def test_to_sdf(datadir, tmp_path):
data_path = datadir / "TUBB3-observations.sdf.gz"
df = dm.read_sdf(data_path, as_df=True)
sdf_path = tmp_path / "mols.sdf"
dm.to_sdf(df, sdf_path, smiles_column="smiles")
new_df = dm.read_sdf(sdf_path, as_df=True)
assert df.equals(new_df)
def test_to_sdf_mols(datadir, tmp_path):
data_path = datadir / "TUBB3-observations.sdf.gz"
mols = dm.read_sdf(data_path, as_df=False)
sdf_path = tmp_path / "mols.sdf"
dm.to_sdf(mols, sdf_path)
new_mols = dm.read_sdf(sdf_path, as_df=False)
assert [dm.to_smiles(mol) for mol in mols] == [dm.to_smiles(mol) for mol in new_mols]
def test_read_sdf_gz(datadir):
data_path = datadir / "TUBB3-observations.sdf.gz"
mols = dm.read_sdf(data_path)
assert len(mols) == 10
for mol in mols:
assert isinstance(mol, Chem.rdchem.Mol)
def test_sdf_props_and_conformer_preserved(tmp_path):
sdf_path = tmp_path / "test.sdf"
# Generate an SDF file
props = dict(test_int=588, test_str="hello")
smiles = "CC1(C2C(C3C(C(=O)C(=C(C3(C(=O)C2=C(C4=C1C=CC=C4O)O)O)O)C(=O)N)N(C)C)O)O"
mol = dm.to_mol(smiles)
mol = dm.set_mol_props(mol, props)
mol = dm.conformers.generate(mol, n_confs=1)
pos = mol.GetConformer().GetPositions()
dm.to_sdf(mol, sdf_path)
# Read sdf file
mols = dm.read_sdf(sdf_path)
mol = mols[0]
# Check properties
assert mol.GetPropsAsDict() == props
# Check conformer
conf = mol.GetConformer()
assert mol.GetNumConformers() == 1
assert conf.Is3D()
np.testing.assert_almost_equal(conf.GetPositions(), pos, decimal=4)
def test_to_sdf_single_mol(tmp_path):
sdf_path = tmp_path / "test.sdf"
smiles = "CC1(C2C(C3C(C(=O)C(=C(C3(C(=O)C2=C(C4=C1C=CC=C4O)O)O)O)C(=O)N)N(C)C)O)O"
mol = dm.to_mol(smiles)
dm.to_sdf(mol, sdf_path)
mols = dm.read_sdf(sdf_path)
assert dm.to_smiles(mol) == dm.to_smiles(mols[0])
def test_read_sdf_as_df(datadir):
data_path = datadir / "TUBB3-observations.sdf"
df = dm.read_sdf(data_path, as_df=True)
assert df.shape == (10, 12)
assert set(df.columns) == {
"smiles",
"zinc_id",
"ortholog_name",
"gene_name",
"affinity",
"chembldocid",
"title",
"reference.pubmed_id",
"reference.doi",
"reference.chembl_id",
"reference.journal",
"reference.year",
}
def test_to_df(datadir):
data_path = datadir / "TUBB3-observations.sdf"
mols = dm.read_sdf(data_path)
df = dm.to_df(mols)
assert df.shape == (10, 12)
assert list(df.columns) == [
"smiles",
"zinc_id",
"ortholog_name",
"gene_name",
"affinity",
"chembldocid",
"title",
"reference.pubmed_id",
"reference.doi",
"reference.chembl_id",
"reference.journal",
"reference.year",
]
def test_from_df(datadir):
data_path = datadir / "TUBB3-observations.sdf"
df = dm.read_sdf(data_path, as_df=True)
mols = dm.from_df(df)
assert len(mols) == 10
assert isinstance(mols[0], Chem.rdchem.Mol)
assert set(mols[0].GetPropsAsDict().keys()) == {
"zinc_id",
"ortholog_name",
"gene_name",
"affinity",
"chembldocid",
"title",
"reference.pubmed_id",
"reference.doi",
"reference.chembl_id",
"reference.journal",
"reference.year",
}
assert dm.from_df(pd.DataFrame()) == []
def test_read_sdf_as_df_mol_col(datadir):
data_path = datadir / "TUBB3-observations.sdf"
df = dm.read_sdf(data_path, as_df=True, mol_column="mol")
assert "mol" in df.columns
assert isinstance(df.iloc[0]["mol"], Chem.rdchem.Mol)
