def test_get_assembly(single_model):
"""
Test whether the :func:`get_assembly()` function produces the same
number of peptide chains as the
``_pdbx_struct_assembly.oligomeric_count`` field indicates.
"""
model = 1 if single_model else None
path = join(data_dir, "1f2n.cif")
pdbx_file = pdbx.PDBxFile()
pdbx_file.read(path)
assembly_category = pdbx_file.get_category("pdbx_struct_assembly",
expect_looped=True)
# Test each available assembly
for id, ref_oligomer_count in zip(assembly_category["id"],
assembly_category["oligomeric_count"]):
assembly = pdbx.get_assembly(pdbx_file, assembly_id=id, model=model)
protein_assembly = assembly[..., struc.filter_amino_acids(assembly)]
test_oligomer_count = struc.get_chain_count(protein_assembly)
if single_model:
assert isinstance(assembly, struc.AtomArray)
else:
assert isinstance(assembly, struc.AtomArrayStack)
assert test_oligomer_count == int(ref_oligomer_count)
def test_get_assembly(model):
"""
Test whether the :func:`get_assembly()` function produces the same
number of peptide chains as the
``_pdbx_struct_assembly.oligomeric_count`` field indicates.
Furthermore, check if the number of atoms in the entire assembly
is a multiple of the numbers of atoms in a monomer.
"""
path = join(data_dir("structure"), "1f2n.cif")
pdbx_file = pdbx.PDBxFile.read(path)
assembly_category = pdbx_file.get_category("pdbx_struct_assembly",
expect_looped=True)
# Test each available assembly
for id, ref_oligomer_count in zip(assembly_category["id"],
assembly_category["oligomeric_count"]):
print("Assembly ID:", id)
try:
assembly = pdbx.get_assembly(pdbx_file,
assembly_id=id,
model=model)
except biotite.InvalidFileError:
if model is None:
# The file cannot be parsed into an AtomArrayStack,
# as the models contain different numbers of atoms
# -> skip this test case
return
else:
raise
protein_assembly = assembly[..., struc.filter_amino_acids(assembly)]
test_oligomer_count = struc.get_chain_count(protein_assembly)
if model is None:
assert isinstance(assembly, struc.AtomArrayStack)
else:
assert isinstance(assembly, struc.AtomArray)
assert test_oligomer_count == int(ref_oligomer_count)
# The atom count of the entire assembly should be a multiple
# a monomer,
monomer_atom_count = pdbx.get_structure(pdbx_file).array_length()
assert assembly.array_length() % monomer_atom_count == 0
pdbx_file = pdbx.PDBxFile.read(rcsb.fetch("1M4X", "mmcif"))
assemblies = pdbx.list_assemblies(pdbx_file)
print("ID name")
print()
for assembly_id, name in assemblies.items():
print(f"{assembly_id:2} {name}")
########################################################################
# ``'complete icosahedral assembly'`` sounds good.
# In fact, often the first assembly is the complete one.
# Hence, the :func:`get_assembly()` function builds the first assembly
# by default.
# Since we know the ID we want (``'1'``), we will provide it to this
# function anyway.
# It returns the chosen assembly as :class:`AtomArray`.
# Note that the assembly ID is a string, not an integer.
biological_unit = pdbx.get_assembly(pdbx_file, assembly_id="1", model=1)
print("Number of protein chains:", struc.get_chain_count(biological_unit))
########################################################################
# Now we could do some analysis on the biological unit.
# But for this example we will simply save the entire assembly as *PDB*
# file for later visualization.
# For brevity, save only CA atoms to file for visualization with PyMOL
#biological_unit = biological_unit[biological_unit.atom_name == "CA"]
#strucio.save_structure("biological_assembly.pdb", biological_unit)
# biotite_static_image = biological_assembly.png
def test_get_chain_count(array):
assert struc.get_chain_count(array) == 6