def com(selection,state=None,mass=None,object=None, quiet=1, **kwargs):
"""
DESCRIPTION
Places a pseudoatom at the center of mass
Author: Sean Law
Michigan State University
slaw (at) msu . edu
SEE ALSO
pseudoatom, get_com
"""
quiet = int(quiet)
if (object == None):
object = cmd.get_legal_name(selection)
object = cmd.get_unused_name(object + "_COM", 0)
cmd.delete(object)
if (state != None):
x, y, z=get_com(selection,mass=mass, quiet=quiet)
if not quiet:
print "%f %f %f" % (x, y, z)
cmd.pseudoatom(object,pos=[x, y, z], **kwargs)
cmd.show("spheres",object)
else:
for i in range(cmd.count_states()):
x, y, z=get_com(selection,mass=mass,state=i+1, quiet=quiet)
if not quiet:
print "State %d:%f %f %f" % (i+1, x, y, z)
cmd.pseudoatom(object,pos=[x, y, z],state=i+1, **kwargs)
cmd.show("spheres", 'last ' + object)
def minimize(selection='all', forcefield='MMFF94s', method='conjugate gradients', nsteps=500, conv=0.0001, cutoff=False, cut_vdw=6.0, cut_elec=8.0):
pdb_string = cmd.get_pdbstr(selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats('pdb', 'pdb')
mol = ob.OBMol()
obconversion.ReadString(mol, pdb_string)
mol.AddHydrogens()
ff = ob.OBForceField.FindForceField(forcefield) # GAFF, MMFF94s, MMFF94, UFF, Ghemical
ff.Setup(mol)
if cutoff == True:
ff.EnableCutOff(True)
ff.SetVDWCutOff(cut_vdw)
ff.SetElectrostaticCutOff(cut_elec)
if method == 'conjugate gradients':
ff.ConjugateGradients(nsteps, conv)
else:
ff.SteepestDescent(nsteps, conv)
ff.GetCoordinates(mol)
nrg = ff.Energy()
pdb_string = obconversion.WriteString(mol)
cmd.delete(name)
if name == 'all':
name = 'all_'
cmd.read_pdbstr(pdb_string, name)
print '#########################################'
print 'The Energy of %s is %8.2f %s ' % (name, nrg, ff.GetUnit())
print '#########################################'
def minimize(selection='all',
forcefield='MMFF94s',
method='Conjugate Gradients',
nsteps0=500,
conv=0.0001,
cutoff=False,
cut_vdw=6.0,
cut_elec=8.0):
mol_string = cmd.get_str('mol', selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats('mol', 'mol')
mol = ob.OBMol()
obconversion.ReadString(mol, mol_string)
ff = ob.OBForceField.FindForceField(
forcefield) ## GAFF, MMFF94s, MMFF94, UFF, Ghemical
ff.Setup(mol)
if cutoff == True:
ff.EnableCutOff(True)
ff.SetVDWCutOff(cut_vdw)
ff.SetElectrostaticCutOff(cut_elec)
if method == 'Conjugate Gradients':
ff.ConjugateGradients(nsteps0, conv)
else:
ff.SteepestDescent(nsteps0, conv)
ff.GetCoordinates(mol)
nrg = ff.Energy()
mol_string = obconversion.WriteString(mol)
cmd.delete(name)
if name == 'all':
name = 'all_'
cmd.read_molstr(mol_string, name, state=0, finish=1, discrete=1)
print('#########################################')
print('The Energy of %s is %8.2f %s ' % (name, nrg, ff.GetUnit()))
print('#########################################')
def minimize(selection='all',
forcefield='MMFF94s',
method='cg',
nsteps=2000,
conv=1E-6,
cutoff=False,
cut_vdw=6.0,
cut_elec=8.0):
pdb_string = cmd.get_pdbstr(selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats('pdb', 'pdb')
mol = ob.OBMol()
obconversion.ReadString(mol, pdb_string)
ff = ob.OBForceField.FindForceField(forcefield)
ff.Setup(mol)
if cutoff == True:
ff.EnableCutOff(True)
ff.SetVDWCutOff(cut_vdw)
ff.SetElectrostaticCutOff(cut_elec)
if method == 'cg':
ff.ConjugateGradients(nsteps, conv)
else:
ff.SteepestDescent(nsteps, conv)
ff.GetCoordinates(mol)
nrg = ff.Energy()
pdb_string = obconversion.WriteString(mol)
cmd.delete(name)
if name == 'all':
name = 'all_'
cmd.read_pdbstr(pdb_string, name)
return nrg
def dehydron(selection='all', angle_range=40, max_distance=3.5, desolv=6.5, min_wrappers=19, quiet=0):
'''
DESCRIPTION
dehydron calculator
USAGE
dehydron [ selection [, angle_range [, max_distance [, desolv [, min_wrappers ]]]]]
'''
angle, max_distance = float(angle_range), float(max_distance)
desolv, min_wrappers = float(desolv), int(min_wrappers)
quiet = int(quiet)
name = cmd.get_legal_name('DH_%s' % selection)
cmd.delete(name)
selection_hb = '((%s) and polymer)' % (selection)
hb = cmd.find_pairs("((byres "+selection_hb+") and n. n)","((byres "+selection_hb+") and n. o)",mode=1,cutoff=max_distance,angle=angle_range)
if not quiet:
hb.sort(lambda x,y:(cmp(x[0][1],y[0][1])))
print "--------------------------------------------------------------------"
print "--------------------------Dehydron Results--------------------------"
print "--------------------------------------------------------------------"
print " Donor | Aceptor |"
print " Object Chain Residue | Object Chain Residue | # wrappers"
cmd.select('_nonpolar', '(elem C) and not (solvent or (elem N+O) extend 1)', 0)
try:
cmd.select('_selection', '%s' % selection, 0)
except:
pass
sel = []
for pairs in hb:
wrappers = cmd.count_atoms('((%s and _nonpolar and _selection) within %f of byca (%s`%d %s`%d))' %
((pairs[0][0], desolv) + pairs[0] + pairs[1]))
if wrappers < min_wrappers:
cmd.distance(name, pairs[0], pairs[1])
if not quiet:
cmd.iterate(pairs[0], 'stored.nitro = chain, resi, resn')
cmd.iterate(pairs[1], 'stored.oxy = chain, resi, resn')
print ' %12s%4s%6s%5d | %12s%4s%6s%5d |%7s' % (pairs[0][0], stored.nitro[0], stored.nitro[2], int(stored.nitro[1]), pairs[1][0], stored.oxy[0], stored.oxy[2], int(stored.oxy[1]), wrappers)
sel.append(pairs[0])
sel.append(pairs[1])
cmd.delete('_nonpolar')
cmd.delete('_selection')
if len(sel) > 0:
cmd.show_as('dashes', name)
elif not quiet and len(hb) != 0:
print ' - no dehydrons were found - '
else:
print ' - no hydrogen bonds were found - '
def conf_search(selection="all", forcefield="MMFF94s", method="Weighted", nsteps=500, conformers=25, lowest_conf=5):
pdb_string = cmd.get_pdbstr(selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats("pdb", "pdb")
mol = ob.OBMol()
obconversion.ReadString(mol, pdb_string)
mol.AddHydrogens()
ff = ob.OBForceField.FindForceField(forcefield) ## GAFF, MMFF94s, MMFF94, UFF, Ghemical
ff.Setup(mol)
if method == "Weighted":
ff.WeightedRotorSearch(conformers, nsteps)
elif method == "Random":
ff.RandomRotorSearch(conformers, nsteps)
else:
ff.SystematicRotorSearch(nsteps)
if name == "all":
name = "all_"
if method in ["Weighted", "Random"]:
ff.GetConformers(mol)
print "##############################################"
print " Conformer | Energy | RMSD"
nrg_unit = ff.GetUnit()
rmsd = 0
ff.GetCoordinates(mol)
nrg = ff.Energy()
conf_list = []
for i in range(conformers):
mol.SetConformer(i)
ff.Setup(mol)
nrg = ff.Energy()
conf_list.append((nrg, i))
conf_list.sort()
lenght_conf_list = len(conf_list)
if lowest_conf > lenght_conf_list:
lowest_conf = lenght_conf_list
for i in range(lowest_conf):
nrg, orden = conf_list[i]
name_n = "%s%02d" % (name, i)
cmd.delete(name_n)
mol.SetConformer(orden)
pdb_string = obconversion.WriteString(mol)
cmd.read_pdbstr(pdb_string, name_n)
if i != 0:
rmsd = cmd.fit(name_n, "%s00" % name, quiet=1)
print "%15s | %10.2f%9s |%6.1f" % (name_n, nrg, nrg_unit, rmsd)
print "##############################################"
else:
ff.GetCoordinates(mol)
nrg = ff.Energy()
pdb_string = obconversion.WriteString(mol)
cmd.delete(name)
cmd.read_pdbstr(pdb_string, name)
print "#########################################"
print "The Energy of %s is %8.2f %s " % (name, nrg, ff.GetUnit())
print "#########################################"
def conf_search(selection='all', forcefield='MMFF94s', method='Weighted', nsteps=500, conformers=25, lowest_conf=5):
pdb_string = cmd.get_pdbstr(selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats('pdb', 'pdb')
mol = ob.OBMol()
obconversion.ReadString(mol, pdb_string)
mol.AddHydrogens()
ff = ob.OBForceField.FindForceField(forcefield) # GAFF, MMFF94s, MMFF94, UFF, Ghemical
ff.Setup(mol)
if method == 'Weighted':
ff.WeightedRotorSearch(conformers, nsteps)
elif method == 'Random':
ff.RandomRotorSearch(conformers, nsteps)
else:
ff.SystematicRotorSearch(nsteps)
if name == 'all':
name = 'all_'
if method in ['Weighted', 'Random']:
ff.GetConformers(mol)
print '##############################################'
print ' Conformer | Energy | RMSD'
nrg_unit = ff.GetUnit()
rmsd = 0
ff.GetCoordinates(mol)
nrg = ff.Energy()
conf_list = []
for i in range(conformers):
mol.SetConformer(i)
ff.Setup(mol)
nrg = ff.Energy()
conf_list.append((nrg, i))
conf_list.sort()
lenght_conf_list = len(conf_list)
if lowest_conf > lenght_conf_list:
lowest_conf = lenght_conf_list
for i in range(lowest_conf):
nrg, orden = conf_list[i]
name_n = '%s%02d' % (name, i)
cmd.delete(name_n)
mol.SetConformer(orden)
pdb_string = obconversion.WriteString(mol)
cmd.read_pdbstr(pdb_string, name_n)
if i != 0:
rmsd = cmd.fit(name_n, '%s00' % name, quiet=1)
print '%15s | %10.2f%9s |%6.1f' % (name_n, nrg, nrg_unit, rmsd)
print '##############################################'
else:
ff.GetCoordinates(mol)
nrg = ff.Energy()
pdb_string = obconversion.WriteString(mol)
cmd.delete(name)
cmd.read_pdbstr(pdb_string, name)
print '#########################################'
print 'The Energy of %s is %8.2f %s ' % (name, nrg, ff.GetUnit())
print '#########################################'
def minimize(selection='tmp',
forcefield='MMFF94',
method='steepest descent',
nsteps=2000,
conv=1E-6,
cutoff=False,
cut_vdw=6.0,
cut_elec=8.0,
rigid_geometry=True):
"""
Use openbabel to minimize the energy of a molecule.
"""
pdb_string = cmd.get_pdbstr(selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats('pdb', 'pdb')
mol = ob.OBMol()
obconversion.ReadString(mol, pdb_string)
if rigid_geometry:
constraints = ob.OBFFConstraints()
for angle in ob.OBMolAngleIter(mol):
b, a, c = [mol.GetAtom(x + 1) for x in angle]
value = mol.GetAngle(a, b, c)
b, a, c = [x + 1 for x in angle]
constraints.AddAngleConstraint(a, b, c, value)
for i in ob.OBMolBondIter(mol):
a, b = (i.GetBeginAtomIdx(), i.GetEndAtomIdx())
value = i.GetLength()
constraints.AddDistanceConstraint(a, b, value)
ff = ob.OBForceField.FindForceField(forcefield)
ff.Setup(mol, constraints)
ff.SetConstraints(constraints)
else:
ff = ob.OBForceField.FindForceField(forcefield)
ff.Setup(mol)
if cutoff:
ff.EnableCutOff(True)
ff.SetVDWCutOff(cut_vdw)
ff.SetElectrostaticCutOff(cut_elec)
if method == 'conjugate gradients':
ff.ConjugateGradients(nsteps, conv)
else:
ff.SteepestDescent(nsteps, conv)
ff.GetCoordinates(mol)
nrg = ff.Energy()
pdb_string = obconversion.WriteString(mol)
cmd.delete(name)
if name == 'all':
name = 'all_'
cmd.delete(selection)
cmd.read_pdbstr(pdb_string, selection)
return nrg
def sketch_pseudo_coc(selection,
state=None,
name=None,
prefix='',
suffix='_coc',
**kwargs):
"""Create a pseudo atom which indicate the center of coordinate of the
selection
USAGE
sketch_pseudo_coc selection, state=state, name=name,
prefix=prefix, suffix=suffix
ARGUMENTS
selection a selection-expression
state a state-index if positive number or 0 to all, -1 to current
name a name of the pseudoatom, it will
automatically specified if None is specified (Default)
prefix a prefix of the pseudoatom. it will used only when name is
not specified
suffix a suffix of the pseudoatom. it will used only when name is
not specified
EXAMPLE
sketch_pcoc (resn PHE), state=10
"""
if name is None:
try:
name = cmd.get_legal_name(selection)
name = cmd.get_unused_name('{}{}{}'.format(prefix, name, suffix),
0)
except:
name = '%s%s' % (prefix, suffix)
if state is not None:
com = geometry.find_center_of_coordinates(selection)
cmd.pseudoatom(name, pos=com, **kwargs)
else:
for state in range(1, cmd.count_states() + 1):
com = geometry.find_center_of_coordinates(selection, state=state)
cmd.pseudoatom(name, pos=com, state=state, **kwargs)
def minimize(selection='tmp', forcefield='MMFF94', method='steepest descent', nsteps= 2000, conv=1E-6, cutoff=False, cut_vdw=6.0, cut_elec=8.0, rigid_geometry=True):
"""
Write Me!
"""
pdb_string = cmd.get_pdbstr(selection)
name = cmd.get_legal_name(selection)
obconversion = ob.OBConversion()
obconversion.SetInAndOutFormats('pdb', 'pdb')
mol = ob.OBMol()
obconversion.ReadString(mol, pdb_string)
if rigid_geometry:
constraints = ob.OBFFConstraints()
for angle in ob.OBMolAngleIter(mol):
b, a, c = [mol.GetAtom(x+1) for x in angle]
value = mol.GetAngle(a, b, c)
b, a, c = [x+1 for x in angle]
constraints.AddAngleConstraint(a, b, c, value)
for i in ob.OBMolBondIter(mol):
a, b = (i.GetBeginAtomIdx(), i.GetEndAtomIdx())
value = i.GetLength()
constraints.AddDistanceConstraint(a, b, value)
ff = ob.OBForceField.FindForceField(forcefield)
ff.Setup(mol, constraints)
ff.SetConstraints(constraints)
else:
ff = ob.OBForceField.FindForceField(forcefield)
ff.Setup(mol)
if cutoff:
ff.EnableCutOff(True)
ff.SetVDWCutOff(cut_vdw)
ff.SetElectrostaticCutOff(cut_elec)
if method == 'conjugate gradients':
ff.ConjugateGradients(nsteps, conv)
else:
ff.SteepestDescent(nsteps, conv)
ff.GetCoordinates(mol)
nrg = ff.Energy()
pdb_string = obconversion.WriteString(mol)
cmd.delete(name)
if name == 'all':
name = 'all_'
cmd.delete(selection)
cmd.read_pdbstr(pdb_string, selection)
return nrg
def sketch_pseudo_coc(selection, state=None, name=None,
prefix='', suffix='_coc', **kwargs):
"""Create a pseudo atom which indicate the center of coordinate of the
selection
USAGE
sketch_pseudo_coc selection, state=state, name=name,
prefix=prefix, suffix=suffix
ARGUMENTS
selection a selection-expression
state a state-index if positive number or 0 to all, -1 to current
name a name of the pseudoatom, it will
automatically specified if None is specified (Default)
prefix a prefix of the pseudoatom. it will used only when name is
not specified
suffix a suffix of the pseudoatom. it will used only when name is
not specified
EXAMPLE
sketch_pcoc (resn PHE), state=10
"""
if name is None:
try:
name = cmd.get_legal_name(selection)
name = cmd.get_unused_name(
'{}{}{}'.format(prefix, name, suffix), 0
)
except:
name = '%s%s' % (prefix, suffix)
if state is not None:
com = geometry.find_center_of_coordinates(selection)
cmd.pseudoatom(name, pos=com, **kwargs)
else:
for state in range(1, cmd.count_states()+1):
com = geometry.find_center_of_coordinates(selection, state=state)
cmd.pseudoatom(name, pos=com, state=state, **kwargs)
def com(selection, state=None, mass=None, object=None, quiet=1, **kwargs):
quiet = int(quiet)
if (object == None):
try:
object = cmd.get_legal_name(selection)
object = cmd.get_unused_name(object + "_COM", 0)
except AttributeError:
object = 'COM'
cmd.delete(object)
if (state != None):
x, y, z = get_com(selection, mass=mass, quiet=quiet)
if not quiet:
print("%f %f %f" % (x, y, z))
cmd.pseudoatom(object, pos=[x, y, z], **kwargs)
cmd.show("spheres", object)
else:
for i in range(cmd.count_states()):
x, y, z = get_com(selection, mass=mass, state=i + 1, quiet=quiet)
if not quiet:
print("State %d:%f %f %f" % (i + 1, x, y, z))
cmd.pseudoatom(object, pos=[x, y, z], state=i + 1, **kwargs)
cmd.show("spheres", 'last ' + object)
def com(selection, state=None, mass=None, object=None, quiet=1, **kwargs):
quiet = int(quiet)
if (object == None):
try:
object = cmd.get_legal_name(selection)
object = cmd.get_unused_name(object + "_COM", 0)
except AttributeError:
object = 'COM'
cmd.delete(object)
if (state != None):
x, y, z = get_com(selection, mass=mass, quiet=quiet)
if not quiet:
print("%f %f %f" % (x, y, z))
cmd.pseudoatom(object, pos=[x, y, z], **kwargs)
cmd.show("spheres", object)
else:
for i in range(cmd.count_states()):
x, y, z = get_com(selection, mass=mass, state=i + 1, quiet=quiet)
if not quiet:
print("State %d:%f %f %f" % (i + 1, x, y, z))
cmd.pseudoatom(object, pos=[x, y, z], state=i + 1, **kwargs)
cmd.show("spheres", 'last ' + object)
def uniprot_features(uniprot_id, selection='(all)', withss=0,
prefix='feature_', quiet=1):
'''
DESCRIPTION
Fetch feature list from uniprot.org and create named selections.
Requires residue numbering (resi) to match uniprot sequence!
ARGUMENTS
uniprot_id = string: UniProtKB name or accession
selection = string: atom selection {default: all}
withss = 0/1: update secondary structure {default: 0}
'''
import xml.etree.ElementTree as etree
from urllib import urlopen
withss, quiet = int(withss), int(quiet)
url = 'http://www.uniprot.org/uniprot/%s.xml' % uniprot_id
if not quiet:
print 'Downloading', url
doc = etree.parse(urlopen(url))
ns = 'http://uniprot.org/uniprot'
NS = {'u': ns}
if not quiet:
print 'Parsing Features'
features = doc.findall('{%s}entry/{%s}feature' % (ns, ns))
if withss == 1:
cmd.alter(selection, 'ss="L"')
ssmap = {'helix': 'H', 'strand': 'S', 'turn': 'L'}
norange_types = ['disulfide bond']
count = 0
for feature in features:
type = feature.get('type')
begin = feature.find('{%s}location/{%s}begin' % (ns, ns))
if begin is not None:
end = feature.find('{%s}location/{%s}end' % (ns, ns))
values = (selection, begin.get('position'), end.get('position'))
if type in norange_types:
sel = '(%s) and resi %s+%s' % values
else:
sel = '(%s) and resi %s-%s' % values
else:
position = feature.find('{%s}location/{%s}position' % (ns, ns))
sel = '(%s) and resi %s' % (selection, position.get('position'))
if type in ['helix', 'strand', 'turn'] and withss < 2:
if withss == 1:
cmd.alter(sel, 'ss="%s"' % ssmap.get(type, 'L'))
else:
count += 1
name = cmd.get_legal_name('%s%03d_%s' % (prefix, count,
feature.get('description', '').replace('.', '')))
groupname = cmd.get_legal_name('%s%s' % (prefix, type))
cmd.select(name, sel)
cmd.group(groupname, name, 'add')
if not quiet:
print 'Found %d feature records (without secondary structures)' % count
def show_contacts(selection='*', selection2='*',
result="contacts",
cutoff=3.6,
bigcutoff = 4.0,
labels=False,
SC_DEBUG = DEBUG):
"""
USAGE
show_contacts selection, selection2, [result=contacts],[cutoff=3.6],[bigcutoff=4.0]
Show various polar contacts, the good, the bad, and the ugly.
Edit MPB 6-26-14: The distances are heavy atom distances, so I upped the default cutoff to 4.0
Returns:
True/False - if False, something went wrong
"""
if SC_DEBUG > 4:
print('Starting show_contacts')
print('selection = "' + selection + '"')
print('selection2 = "' + selection2 + '"')
result = cmd.get_legal_name(result)
#if the group of contacts already exist, delete them
cmd.delete(result)
# ensure only N and O atoms are in the selection
all_don_acc1 = selection + " and (donor or acceptor)"
all_don_acc2 = selection2 + " and (donor or acceptor)"
if SC_DEBUG > 4:
print('all_don_acc1 = "' + all_don_acc1 + '"')
print('all_don_acc2 = "' + all_don_acc2 + '"')
#if theses selections turn out not to have any atoms in them, pymol throws cryptic errors when calling the dist function like:
#'Selector-Error: Invalid selection name'
#So for each one, manually perform the selection and then pass the reference to the distance command and at the end, clean up the selections
#the return values are the count of the number of atoms
all1_sele_count = cmd.select('all_don_acc1_sele', all_don_acc1)
all2_sele_count = cmd.select('all_don_acc2_sele', all_don_acc2)
#print out some warnings
if DEBUG > 3:
if not all1_sele_count:
print('Warning: all_don_acc1 selection empty!')
if not all2_sele_count:
print('Warning: all_don_acc2 selection empty!')
########################################
allres = result + "_all"
if all1_sele_count and all2_sele_count:
#print(allres)
#print(cmd.get_distance(allres, 'all_don_acc1_sele', 'all_don_acc2_sele', bigcutoff, mode = 0))
any = cmd.distance(allres, 'all_don_acc1_sele', 'all_don_acc2_sele', bigcutoff, mode = 0)
# if any is 0 it seems that there is no distance!
if any:
cmd.set("dash_radius", "0.05", allres)
if not labels:
cmd.hide("labels", allres)
else:
# just do nothing and clena up
print('no contacts')
cmd.delete('all_don_acc1_sele')
cmd.delete('all_don_acc2_sele')
cmd.delete(result + "_all")
return None
########################################
# compute good polar interactions according to pymol
polres = result + "_polar"
if all1_sele_count and all2_sele_count:
cmd.distance(polres, 'all_don_acc1_sele', 'all_don_acc2_sele', cutoff, mode = 2) #hopefully this checks angles? Yes
#cmd.set("dash_color", "marine", allres)
#cmd.set('dash_gap', '0')
cmd.set("dash_radius","0.2", polres) #"0.126"
#cmd.set("dash_color", "marine", allres)
if not labels:
cmd.hide("labels", polres)
########################################
# When running distance in mode=2, the cutoff parameter is ignored if set higher then the default of 3.6
# so set it to the passed in cutoff and change it back when you are done.
old_h_bond_cutoff_center = cmd.get('h_bond_cutoff_center') # ideal geometry
old_h_bond_cutoff_edge = cmd.get('h_bond_cutoff_edge') # minimally acceptable geometry
cmd.set('h_bond_cutoff_center', bigcutoff)
cmd.set('h_bond_cutoff_edge', bigcutoff)
# compute possibly suboptimal polar interactions using the user specified distance
pol_ok_res = result + "_polar_ok"
if all1_sele_count and all2_sele_count:
cmd.distance(pol_ok_res, 'all_don_acc1_sele', 'all_don_acc2_sele', bigcutoff, mode = 2)
cmd.set("dash_radius", "0.06", pol_ok_res)
if not labels:
cmd.hide("labels", pol_ok_res)
#now reset the h_bond cutoffs
cmd.set('h_bond_cutoff_center', old_h_bond_cutoff_center)
cmd.set('h_bond_cutoff_edge', old_h_bond_cutoff_edge)
########################################
onlyacceptors1 = selection + " and (acceptor and !donor)"
onlyacceptors2 = selection2 + " and (acceptor and !donor)"
onlydonors1 = selection + " and (!acceptor and donor)"
onlydonors2 = selection2 + " and (!acceptor and donor)"
#perform the selections
onlyacceptors1_sele_count = cmd.select('onlyacceptors1_sele', onlyacceptors1)
onlyacceptors2_sele_count = cmd.select('onlyacceptors2_sele', onlyacceptors2)
onlydonors1_sele_count = cmd.select('onlydonors1_sele', onlydonors1)
onlydonors2_sele_count = cmd.select('onlydonors2_sele', onlydonors2)
#print out some warnings
if SC_DEBUG > 2:
if not onlyacceptors1_sele_count:
print('Warning: onlyacceptors1 selection empty!')
if not onlyacceptors2_sele_count:
print('Warning: onlyacceptors2 selection empty!')
if not onlydonors1_sele_count:
print('Warning: onlydonors1 selection empty!')
if not onlydonors2_sele_count:
print('Warning: onlydonors2 selection empty!')
# acceptors acceptors
accres = result+"_aa"
if onlyacceptors1_sele_count and onlyacceptors2_sele_count:
aa_dist_out = cmd.distance(accres, 'onlyacceptors1_sele', 'onlyacceptors2_sele', cutoff, 0)
if aa_dist_out < 0:
print('\n\nCaught a pymol selection error in acceptor-acceptor selection of show_contacts')
print('accres:', accres)
print('onlyacceptors1', onlyacceptors1)
print('onlyacceptors2', onlyacceptors2)
return False
cmd.set("dash_color","red",accres)
cmd.set("dash_radius","0.125",accres)
if not labels:
cmd.hide("labels", accres)
########################################
# donors donors
donres = result+"_dd"
if onlydonors1_sele_count and onlydonors2_sele_count:
dd_dist_out = cmd.distance(donres, 'onlydonors1_sele', 'onlydonors2_sele', cutoff, 0)
#try to catch the error state
if dd_dist_out < 0:
print('\n\nCaught a pymol selection error in dd selection of show_contacts')
print('donres:', donres)
print('onlydonors1', onlydonors1)
print('onlydonors2', onlydonors2)
print("cmd.distance('" + donres + "', '" + onlydonors1 + "', '" + onlydonors2 + "', " + str(cutoff) + ", 0)")
return False
cmd.set("dash_color","red",donres)
cmd.set("dash_radius","0.125",donres)
if not labels:
cmd.hide("labels", donres)
##########################################################
##### find the buried unpaired atoms of the receptor #####
##########################################################
#initialize the variable for when CALC_SASA is False
unpaired_atoms = ''
## Group
print(allres) # contacts_all
cmd.group(result,"%s %s %s %s %s %s" % (polres, allres, accres, donres, pol_ok_res, unpaired_atoms))
## Clean up the selection objects
#if the show_contacts debug level is high enough, don't delete them.
if SC_DEBUG < 5:
cmd.delete('all_don_acc1_sele')
cmd.delete('all_don_acc2_sele')
cmd.delete('onlyacceptors1_sele')
cmd.delete('onlyacceptors2_sele')
cmd.delete('onlydonors1_sele')
cmd.delete('onlydonors2_sele')
cmd.disable('contacts_all')
cmd.disable('contacts_polar_ok')
cmd.disable('contacts_aa')
cmd.disable('contacts_dd')
return True
def testGetLegalName(self):
self.assertEqual(cmd.get_legal_name("foo bar baz"), "foo_bar_baz")
def dehydron(selection='all',
angle_range=40,
max_distance=3.5,
desolv=6.5,
min_wrappers=19,
quiet=0):
'''
DESCRIPTION
dehydron calculator
USAGE
dehydron [ selection [, angle_range [, max_distance [, desolv [, min_wrappers ]]]]]
'''
angle, max_distance = float(angle_range), float(max_distance)
desolv, min_wrappers = float(desolv), int(min_wrappers)
quiet = int(quiet)
name = cmd.get_legal_name('DH_%s' % selection)
cmd.delete(name)
selection_hb = '((%s) and polymer)' % (selection)
hb = cmd.find_pairs("((byres " + selection_hb + ") and n. n)",
"((byres " + selection_hb + ") and n. o)",
mode=1,
cutoff=max_distance,
angle=angle_range)
if not quiet:
hb.sort(lambda x, y: (cmp(x[0][1], y[0][1])))
print "--------------------------------------------------------------------"
print "--------------------------Dehydron Results--------------------------"
print "--------------------------------------------------------------------"
print " Donor | Aceptor |"
print " Object Chain Residue | Object Chain Residue | # wrappers"
cmd.select('_nonpolar',
'(elem C) and not (solvent or (elem N+O) extend 1)', 0)
try:
cmd.select('_selection', '%s' % selection, 0)
except:
pass
sel = []
total_wrappers = 0
for pairs in hb:
wrappers = cmd.count_atoms(
'((%s and _nonpolar and _selection) within %f of byca (%s`%d %s`%d))'
% ((pairs[0][0], desolv) + pairs[0] + pairs[1]))
total_wrappers = total_wrappers + wrappers
if wrappers < min_wrappers:
cmd.distance(name, pairs[0], pairs[1])
if not quiet:
cmd.iterate(pairs[0], 'stored.nitro = chain, resi, resn')
cmd.iterate(pairs[1], 'stored.oxy = chain, resi, resn')
print ' %12s%4s%6s%6s | %12s%4s%6s%6s |%7s' % (
pairs[0][0], stored.nitro[0], stored.nitro[2],
stored.nitro[1], pairs[1][0], stored.oxy[0], stored.oxy[2],
stored.oxy[1], wrappers)
sel.append(pairs[0])
sel.append(pairs[1])
cmd.delete('_nonpolar')
cmd.delete('_selection')
#compute the z_scores for validation porpoises.
stored.ResiduesNames = []
cmd.iterate('(name ca)', 'stored.ResiduesNames.append((resn))')
total_residues = float(len(stored.ResiduesNames))
z_score_wrappers = ((total_wrappers / total_residues) - 17) / 2
z_score_hb = ((len(hb) / total_residues) - 0.62) / 0.06
if len(sel) > 0:
cmd.show_as('dashes', name)
print '\nz-score wrappers = %6.2f\nz-score hydrogen bonds = %6.2f\n' % (
z_score_wrappers, z_score_hb)
elif not quiet and len(hb) != 0:
print '\n - no dehydrons were found - '
print '\nz-score hydrogen bonds = %6.2f\n' % (z_score_hb)
else:
print '\n - no hydrogen bonds were found - '
def uniprot_features(uniprot_id, selection='(all)', withss=0,
prefix='feature_', sm=None, quiet=1):
'''
DESCRIPTION
Fetch feature list from uniprot.org and create named selections.
Requires residue numbering (resi) to match uniprot sequence!
ARGUMENTS
uniprot_id = string: UniProtKB name or accession
selection = string: atom selection {default: all}
withss = 0/1: update secondary structure {default: 0}
'''
import xml.etree.ElementTree as etree
try:
from urllib import urlopen
except ImportError:
from urllib.request import urlopen
withss, quiet = int(withss), int(quiet)
url = 'http://www.uniprot.org/uniprot/%s.xml' % uniprot_id
if not quiet:
print('Downloading', url)
doc = etree.parse(urlopen(url))
ns = 'http://uniprot.org/uniprot'
NS = {'u': ns}
if not quiet:
print('Parsing Features')
features = doc.findall('{%s}entry/{%s}feature' % (ns, ns))
sequence = doc.findtext('{%s}entry/{%s}sequence' % (ns, ns))
sequence = ''.join(sequence.split())
try:
if sm is None:
sm = resid_mapper.from_seq_sel(sequence, selection)
except:
print(' Warning: sequence mapping failed')
sm = lambda x: x
if withss == 1:
cmd.alter(selection, 'ss="L"')
ssmap = {'helix': 'H', 'strand': 'S', 'turn': 'L'}
norange_types = ['disulfide bond']
count = 0
for feature in features:
type_ = feature.get('type')
begin = feature.find('{%s}location/{%s}begin' % (ns, ns))
if begin is not None:
end = feature.find('{%s}location/{%s}end' % (ns, ns))
values = begin.get('position'), end.get('position')
if type_ in norange_types:
try:
values = sm(values[0]), sm(values[1])
sel = '(' + selection + ') and resi %s+%s' % values
except KeyError:
print(' Warning: could not map', values)
sel = 'none'
else:
try:
values = sm(values)
sel = '(' + selection + ') and resi %s-%s' % values
except KeyError:
print(' Warning: could not map', values)
sel = 'none'
else:
position = feature.find('{%s}location/{%s}position' % (ns, ns))
try:
value = sm(position.get('position'))
sel = '(%s) and resi %s' % (selection, value)
except KeyError:
sel = 'none'
if type_ in ['helix', 'strand', 'turn'] and withss < 2:
if withss == 1:
cmd.alter(sel, 'ss="%s"' % ssmap.get(type_, 'L'))
else:
count += 1
name = cmd.get_legal_name('%s%03d_%s' % (prefix, count,
feature.get('description', '').replace('.', '')))
groupname = cmd.get_legal_name('%s%s' % (prefix, type_))
cmd.select(name, sel)
cmd.group(groupname, name, 'add')
if not quiet:
print('Found %d feature records (without secondary structures)' % count)
def uniprot_features(uniprot_id, selection="(all)", withss=0, prefix="feature_", sm=None, quiet=1):
"""
DESCRIPTION
Fetch feature list from uniprot.org and create named selections.
Requires residue numbering (resi) to match uniprot sequence!
ARGUMENTS
uniprot_id = string: UniProtKB name or accession
selection = string: atom selection {default: all}
withss = 0/1: update secondary structure {default: 0}
"""
import xml.etree.ElementTree as etree
from urllib import urlopen
withss, quiet = int(withss), int(quiet)
url = "http://www.uniprot.org/uniprot/%s.xml" % uniprot_id
if not quiet:
print "Downloading", url
doc = etree.parse(urlopen(url))
ns = "http://uniprot.org/uniprot"
NS = {"u": ns}
if not quiet:
print "Parsing Features"
features = doc.findall("{%s}entry/{%s}feature" % (ns, ns))
sequence = doc.findtext("{%s}entry/{%s}sequence" % (ns, ns))
sequence = "".join(sequence.split())
try:
if sm is None:
sm = resid_mapper.from_seq_sel(sequence, selection)
except:
print " Warning: sequence mapping failed"
sm = lambda x: x
if withss == 1:
cmd.alter(selection, 'ss="L"')
ssmap = {"helix": "H", "strand": "S", "turn": "L"}
norange_types = ["disulfide bond"]
count = 0
for feature in features:
type = feature.get("type")
begin = feature.find("{%s}location/{%s}begin" % (ns, ns))
if begin is not None:
end = feature.find("{%s}location/{%s}end" % (ns, ns))
values = begin.get("position"), end.get("position")
if type in norange_types:
try:
values = sm(values[0]), sm(values[1])
sel = "(" + selection + ") and resi %s+%s" % values
except KeyError:
print " Warning: could not map", values
sel = "none"
else:
try:
values = sm(values)
sel = "(" + selection + ") and resi %s-%s" % values
except KeyError:
print " Warning: could not map", values
sel = "none"
else:
position = feature.find("{%s}location/{%s}position" % (ns, ns))
try:
value = sm(position.get("position"))
sel = "(%s) and resi %s" % (selection, value)
except KeyError:
sel = "none"
if type in ["helix", "strand", "turn"] and withss < 2:
if withss == 1:
cmd.alter(sel, 'ss="%s"' % ssmap.get(type, "L"))
else:
count += 1
name = cmd.get_legal_name("%s%03d_%s" % (prefix, count, feature.get("description", "").replace(".", "")))
groupname = cmd.get_legal_name("%s%s" % (prefix, type))
cmd.select(name, sel)
cmd.group(groupname, name, "add")
if not quiet:
print "Found %d feature records (without secondary structures)" % count
def dehydron(selection='all',
angle_range=40.,
max_distance=3.5,
desolv=6.5,
min_wrappers=19,
max_wrappers=35,
quiet=0):
'''
DESCRIPTION
dehydron calculator
USAGE
wrappy [ selection [, angle_range [, max_distance [, desolv [, min_wrappers [, max_wrappers ]]]]]]
'''
angle, max_distance = float(angle_range), float(max_distance)
desolv, min_wrappers, max_wrappers = float(desolv), int(min_wrappers), int(
max_wrappers)
quiet = int(quiet)
DH_name = cmd.get_legal_name('DH_%s' % selection)
cmd.delete(DH_name)
HBA_name = cmd.get_legal_name('HBA_%s' % selection)
cmd.delete(HBA_name)
HBO_name = cmd.get_legal_name('HBO_%s' % selection)
cmd.delete(HBO_name)
selection_hb = '((%s) and polymer)' % (selection)
hb = cmd.find_pairs("((byres " + selection_hb + ") and n. n)",
"((byres " + selection_hb + ") and n. o)",
mode=1,
cutoff=max_distance,
angle=angle_range)
cmd.select('_nonpolar',
'(elem C) and not (solvent or (elem N+O) extend 1)', 0)
try:
cmd.select('_selection', '%s' % selection, 0)
except:
pass
low_sel = []
mean_sel = []
high_sel = []
total_wrappers = 0
for pairs in hb:
wrappers = cmd.count_atoms(
'((%s and _nonpolar and _selection) within %f of byca (%s`%d %s`%d))'
% ((pairs[0][0], desolv) + pairs[0] + pairs[1]))
total_wrappers = total_wrappers + wrappers
cmd.iterate(pairs[0], 'stored.donor = chain, resi, resn')
cmd.iterate(pairs[1], 'stored.aceptor = chain, resi, resn')
if wrappers < min_wrappers:
cmd.distance(DH_name, pairs[0], pairs[1])
line = (wrappers,
'%12s%4s%6s%6s | %12s%4s%6s%6s |%7s below' %
(pairs[0][0], stored.donor[0], stored.donor[2],
stored.donor[1], pairs[1][0], stored.aceptor[0],
stored.aceptor[2], stored.aceptor[1], wrappers))
low_sel.append(line)
elif wrappers < max_wrappers:
cmd.distance(HBA_name, pairs[0], pairs[1])
line = (wrappers,
'%12s%4s%6s%6s | %12s%4s%6s%6s |%7s average' %
(pairs[0][0], stored.donor[0], stored.donor[2],
stored.donor[1], pairs[1][0], stored.aceptor[0],
stored.aceptor[2], stored.aceptor[1], wrappers))
mean_sel.append(line)
else:
cmd.distance(HBO_name, pairs[0], pairs[1])
line = (wrappers,
'%12s%4s%6s%6s | %12s%4s%6s%6s |%7s over' %
(pairs[0][0], stored.donor[0], stored.donor[2],
stored.donor[1], pairs[1][0], stored.aceptor[0],
stored.aceptor[2], stored.aceptor[1], wrappers))
high_sel.append(line)
cmd.delete('_nonpolar')
cmd.delete('_selection')
# compute the z_scores. Useful for protein structure validation.
stored.ResiduesNames = []
cmd.iterate('(name ca)', 'stored.ResiduesNames.append((resn))')
total_residues = float(len(stored.ResiduesNames))
z_score_wrappers = ((total_wrappers / total_residues) - 17) / 2
z_score_hb = ((len(hb) / total_residues) - 0.62) / 0.06
if len(low_sel) > 0:
cmd.show_as('dashes', DH_name)
cmd.color('red', DH_name)
low_sel.sort()
if len(mean_sel) > 0:
cmd.show_as('dashes', HBA_name)
cmd.color('yellow', HBA_name)
mean_sel.sort()
if len(high_sel) > 0:
cmd.show_as('dashes', HBO_name)
cmd.color('green', HBO_name)
high_sel.sort()
if not quiet:
hb.sort()
print "--------------------------------------------------------------------"
print "------------------------------Results ------------------------------"
print "--------------------------------------------------------------------"
print " Donor | Aceptor |"
print " Object Chain Residue | Object Chain Residue | # wrappers wrapping"
for line in low_sel:
print line[1]
for line in mean_sel:
print line[1]
for line in high_sel:
print line[1]
print '\nProtein global statistics:'
print '\nz-score wrappers = %6.2f\nz-score hydrogen bonds = %6.2f\n' % (
z_score_wrappers, z_score_hb)
elif not quiet and len(hb) != 0:
print '\n - no dehydrons were found - '
print '\nz-score hydrogen bonds = %6.2f\n' % (z_score_hb)
else:
print '\n - no hydrogen bonds were found - '
def process_session(
ensemble_collector,
patterns,
group,
max_size,
plot,
plot_annot,
plot_class,
plot_method,
table,
base_root=None,
):
"""Main plugin code."""
results = {}
for pattern in sorted(patterns):
for path in glob(pattern):
if base_root is None:
root = pm.get_legal_name(splitext(basename(path))[0])
else:
root = base_root
results[root] = [], []
ensembles, clusters = results[root]
if group:
root = f"{group}.{root}"
else:
root = root
with disable_feedback("all", "warnings"):
with settings(group_auto_mode=1):
pm.load(path)
try:
collected_ensembles = list(ensemble_collector(max_size))
if len(collected_ensembles) == 0:
raise Exception("No ensembles found.")
except:
raise CmdException(f"File {path} is invalid.")
with settings(group_auto_mode=2):
pm.create(f"{root}.protein", "protein")
pm.delete("protein")
i = 0
for ensemble in collected_ensembles:
klass = ensemble.klass
if klass:
i += 1
pm.hide("sticks", ensemble.selection)
pm.show("line", ensemble.selection)
pm.util.cbas(ensemble.selection)
obj = f"{root}.{klass}.{i:03}"
pm.create(obj, ensemble.selection)
if hasattr(pm, "set_property"):
pm.set_property("Class", ensemble.klass, obj)
pm.set_property("S", ensemble.strength, obj)
pm.set_property("S (CS0)",
ensemble.clusters[0].strength, obj)
pm.set_property("CD",
ensemble.max_center_to_center, obj)
pm.set_property("MD", ensemble.max_dist, obj)
ensemble.selection = obj
ensembles.append(ensemble)
for i, cluster in enumerate(Cluster.collect_atlas()):
pm.hide("sticks", cluster.selection)
pm.show("line", cluster.selection)
pm.util.cbay(cluster.selection)
obj = f"{root}.CS.{i:03}_{cluster.strength:03}"
pm.create(obj, cluster.selection)
pm.delete(cluster.selection)
cluster.selection = obj
clusters.append(cluster)
pm.color("yellow", f"{root}.CS.*")
pm.color("salmon", f"{root}.B.* or {root}.Bs.* {root}.Bl.*")
pm.color("red", f"{root}.D.* or {root}.Ds.* {root}.Dl.*")
pm.hide("lines", f"{root}.*")
pm.disable(f"{root}.CS")
pm.show("mesh", f"{root}.B.* or {root}.Bs.* {root}.Bl.*")
pm.show("mesh", f"{root}.D.* or {root}.Ds.* {root}.Dl.*")
pm.show("mesh", f"{root}.CS.*")
pm.hide("nb_spheres", "*label")
pm.orient(root)
if plot:
roots = []
selections = []
for root in sorted(results):
for ensemble in results[root][0]:
roots.append(root)
selections.append(ensemble.selection)
if plot_class:
selections = [s for s in selections if "." + plot_class + "." in s]
matrix_sim = np.zeros((len(selections), len(selections)))
matrix_over = np.zeros((len(selections), len(selections)))
for i, (root1, selection1) in enumerate(zip(roots, selections)):
for j, (root2, selection2) in enumerate(zip(roots, selections)):
matrix_sim[i][j] = nearby_aminoacids_similarity(
selection1,
selection2,
polymer1=root1 + ".protein",
polymer2=root2 + ".protein",
method=plot_method,
verbose=False,
)
matrix_over[i][j] = get_fractional_overlap(selection1,
selection2,
verbose=0)
fig, ax = plt.subplots(1, 2)
sb.heatmap(
matrix_sim,
vmax=1,
vmin=0,
xticklabels=selections,
yticklabels=selections,
annot=plot_annot,
cmap="YlGnBu",
ax=ax[0],
)
sb.heatmap(
matrix_over,
vmax=1,
vmin=0,
yticklabels=selections,
xticklabels=selections,
annot=plot_annot,
cmap="YlGnBu",
ax=ax[1],
)
ax[0].set_title(plot_method + " coefficient")
ax[0].set_xticklabels(selections, rotation=45, ha="right")
ax[1].set_title("fractional overlap")
ax[1].set_xticklabels(selections, rotation=45, ha="right")
plt.show()
if table:
tmpl = Template("""
<table>
<tr>
<th>Object</th>
<th>Class</th>
<th>S</th>
<th>S0</th>
<th>CD</th>
<th>MD</th>
</tr>
{% for root in results %}
{% for e in results[root][0] %}
<tr>
<td>{{ e.selection }}</td>
<td>{{ e.klass }}</td>
<td>{{ e.strength }}</td>
<td>{{ e.strength0 }}</td>
<td>{{ e.max_center_to_center | round(2) }}</td>
<td>{{ e.max_dist | round(2) }}</td>
</tr>
{% endfor %}
{% endfor %}
</table>
""")
_, path = tempfile.mkstemp(suffix=".html")
with open(path, "w") as fd:
fd.write(tmpl.render(results=results))
webbrowser.open(path)
return results
def dehydron(selection='all', angle_range=40., max_distance=3.5, desolv=6.5, min_wrappers=19, max_wrappers=35, quiet=0):
'''
DESCRIPTION
dehydron calculator
USAGE
wrappy [ selection [, angle_range [, max_distance [, desolv [, min_wrappers [, max_wrappers ]]]]]]
'''
angle, max_distance = float(angle_range), float(max_distance)
desolv, min_wrappers, max_wrappers = float(desolv), int(min_wrappers), int(max_wrappers)
quiet = int(quiet)
DH_name = cmd.get_legal_name('DH_%s' % selection)
cmd.delete(DH_name)
HBA_name = cmd.get_legal_name('HBA_%s' % selection)
cmd.delete(HBA_name)
HBO_name = cmd.get_legal_name('HBO_%s' % selection)
cmd.delete(HBO_name)
selection_hb = '((%s) and polymer)' % (selection)
hb = cmd.find_pairs("((byres " + selection_hb + ") and n. n)", "((byres " + selection_hb + ") and n. o)", mode=1, cutoff=max_distance, angle=angle_range)
cmd.select('_nonpolar', '(elem C) and not (solvent or (elem N+O) extend 1)', 0)
try:
cmd.select('_selection', '%s' % selection, 0)
except:
pass
low_sel = []
mean_sel = []
high_sel = []
total_wrappers = 0
for pairs in hb:
wrappers = cmd.count_atoms('((%s and _nonpolar and _selection) within %f of byca (%s`%d %s`%d))' %
((pairs[0][0], desolv) + pairs[0] + pairs[1]))
total_wrappers = total_wrappers + wrappers
cmd.iterate(pairs[0], 'stored.donor = chain, resi, resn')
cmd.iterate(pairs[1], 'stored.aceptor = chain, resi, resn')
if wrappers < min_wrappers:
cmd.distance(DH_name, pairs[0], pairs[1])
line = (wrappers, '%12s%4s%6s%6s | %12s%4s%6s%6s |%7s below' % (pairs[0][0], stored.donor[0], stored.donor[2], stored.donor[1], pairs[1][0], stored.aceptor[0], stored.aceptor[2], stored.aceptor[1], wrappers))
low_sel.append(line)
elif wrappers < max_wrappers:
cmd.distance(HBA_name, pairs[0], pairs[1])
line = (wrappers, '%12s%4s%6s%6s | %12s%4s%6s%6s |%7s average' % (pairs[0][0], stored.donor[0], stored.donor[2], stored.donor[1], pairs[1][0], stored.aceptor[0], stored.aceptor[2], stored.aceptor[1], wrappers))
mean_sel.append(line)
else:
cmd.distance(HBO_name, pairs[0], pairs[1])
line = (wrappers, '%12s%4s%6s%6s | %12s%4s%6s%6s |%7s over' % (pairs[0][0], stored.donor[0], stored.donor[2], stored.donor[1], pairs[1][0], stored.aceptor[0], stored.aceptor[2], stored.aceptor[1], wrappers))
high_sel.append(line)
cmd.delete('_nonpolar')
cmd.delete('_selection')
# compute the z_scores. Useful for protein structure validation.
stored.ResiduesNames = []
cmd.iterate('(name ca)', 'stored.ResiduesNames.append((resn))')
total_residues = float(len(stored.ResiduesNames))
z_score_wrappers = ((total_wrappers / total_residues) - 17) / 2
z_score_hb = ((len(hb) / total_residues) - 0.62) / 0.06
if len(low_sel) > 0:
cmd.show_as('dashes', DH_name)
cmd.color('red', DH_name)
low_sel.sort()
if len(mean_sel) > 0:
cmd.show_as('dashes', HBA_name)
cmd.color('yellow', HBA_name)
mean_sel.sort()
if len(high_sel) > 0:
cmd.show_as('dashes', HBO_name)
cmd.color('green', HBO_name)
high_sel.sort()
if not quiet:
hb.sort()
print "--------------------------------------------------------------------"
print "------------------------------Results ------------------------------"
print "--------------------------------------------------------------------"
print " Donor | Aceptor |"
print " Object Chain Residue | Object Chain Residue | # wrappers wrapping"
for line in low_sel:
print line[1]
for line in mean_sel:
print line[1]
for line in high_sel:
print line[1]
print '\nProtein global statistics:'
print '\nz-score wrappers = %6.2f\nz-score hydrogen bonds = %6.2f\n' % (z_score_wrappers, z_score_hb)
elif not quiet and len(hb) != 0:
print '\n - no dehydrons were found - '
print '\nz-score hydrogen bonds = %6.2f\n' % (z_score_hb)
else:
print '\n - no hydrogen bonds were found - '
def show_contacts(selection,selection2,result="contacts",cutoff=3.6, bigcutoff = 4.0, SC_DEBUG = DEBUG):
"""
USAGE
show_contacts selection, selection2, [result=contacts],[cutoff=3.6],[bigcutoff=4.0]
Show various polar contacts, the good, the bad, and the ugly.
Edit MPB 6-26-14: The distances are heavy atom distances, so I upped the default cutoff to 4.0
Returns:
True/False - if False, something went wrong
"""
if SC_DEBUG > 4:
print('Starting show_contacts')
print('selection = "' + selection + '"')
print('selection2 = "' + selection2 + '"')
result = cmd.get_legal_name(result)
#if the group of contacts already exist, delete them
cmd.delete(result)
# ensure only N and O atoms are in the selection
all_don_acc1 = selection + " and (donor or acceptor)"
all_don_acc2 = selection2 + " and (donor or acceptor)"
if SC_DEBUG > 4:
print('all_don_acc1 = "' + all_don_acc1 + '"')
print('all_don_acc2 = "' + all_don_acc2 + '"')
#if theses selections turn out not to have any atoms in them, pymol throws cryptic errors when calling the dist function like:
#'Selector-Error: Invalid selection name'
#So for each one, manually perform the selection and then pass the reference to the distance command and at the end, clean up the selections
#the return values are the count of the number of atoms
all1_sele_count = cmd.select('all_don_acc1_sele', all_don_acc1)
all2_sele_count = cmd.select('all_don_acc2_sele', all_don_acc2)
#print out some warnings
if DEBUG > 3:
if not all1_sele_count:
print('Warning: all_don_acc1 selection empty!')
if not all2_sele_count:
print('Warning: all_don_acc2 selection empty!')
########################################
allres = result + "_all"
if all1_sele_count and all2_sele_count:
cmd.distance(allres, 'all_don_acc1_sele', 'all_don_acc2_sele', bigcutoff, mode = 0)
cmd.set("dash_radius", "0.05", allres)
cmd.set("dash_color", "purple", allres)
cmd.hide("labels", allres)
########################################
#compute good polar interactions according to pymol
polres = result + "_polar"
if all1_sele_count and all2_sele_count:
cmd.distance(polres, 'all_don_acc1_sele', 'all_don_acc2_sele', cutoff, mode = 2) #hopefully this checks angles? Yes
cmd.set("dash_radius","0.126",polres)
########################################
#When running distance in mode=2, the cutoff parameter is ignored if set higher then the default of 3.6
#so set it to the passed in cutoff and change it back when you are done.
old_h_bond_cutoff_center = cmd.get('h_bond_cutoff_center') # ideal geometry
old_h_bond_cutoff_edge = cmd.get('h_bond_cutoff_edge') # minimally acceptable geometry
cmd.set('h_bond_cutoff_center', bigcutoff)
cmd.set('h_bond_cutoff_edge', bigcutoff)
#compute possibly suboptimal polar interactions using the user specified distance
pol_ok_res = result + "_polar_ok"
if all1_sele_count and all2_sele_count:
cmd.distance(pol_ok_res, 'all_don_acc1_sele', 'all_don_acc2_sele', bigcutoff, mode = 2)
cmd.set("dash_radius", "0.06", pol_ok_res)
#now reset the h_bond cutoffs
cmd.set('h_bond_cutoff_center', old_h_bond_cutoff_center)
cmd.set('h_bond_cutoff_edge', old_h_bond_cutoff_edge)
########################################
onlyacceptors1 = selection + " and (acceptor and !donor)"
onlyacceptors2 = selection2 + " and (acceptor and !donor)"
onlydonors1 = selection + " and (!acceptor and donor)"
onlydonors2 = selection2 + " and (!acceptor and donor)"
#perform the selections
onlyacceptors1_sele_count = cmd.select('onlyacceptors1_sele', onlyacceptors1)
onlyacceptors2_sele_count = cmd.select('onlyacceptors2_sele', onlyacceptors2)
onlydonors1_sele_count = cmd.select('onlydonors1_sele', onlydonors1)
onlydonors2_sele_count = cmd.select('onlydonors2_sele', onlydonors2)
#print out some warnings
if SC_DEBUG > 2:
if not onlyacceptors1_sele_count:
print('Warning: onlyacceptors1 selection empty!')
if not onlyacceptors2_sele_count:
print('Warning: onlyacceptors2 selection empty!')
if not onlydonors1_sele_count:
print('Warning: onlydonors1 selection empty!')
if not onlydonors2_sele_count:
print('Warning: onlydonors2 selection empty!')
accres = result+"_aa"
if onlyacceptors1_sele_count and onlyacceptors2_sele_count:
aa_dist_out = cmd.distance(accres, 'onlyacceptors1_sele', 'onlyacceptors2_sele', cutoff, 0)
if aa_dist_out < 0:
print('\n\nCaught a pymol selection error in acceptor-acceptor selection of show_contacts')
print('accres:', accres)
print('onlyacceptors1', onlyacceptors1)
print('onlyacceptors2', onlyacceptors2)
return False
cmd.set("dash_color","red",accres)
cmd.set("dash_radius","0.125",accres)
########################################
donres = result+"_dd"
if onlydonors1_sele_count and onlydonors2_sele_count:
dd_dist_out = cmd.distance(donres, 'onlydonors1_sele', 'onlydonors2_sele', cutoff, 0)
#try to catch the error state
if dd_dist_out < 0:
print('\n\nCaught a pymol selection error in dd selection of show_contacts')
print('donres:', donres)
print('onlydonors1', onlydonors1)
print('onlydonors2', onlydonors2)
print("cmd.distance('" + donres + "', '" + onlydonors1 + "', '" + onlydonors2 + "', " + str(cutoff) + ", 0)")
return False
cmd.set("dash_color","red",donres)
cmd.set("dash_radius","0.125",donres)
##########################################################
##### find the buried unpaired atoms of the receptor #####
##########################################################
#initialize the variable for when CALC_SASA is False
unpaired_atoms = ''
## Group
cmd.group(result,"%s %s %s %s %s %s" % (polres, allres, accres, donres, pol_ok_res, unpaired_atoms))
## Clean up the selection objects
#if the show_contacts debug level is high enough, don't delete them.
if SC_DEBUG < 5:
cmd.delete('all_don_acc1_sele')
cmd.delete('all_don_acc2_sele')
cmd.delete('onlyacceptors1_sele')
cmd.delete('onlyacceptors2_sele')
cmd.delete('onlydonors1_sele')
cmd.delete('onlydonors2_sele')
return True
def uniprot_features(uniprot_id,
selection='(all)',
withss=0,
prefix='feature_',
sm=None,
quiet=1):
'''
DESCRIPTION
Fetch feature list from uniprot.org and create named selections.
Requires residue numbering (resi) to match uniprot sequence!
ARGUMENTS
uniprot_id = string: UniProtKB name or accession
selection = string: atom selection {default: all}
withss = 0/1: update secondary structure {default: 0}
'''
import xml.etree.ElementTree as etree
try:
from urllib import urlopen
except ImportError:
from urllib.request import urlopen
withss, quiet = int(withss), int(quiet)
url = 'http://www.uniprot.org/uniprot/%s.xml' % uniprot_id
if not quiet:
print('Downloading', url)
doc = etree.parse(urlopen(url))
ns = 'http://uniprot.org/uniprot'
NS = {'u': ns}
if not quiet:
print('Parsing Features')
features = doc.findall('{%s}entry/{%s}feature' % (ns, ns))
sequence = doc.findtext('{%s}entry/{%s}sequence' % (ns, ns))
sequence = ''.join(sequence.split())
try:
if sm is None:
sm = resid_mapper.from_seq_sel(sequence, selection)
except:
print(' Warning: sequence mapping failed')
sm = lambda x: x
if withss == 1:
cmd.alter(selection, 'ss="L"')
ssmap = {'helix': 'H', 'strand': 'S', 'turn': 'L'}
norange_types = ['disulfide bond']
count = 0
for feature in features:
type_ = feature.get('type')
begin = feature.find('{%s}location/{%s}begin' % (ns, ns))
if begin is not None:
end = feature.find('{%s}location/{%s}end' % (ns, ns))
values = begin.get('position'), end.get('position')
if type_ in norange_types:
try:
values = sm(values[0]), sm(values[1])
sel = '(' + selection + ') and resi %s+%s' % values
except KeyError:
print(' Warning: could not map', values)
sel = 'none'
else:
try:
values = sm(values)
sel = '(' + selection + ') and resi %s-%s' % values
except KeyError:
print(' Warning: could not map', values)
sel = 'none'
else:
position = feature.find('{%s}location/{%s}position' % (ns, ns))
try:
value = sm(position.get('position'))
sel = '(%s) and resi %s' % (selection, value)
except KeyError:
sel = 'none'
if type_ in ['helix', 'strand', 'turn'] and withss < 2:
if withss == 1:
cmd.alter(sel, 'ss="%s"' % ssmap.get(type_, 'L'))
else:
count += 1
name = cmd.get_legal_name(
'%s%03d_%s' % (prefix, count, feature.get(
'description', '').replace('.', '')))
groupname = cmd.get_legal_name('%s%s' % (prefix, type_))
cmd.select(name, sel)
cmd.group(groupname, name, 'add')
if not quiet:
print('Found %d feature records (without secondary structures)' %
count)
def read_maestr(maestr,name,state=0,finish=1,discrete=-1,
quiet=1,zoom=-1,multiplex=-1,mimic=1,
object_props=None, atom_props=None, _self=cmd):
import sys
if sys.version_info[0] > 2 and isinstance(maestr, bytes):
maestr = maestr.decode()
cmd = _self
mimic = int(mimic)
if not quiet:
print(" Load: Processing Schrodinger .MAE file...")
if object_props is None:
object_props = _self.get('load_object_props_default')
if atom_props is None:
atom_props = _self.get('load_atom_props_default')
mr = MAEReader(mimic, object_props, atom_props)
list = mr.listFromStr(maestr)
if list and getattr(list[0], 'ct_type', None) == 'full_system':
if not quiet:
print(" Desmond CMS file detected, only loading first CT")
list = list[:1]
len_list = len(list);
if len_list<2: # MAE contains no more than one molecule
if discrete<0:
discrete = 0
else: # MAE contains multiple molecules
if multiplex > 0:
if len_list > mr.full_count:
multiplex = 2 # multiplex at the record level (break out conformers, etc.)
if multiplex < 0:
if len_list == mr.full_count: # file only contains full records
multiplex = 1 # then multiplex at the object level
elif multiplex == 0:
if mr.full_count>1: # multiple full records present
if discrete<0:
discrete = 1 # use discrete object to prevent collisions
if discrete < 0:
discrete = 0
preserve_chempy_ids=cmd.get('preserve_chempy_ids')
cmd.set('preserve_chempy_ids')
# nest p_m_ct behind f_m_ct
nested_list = []
for mdl in list:
mdl.molecule.title = getattr(mdl, 'title', None) or getattr(mdl, 'name', '')
if (mdl.mae_record == 'f_m_ct') or not nested_list:
nested_list.append( [mdl] )
else:
nested_list[-1].append(mdl)
if multiplex < 0:
multiplex = 1 if (len(nested_list) > 1) else 0
if multiplex:
cmd.group(name)
unnamed_cnt = itertools.count(1)
object_name_dict = {}
nested_list.reverse()
while nested_list:
mdl_list = nested_list.pop()
len_mdl_list = len(mdl_list)
if len_mdl_list:
mdl = mdl_list[0]
mdl_name = name
if multiplex:
mdl_name += "." + (mdl.molecule.title or str(next(unnamed_cnt)))
mdl_name = cmd.get_legal_name(mdl_name)
if mdl_name in object_name_dict and multiplex == 1:
mdl_name = cmd.get_unused_name(mdl_name + "_")
if multiplex > 0:
name_dict = {}
if multiplex > 1: # break coordinate sets into separate objects
cnt = 0
for mdl in mdl_list:
cnt = cnt + 1
mr.fix_bond_reps(mdl)
state_name = mdl_name
if len_mdl_list>1:
state_name += "_" + str(cnt)
_self.load_model(mdl,state_name,state,zoom=zoom,
discrete=discrete,quiet=quiet, _self=_self)
object_name_dict[state_name] = 1
else: # only break objects
for mdl in mdl_list:
mr.fix_bond_reps(mdl)
_self.load_model(mdl,mdl_name,state,zoom=zoom,
discrete=discrete,quiet=quiet, _self=_self)
object_name_dict[mdl_name] = 1
else: # multiplex = 0 , so stuff everything into a single object
for mdl in mdl_list:
mr.fix_bond_reps(mdl)
_self.load_model(mdl,mdl_name,state,zoom=zoom,
discrete=discrete,quiet=quiet, _self=_self)
object_name_dict[mdl_name] = 1
cmd.set('preserve_chempy_ids',preserve_chempy_ids)
if mimic:
for name in object_name_dict.keys():
# maestro-like settings
_self.flag("ignore",name+" and not polymer","set")
_self.set("stick_radius",0.18,name)
_self.set("line_width",1.2,name)
_self.set("valence_size",0.075,name)
_self.set("valence", 1, name)
_self.set("light",[0.4,-0.4,-1.0])
_self.set("shininess",100)
_self.set("light2",[-0.5,-0.7,-0.2])
_self.set("light_count",3)
_self.set("label_position",[1,0,2])
_self.set("label_font_id",7)
def load_vina_results(project_file, group, max_load, max_rank, interactions_check):
# Load project data
with open(project_file) as _project_file:
project_data = json.load(_project_file)
# Load target
target_name = f"{group}.target"
if project_data["flexible"]:
cmd.load(project_data["rigid_pdbqt"], target_name)
else:
cmd.load(project_data["target_pdbqt"], target_name)
cmd.group(group)
cmd.group(group, target_name)
# Show box
box_name = f"{group}.box"
display_box(
box_name,
(
project_data["center_x"] + project_data["size_x"] / 2,
project_data["center_y"] + project_data["size_y"] / 2,
project_data["center_z"] + project_data["size_z"] / 2,
),
(
project_data["center_x"] - project_data["size_x"] / 2,
project_data["center_y"] - project_data["size_y"] / 2,
project_data["center_z"] - project_data["size_z"] / 2,
),
)
cmd.group(group, box_name)
# Parse results
results_dir = project_data["results_dir"]
results = itertools.chain.from_iterable(
map(parse_vina_log, glob(f"{results_dir}/poses/*.pdbqt"))
)
results = sorted(results, key=itemgetter("affinity"))
cache = set()
objects = set()
count = 0
for pose in results:
# Ignore poses which mode is greater than max
if pose["mode"] > max_rank:
continue
# Load molecule into cache
cache_name = cmd.get_legal_name(pose["filename"].replace(".", "_"))
if cache_name not in cache:
cmd.load(pose["filename"], cache_name)
cache.add(cache_name)
# Compute object names
score = int(-10 * pose["affinity"])
state = pose["mode"]
base_name = f'{group}.{pose["name"]}_{pose["mode"]}_{score}'
obj_name = f"{base_name}.mol"
polar_name = f"{base_name}.polar"
# Create group
cmd.group(base_name)
# Create molecule object
cmd.create(obj_name, cache_name, state, 1)
cmd.group(base_name, obj_name)
if interactions_check:
cmd.distance(polar_name, target_name, obj_name, 2)
cmd.group(base_name, polar_name)
objects.add(obj_name)
count += 1
if count >= max_load:
break
cmd.delete("delete " + " ".join(cache))
