def load_structure(fpath, chain=None):
"""
Args:
fpath: filepath to either pdb or cif file
chain: the chain id
Returns:
biotite.structure.AtomArray
"""
if fpath.endswith('cif'):
with open(fpath) as fin:
pdbxf = pdbx.PDBxFile.read(fin)
structure = pdbx.get_structure(pdbxf, model=1)
elif fpath.endswith('pdb'):
with open(fpath) as fin:
pdbf = pdb.PDBFile.read(fin)
structure = pdb.get_structure(pdbf, model=1)
issolvent = filter_solvent(structure)
structure = structure[~issolvent]
chains = get_chains(structure)
print(f'Found {len(chains)} chains:', chains, '\n')
if len(chains) == 0:
raise ValueError('No chains found in the input file.')
if chain is None:
chain = chains[0]
if chain not in chains:
raise ValueError(f'Chain {chain} not found in input file')
structure = structure[structure.chain_id == chain]
print(f'Loaded chain {chain}\n')
return structure
def test_remove_pbc_unsegmented():
"""
`remove_pbc()` should not alter unsegmented structures,
when the structure is entirely in the box.
Exclude the solvent, due to high distances between each atom.
"""
ref_array = load_structure(join(data_dir("structure"), "3o5r.mmtf"))
# Center structure in box
centroid = struc.centroid(ref_array)
box_center = np.diag(ref_array.box) / 2
ref_array = struc.translate(ref_array, box_center - centroid)
# Remove solvent
ref_array = ref_array[~struc.filter_solvent(ref_array)]
array = struc.remove_pbc(ref_array)
assert ref_array.equal_annotation_categories(array)
assert np.allclose(ref_array.coord, array.coord)
import biotite.database.rcsb as rcsb
import numpy as np
# The output file names
# Modify these values for actual file output
ku_dna_file = biotite.temp_file("ku_dna.cif")
ku_file = biotite.temp_file("ku.cif")
# Download and parse structure files
file = rcsb.fetch("1JEY", "mmtf", biotite.temp_dir())
ku_dna = strucio.load_structure(file)
file = rcsb.fetch("1JEQ", "mmtf", biotite.temp_dir())
ku = strucio.load_structure(file)
# Remove DNA and water
ku_dna = ku_dna[(ku_dna.chain_id == "A") | (ku_dna.chain_id == "B")]
ku_dna = ku_dna[~struc.filter_solvent(ku_dna)]
ku = ku[~struc.filter_solvent(ku)]
# The structures have a differing amount of atoms missing
# at the the start and end of the structure
# -> Find common structure
ku_dna_common = ku_dna[struc.filter_intersection(ku_dna, ku)]
ku_common = ku[struc.filter_intersection(ku, ku_dna)]
# Superimpose
ku_superimposed, transformation = struc.superimpose(
ku_dna_common, ku_common, (ku_common.atom_name == "CA"))
# We do not want the cropped structures
# -> apply superimposition on structures before intersection filtering
ku_superimposed = struc.superimpose_apply(ku, transformation)
# Write PDBx files as input for PyMOL
cif_file = pdbx.PDBxFile()
pdbx.set_structure(cif_file, ku_dna, data_block="ku_dna")
def test_solvent_filter(sample_array):
assert len(sample_array[struc.filter_solvent(sample_array)]) == 287
import biotite.structure.io.pdb as pdb import biotite.structure as struc pdb_file = pdb.PDBFile() pdb_file.read(snakemake.input[0]) # Only use one model structure = pdb_file.get_structure(model=1) # Remove water structure = structure[~struc.filter_solvent(structure)] # Remove hydrogens structure = structure[structure.element != "H"] pdb_file.set_structure(structure) pdb_file.write(snakemake.output[0])
traj_file_path = "../../download/waterbox_md.xtc" # Load the trajectory traj = strucio.load_structure(traj_file_path, template=templ_file_path) # Sanitize the PDB file produced by Gromacs: # Use capital letters for atom elements... traj.element = np.array([element.upper() for element in traj.element]) # ...and set 'hetero' to true for all atoms, # as the file does not contain any regular chains. traj.hetero[:] = True # Create boolean masks for all sodium or chloride ions, respectively na = traj.coord[:, traj.element == "NA"] cl = traj.coord[:, traj.element == "CL"] # Create a boolean mask for all watewr molecules solvent = traj[:, struc.filter_solvent(traj)] # Calculate the RDF of water molecules # centered on sodium or chloride ions, respectively N_BINS = 200 bins, rdf_na = struc.rdf(center=na, atoms=solvent, periodic=True, bins=N_BINS) bins, rdf_cl = struc.rdf(center=cl, atoms=solvent, periodic=True, bins=N_BINS) # Find peaks # This requires a bit trial and error on the parameters # The 'x' in '[x * N_BINS/10]' is the expected peak width in Å, # that is transformed into a peak width in amount of values peak_indices_na = signal.find_peaks_cwt(rdf_na, widths=[0.2 * N_BINS / 10]) peak_indices_cl = signal.find_peaks_cwt(rdf_cl, widths=[0.3 * N_BINS / 10]) peak_indices_na, peak_indices_cl = peak_indices_na[:3], peak_indices_cl[:3] # Create plots