["uno", "dos", "tres", "cuatro", "cinco"])
source.add_info("load_only", [2, 3])
self.compare_loaded_results(data.expected_3, loader.load_data_from_source(source))
def get_elements_with_source(self, source_str):
"""
"""
return [i for i in self.elements[DataSource(source_str)]]
def __init__(self, source_list):
self.source_list = [
DataSource(source)
for source in SourceGenerator.inflate_source_list(source_list)
]