def test_delete_clashing_sidechains(package, resource, cutoff, sequence):
"""Compare results to expected sequence."""
with resources.path(package, resource) as path:
structure = read_molecules(str(path))[0]
structure = delete_clashing_sidechains(structure, cutoff)
structure = delete_partial_residues(structure)
assert get_sequence(structure) == sequence
def test_apply_insertions(package_list, resource_list, sequence):
"""Compare results to expected sequence."""
with resources.path(package_list[0], resource_list[0]) as pdb_path:
with resources.path(package_list[1], resource_list[1]) as loop_db_path:
structure = read_molecules(str(pdb_path))[0]
structure = remove_non_protein(structure, remove_water=True)
structure_with_insertions = apply_insertions(
structure, sequence, loop_db_path)
sequence_with_insertions = get_sequence(structure_with_insertions)
assert sequence_with_insertions == sequence
def test_apply_mutations(package, resource, sequence, delete_fallback,
expectation, expected_sequence):
"""Compare results to expected sequence."""
with resources.path(package, resource) as pdb_path:
structure = read_molecules(str(pdb_path))[0]
structure = remove_non_protein(structure, remove_water=True)
with expectation:
structure_with_mutations = apply_mutations(structure, sequence,
delete_fallback)
sequence_with_mutations = get_sequence(structure_with_mutations)
assert sequence_with_mutations == expected_sequence
def _get_kinase_residue_numbers(
kinase_domain_structure: oechem.OEGraphMol,
canonical_kinase_domain_sequence: KinaseDomainAminoAcidSequence,
) -> List[int]:
"""
Get the canonical residue numbers of a kinase domain structure.
Parameters
----------
kinase_domain_structure: oechem.OEGraphMol
The kinase domain structure.
canonical_kinase_domain_sequence: KinaseDomainAminoAcidSequence
The canonical kinase domain sequence.
Returns
-------
residue_number: list of int
A list of canonical residue numbers in the same order
as the residues in the given kinase domain structure.
"""
from Bio import pairwise2
from kinoml.modeling.OEModeling import get_sequence
logging.debug("Getting sequence of given kinase domain ...")
target_sequence = get_sequence(kinase_domain_structure)
logging.debug("Aligning sequences ...")
template_sequence, target_sequence = pairwise2.align.globalxs(
canonical_kinase_domain_sequence, target_sequence, -10, 0)[0][:2]
logging.debug(f"Template sequence:\n{template_sequence}")
logging.debug(f"Target sequence:\n{target_sequence}")
logging.debug("Generating residue numbers ...")
residue_numbers = []
residue_number = canonical_kinase_domain_sequence.metadata["begin"]
for template_sequence_residue, target_sequence_residue in zip(
template_sequence, target_sequence):
if template_sequence_residue != "-":
if target_sequence_residue != "-":
residue_numbers.append(residue_number)
residue_number += 1
else:
# TODO: This situation occurs if the given protein contains
# sequence segments that are not part of the
# canonical kinase domain sequence from UniProt.
# I don't think this will ever happen in the
# current implementation.
raise NotImplementedError
return residue_numbers
def test_delete_partial_residues(package, resource, delete_backbone_C,
sequence):
"""Compare results to expected sequence."""
from openeye import oechem
with resources.path(package, resource) as path:
structure = read_molecules(str(path))[0]
if delete_backbone_C:
hier_view = oechem.OEHierView(structure)
hier_residue = hier_view.GetResidue("A", delete_backbone_C[:3],
int(delete_backbone_C[3:]))
for atom in hier_residue.GetAtoms():
atom_name = atom.GetName().strip()
if atom_name == "C":
structure.DeleteAtom(atom)
structure = delete_partial_residues(structure)
assert get_sequence(structure) == sequence
def test_delete_short_protein_segments(package, resource, sequence):
"""Compare results to expected sequence."""
with resources.path(package, resource) as path:
structure = read_molecules(str(path))[0]
structure = delete_short_protein_segments(structure)
assert get_sequence(structure) == sequence
def test_get_sequence(package, resource, sequence):
"""Compare results to expected sequence."""
with resources.path(package, resource) as path:
structure = read_molecules(str(path))[0]
assert get_sequence(structure) == sequence