def test_pdb_element_names_from_amber(self):
st = gemmi.read_pdb_string(AMBER_FRAGMENT)
residue = st[0][''][0]
self.assertEqual(residue.sole_atom('CB').element, gemmi.Element('C'))
self.assertEqual(residue.sole_atom('HB').element, gemmi.Element('H'))
self.assertEqual(residue.sole_atom('CG1').element, gemmi.Element('C'))
self.assertEqual(residue.sole_atom('HG11').element, gemmi.Element('H'))
chain = gemmi.read_pdb_string(FRAGMENT_WITH_HG)[0]['P']
self.assertEqual(chain[0].sole_atom('HG').element, gemmi.Element('Hg'))
self.assertEqual(chain[1].sole_atom('HG1').element, gemmi.Element('Hg'))
def test_pdb_element_names_from_amber(self):
st = gemmi.read_pdb_string(AMBER_FRAGMENT)
residue = st[0][''][0]
self.assertEqual(residue.sole_atom('CB').element, gemmi.Element('C'))
self.assertEqual(residue.sole_atom('HB').element, gemmi.Element('H'))
self.assertEqual(residue.sole_atom('CG1').element, gemmi.Element('C'))
self.assertEqual(residue.sole_atom('HG11').element, gemmi.Element('H'))
lines = AMBER_FRAGMENT.splitlines()
for n, atom in enumerate(residue):
self.assertEqual(atom.padded_name(), lines[n][12:16].rstrip())
chain = gemmi.read_pdb_string(FRAGMENT_WITH_HG)[0]['P']
self.assertEqual(chain[0].sole_atom('HG').element, gemmi.Element('Hg'))
self.assertEqual(chain[1].sole_atom('HG1').element,
gemmi.Element('Hg'))
def test_short_ssbond(self):
st = gemmi.read_pdb_string(SHORT_SSBOND)
out = st.make_pdb_headers()
self.assertEqual(
out.splitlines()[0],
SHORT_SSBOND.splitlines()[0] +
" 1555 1555 2.06 ")
def test_add_remove(self):
st = gemmi.read_pdb_string(SSBOND_FRAGMENT)
st.add_model(st[0])
st.renumber_models()
res = st[0].sole_residue('A', gemmi.SeqId('310'))
self.assertEqual(len(res), 1)
res.remove_atom('SG', ' ')
self.assertEqual(len(res), 0)
res = st[1].sole_residue('A', gemmi.SeqId('310'))
self.assertEqual(len(res), 1)
self.assertEqual(len(st[0]['A']), 7)
del st[0]['A'][3]
self.assertEqual(len(st[0]['A']), 6)
self.assertEqual(len(st), 2)
self.assertEqual(st[0].name, '1')
del st['1']
self.assertEqual(len(st), 1)
self.assertEqual(st[0].name, '2')
st.renumber_models()
self.assertEqual(st[0].name, '1')
st.add_model(st[0])
st.add_model(st[0])
st.renumber_models()
self.assertEqual(st[0].name, '1')
self.assertEqual(st[-1].name, '3')
del st[:-1]
self.assertEqual(len(st), 1)
self.assertEqual(st[0].name, '3')
del st[0]
self.assertEqual(len(st), 0)
def get_reference_models(project_code):
print("Project code is: {}".format(project_code))
summary = xcextracter(project_code)
print(summary)
print("Number of records is: {}".format(len(summary)))
pdb_grabber = GetPdbData()
reference_models = {}
for index, row in summary.iterrows():
protein_code = row["protein_code"]
# print(protein_code)
dtag = protein_code_to_dtag(protein_code)
# print(dtag)
pdb_block = pdb_grabber.get_bound_pdb_file(protein_code)
try:
# print(pdb_block)
structure = gemmi.read_pdb_string(pdb_block)
model = structure
reference_models[dtag] = model
except Exception as e:
print(e)
reference_models[dtag] = None
return reference_models
def test_pdb_misaligned_element(self):
pdb_line = "ATOM 7 S SUB A 7 34.489 -14.293 34.343" \
" 0.29 43.77 S"
for line in [pdb_line, pdb_line + '\n', pdb_line + '\r\n']:
st = gemmi.read_pdb_string(line)
atom = st[0].sole_residue('A', gemmi.SeqId('7')).sole_atom('S')
self.assertEqual(atom.element.name, 'S')
def test_pdb_fragment(self):
pdb_line = "HETATM 4154 MG MG A 341 1.384 19.340 11.968" \
" 1.00 67.64 MG"
for line in [pdb_line, pdb_line.strip(' MG'), pdb_line[:-2] + ' ']:
st = gemmi.read_pdb_string(line)
mg_atom = st[0].sole_residue('A', 341, ' ')['MG']
self.assertEqual(mg_atom.element.name, 'Mg')
self.assertAlmostEqual(mg_atom.b_iso, 67.64, delta=1e-6)
def test_pdb_element_names_from_trjconv(self):
st = gemmi.read_pdb_string(TRJCONV_FRAGMENT)
residue = st[0][''][0]
expected = ['C', 'H', 'H', 'O', 'P', 'O', 'O', 'O']
lines = TRJCONV_FRAGMENT.splitlines()
for n, atom in enumerate(residue):
self.assertEqual(atom.element.name, expected[n])
self.assertEqual(atom.padded_name(), lines[n][12:16].rstrip())
def test_1gtv(self):
st = gemmi.read_pdb_string(FRAGMENT_1GTV)
a1 = st[0].sole_residue('A', 85, ' ')[0]
subcells = gemmi.SubCells(st[0], st.cell, 5)
marks = subcells.find_atoms(a1.pos, a1.altloc, 3)
self.assertEqual(len(marks), 2)
for mark in marks:
d = subcells.dist(a1.pos, mark.pos())
self.assertAlmostEqual(d, 0, delta=5e-6)
def test_blank_chain(self):
st = gemmi.read_pdb_string(BLANK_CHAIN_FRAGMENT)
out_name = get_path_for_tempfile()
st.write_minimal_pdb(out_name)
out = read_lines_and_remove(out_name)
# CRYST1 differs (50.000 not 50.00 and added P1).
# ATOM lines have added element names.
trimmed_out = [line[:66] for line in out[1:]]
self.assertEqual(trimmed_out, BLANK_CHAIN_FRAGMENT.splitlines()[1:])
def test_remove_atom(self):
st = gemmi.read_pdb_string(SSBOND_FRAGMENT)
res = st[0].sole_residue('A', 310, ' ')
self.assertEqual(len(res), 1)
del res['SG']
self.assertEqual(len(res), 0)
self.assertEqual(len(st), 1)
self.assertEqual(st[0].name, '1')
del st['1']
self.assertEqual(len(st), 0)
def test_pdb_element_names(self):
pdb_line = "HETATM 4154 MG MG A 341 1.384 19.340 11.968" \
" 1.00 67.64 MG"
for line in [pdb_line, pdb_line.strip(' MG'), pdb_line[:-2] + ' ']:
st = gemmi.read_pdb_string(line)
residue = st[0].sole_residue('A', gemmi.SeqId(341, ' '))
mg_atom = residue.sole_atom('MG')
self.assertEqual(mg_atom.element.name, 'Mg')
self.assertAlmostEqual(mg_atom.b_iso, 67.64, delta=1e-6)
mg_atom.element = gemmi.Element('Cu')
self.assertEqual(mg_atom.element.name, 'Cu')
def test_5a11(self):
st = gemmi.read_pdb_string(FRAGMENT_5A11)
a1 = st[0].sole_residue('A', 37, ' ')[0]
sc = gemmi.SubCells(st[0], st.cell, 5)
marks = sc.find_atoms(a1.pos, a1.altloc, 3)
m1, m2 = sorted(marks, key=lambda m: sc.dist(a1.pos, m.pos()))
self.assertAlmostEqual(sc.dist(a1.pos, m1.pos()), 0, delta=5e-6)
self.assertAlmostEqual(sc.dist(a1.pos, m2.pos()), 0.13, delta=5e-3)
cra2 = m2.to_cra(st[0])
self.assertEqual(cra2.chain.name, 'B')
self.assertEqual(str(cra2.residue.seqid), '37')
self.assertEqual(cra2.atom.name, 'SG')
def test_4hhh_frag(self):
path = full_path('4hhh_frag.pdb')
with open(path) as f:
frag = f.read()
st = gemmi.read_pdb_string(frag)
in_headers = frag.splitlines()
out_headers = st.make_pdb_headers().splitlines()
self.assertEqual(in_headers[0], out_headers[0])
# the difference 4555 vs 2555 doesn't matter for us
self.assertEqual(in_headers[1],
out_headers[1].replace(' 4555 ', ' 2555 '))
self.assertEqual(in_headers[2], out_headers[2])
def test_1gtv(self):
st = gemmi.read_pdb_string(FRAGMENT_1GTV)
a1 = st[0].sole_residue('A', gemmi.SeqId(85, ' '))[0]
ns = gemmi.NeighborSearch(st[0], st.cell, 5)
ns.populate()
marks = ns.find_atoms(a1.pos, a1.altloc, 3)
self.assertEqual(len(marks), 2)
for mark in marks:
d = ns.dist(a1.pos, mark.pos())
self.assertAlmostEqual(d, 0, delta=5e-6)
marks2 = ns.find_neighbors(a1, 0.1, 3)
self.assertEqual(len(marks2), 0)
def parsestr(instream):
"""Parse structure file from a string.
:param instream: Imported-to-string structure file.
:type instream: str
:return: Parsed structure
:rtype: :obj:`gemmi.Structure`
"""
strout = gemmi.read_pdb_string(instream)
return strout
def test_different_altloc_order(self):
st = gemmi.read_pdb_string(UNORDERED_ALTLOC_FRAGMENT)
chain = st[0]['A']
cb = chain['9']['SER']['CB']
cb_numbers = [atom.serial for atom in cb]
self.assertEqual(cb_numbers, [59, 65])
self.assertEqual(cb[0].serial, 59)
self.assertEqual(cb[1].serial, 65)
self.assertEqual(cb[-1].serial, 65)
self.assertEqual(cb[-2].serial, 59)
self.assertEqual(chain.count_atom_sites(), 14)
self.assertEqual(chain.count_occupancies(), 8)
st.remove_alternative_conformations()
self.assertEqual(chain.count_atom_sites(), 8)
def parse_structure3d(struct_str, filename, logger):
t_start = timer()
output_filename = filename
if filename.endswith('.gz'):
struct_str = gzip.decompress(struct_str)
output_filename = filename[:-3]
if output_filename.endswith('.cif'):
cif_block = gemmi.cif.read_string(struct_str)[0]
output_string = gemmi.make_structure_from_block(cif_block)
elif output_filename.endswith('.pdb'):
output_string = gemmi.read_pdb_string(struct_str)
else:
# Should never reach here
raise RnaspiderUserError('Unknown file format was provided.')
t_end = timer()
logger.info(
f"Finished parsing input. Elapsed time: {int((t_end - t_start) * 1000)} ms"
)
return output_string
def test_5a11(self, use_populate=True):
st = gemmi.read_pdb_string(FRAGMENT_5A11)
a1 = st[0].sole_residue('A', gemmi.SeqId(37, ' '))[0]
ns = gemmi.NeighborSearch(st[0], st.cell, 5)
if use_populate:
ns.populate()
else:
for n_ch, chain in enumerate(st[0]):
for n_res, res in enumerate(chain):
for n_atom, atom in enumerate(res):
ns.add_atom(atom, n_ch, n_res, n_atom)
marks = ns.find_atoms(a1.pos, a1.altloc, 3)
m1, m2 = sorted(marks, key=lambda m: ns.dist(a1.pos, m.pos()))
self.assertAlmostEqual(ns.dist(a1.pos, m1.pos()), 0, delta=5e-6)
self.assertAlmostEqual(ns.dist(a1.pos, m2.pos()), 0.13, delta=5e-3)
cra2 = m2.to_cra(st[0])
self.assertEqual(cra2.chain.name, 'B')
self.assertEqual(str(cra2.residue.seqid), '37')
self.assertEqual(cra2.atom.name, 'SG')
marks2 = ns.find_neighbors(a1, 0.1, 3)
self.assertEqual(len(marks2), 1)
self.assertEqual(marks2[0], m2)
def test_ssbond_again(self):
st = gemmi.read_pdb_string(SSBOND_FRAGMENT)
doc = st.make_mmcif_document()
st2 = gemmi.make_structure_from_block(doc[0])
out = st2.make_pdb_headers()
self.assertEqual(out.splitlines(), SSBOND_FRAGMENT.splitlines()[:3])
def test_ssbond(self):
st = gemmi.read_pdb_string(SSBOND_FRAGMENT)
out = st.make_pdb_headers()
self.assertEqual(out.splitlines(), SSBOND_FRAGMENT.splitlines()[:3])
def from_pdb_string(pdb_string: str):
structure = gemmi.read_pdb_string(pdb_string)
return Structure(structure)
def to_gemmi(self):
structure = gemmi.read_pdb_string(self.string)
return structure
