def mutate_all(selection, new_resn, inplace=1, sculpt=0, *args, **kwargs):
'''
DESCRIPTION
Mutate all residues in selection. By default do mutation in-place (unlike
the 'mutate' command which by default works on a copy).
FOR SCULPTING ONLY SUPPORTS SELECTIONS WITHIN THE SAME MODEL!
SEE ALSO
mutate
'''
inplace, sculpt = int(inplace), int(sculpt)
if sculpt and len(cmd.get_object_list('(' + selection + ')')) > 1:
print(' Error: Sculpting in multiple models not supported')
raise CmdException
kwargs.pop('_self', None)
sele_list = set()
cmd.iterate(selection,
'sele_list.add("/%s/%s/%s/%s" % (model, segi, chain, resi))',
space={'sele_list': sele_list})
for sele in sele_list:
mutate(sele, new_resn, inplace, sculpt and not inplace, *args, **kwargs)
if sculpt and inplace:
sculpt_relax('(' + ' '.join(sele_list) + ')', 0, sculpt == 2)
def get_dehydrons():
cmd.delete('dehydrons')
cmd.delete('DH_pairs')
cmd.hide()
angle = float(angle_value.get())
cutoff = float(dist_cutoff_value.get())
desolv = float(desolv_sphere.get())
min_wrappers = float(min_value.get())
selection = 'name n or name o and not resn hoh'
hb = cmd.find_pairs("((byres "+selection+") and n. n)","((byres "+selection+") and n. o)",mode=1,cutoff=cutoff,angle=angle)
# sort the list for easier reading
hb.sort(lambda x,y:(cmp(x[0][1],y[0][1])))
print "------------------------------------------------\n----------------dehydron Results-----------------\n------------------------------------------------\n Donor | Aceptor | \nChain Residue | Chain Residue | # dehydrons"
sel = []
wra = 0
for pairs in hb:
cmd.iterate("%s and index %s" % (pairs[0][0], pairs[0][1]), 'stored.nitro = chain, resi, resn') # extracts the nitrogen
cmd.iterate("%s and index %s" % (pairs[1][0], pairs[1][1]), 'stored.oxy = chain, resi, resn') # extracts the oxygen
wrappers = cmd.select('wrap', '(((chain %s and name ca and resi %s) around %f) or ((chain %s and name ca and resi %s) around %f)) and (not ((neighbor name n*+o*) or (name n*+o*+h*)))' % (stored.nitro[0], stored.nitro[1], desolv, stored.oxy[0], stored.oxy[1], desolv)) #Identifies the putative dehydrons
if wrappers < min_wrappers:
wra = 1
cmd.distance('Dehydrons',"%s and index %s" % (pairs[0][0],pairs[0][1]),"%s and index %s" % (pairs[1][0],pairs[1][1]))
print ' %s%7s%5d | %s%7s%5d |%7s' % (stored.nitro[0], stored.nitro[2], int(stored.nitro[1]), stored.oxy[0], stored.oxy[2], int(stored.oxy[1]), wrappers)
if stored.nitro[1] not in sel:
sel.append(stored.nitro[1])
if stored.oxy[1] not in sel:
sel.append(stored.oxy[1])
if wra == 1:
cmd.select('DH_pairs', 'resi %s' % ('+').join(sel))
cmd.show('lines', 'DH_pairs')
cmd.disable('DH_pairs')
cmd.hide('labels')
cmd.delete('wrap')
cmd.show('cartoon')
cmd.show('dashes')
def test_cbss(self):
cmd.load(self.datafile('1oky-frag.pdb'))
c = [2, 13, 22] # blue orange forest
pymol.util.cbss('*', *c)
stored.colors = set()
cmd.iterate('*', 'stored.colors.add((ss or "L", color))')
self.assertEqual(stored.colors, set(zip('HSL', c)))
def test_colors(self):
cmd.fragment('gly')
pymol.util.colors("jmol")
stored.colors = set()
cmd.iterate('elem C', 'stored.colors.add(color)')
colors = [get_color_tuple(c, 3) for c in stored.colors]
self.assertEqual(colors, [(0.567, 0.567, 0.567)])
def fade_color( sele = "all", fade_extent = 0.7, by_chain = True ):
"""
Fade out RGB color values associated with a selection.
The by_chain variable assumes that chains are uniform color, to
allow speedy color setting based on chain, rather than based on
residue (which can take a long time!)
"""
pymol.stored.colors = []
cmd.iterate( sele, "stored.colors.append( (chain, resi, name, color))")
res_colors = {}
stored_colors = pymol.stored.colors;
cols = []
colorname = ''
chain = ''
prev_chain = ''
for chain, resi, name, color in stored_colors:
if name == 'CA' or name == 'P': # c-alpha atom
if by_chain and chain != prev_chain and len( colorname ) > 0:
cmd.color( colorname, 'chain %s and %s' % (prev_chain,sele) )
color_tuple = cmd.get_color_tuple(color)
cols = [];
for i in range(3):
cols.append( fade_extent + ( 1.0 - fade_extent ) * float(color_tuple[ i ]) )
colorname = 'temp_chain%s' % chain
cmd.set_color( colorname, cols )
if not by_chain: cmd.color( colorname, 'resi %s and chain %s and %s' % (resi,chain,sele) )
prev_chain = chain
if by_chain: cmd.color( colorname, 'chain %s and %s' % (chain,sele) )
def ss(selection):
class SSE(object):
def __init__(self, start, typ):
self.start, self.typ = start, typ
self.end = -1
def __repr__(self):
return "%s-%s %s" % (self.start, self.end, self.typ)
stored.pairs = []
cmd.iterate("%s and n. ca" % selection, "stored.pairs.append((resi, ss))")
num, currentType = stored.pairs[0]
sses = [SSE(num, currentType)]
currentSSE = sses[0]
for resi, ssType in stored.pairs:
if ssType == currentType:
currentSSE.end = resi
else:
sses.append(SSE(resi, ssType))
currentSSE = sses[-1]
currentType = ssType
for sse in sses:
print sse
def cartoonize(color, rep):
"""draw a cartoon representation of glycans"""
stored.ResiduesNumber = []
cmd.iterate('name c1', 'stored.ResiduesNumber.append((resi))')
resn_list = [int(i) for i in stored.ResiduesNumber]
bonds = get_glyco_bonds(resn_list[0], resn_list[-1]+1)
con_matrix = writer(bonds)
#con_matrix = writer2(bonds)
rings = find_rings(resn_list)
rings_coords = get_ring_coords(resn_list, rings)
bonds_coords = get_bonds_coords(resn_list, con_matrix)
colors = get_colors_c1(resn_list, color)
bonds_colors = get_bonds_colors(resn_list, con_matrix, color)
cmd.set('suspend_updates', 'on')
for state, coords in enumerate(rings_coords):
obj = []
if rep == 'beads':
radius_s = 1.8
radius_b = 0.18
obj = beads(obj, coords, colors[state], radius_s)
obj = cylinder(obj, bonds_coords[state], bonds_colors[state], radius_b)
else:
if rep == 'cartoon':
radius = 0.075
else:
radius = 0.035
obj = hexagon(obj, coords, colors[state], rep, radius)
obj = cylinder(obj, bonds_coords[state], bonds_colors[state], radius)
cmd.load_cgo(obj,'cgo01', state+1)
cmd.select('glycan', 'byres name C1')
cmd.delete('glycan')
cmd.delete('tmp')
cmd.set('two_sided_lighting', 1)
cmd.set('suspend_updates', 'off')
def defineShiftRange(displayed, cs_fn, f3width, f3left):
""" Define shift range of selected atoms
@param displayed: hydrogen atoms in the center selection or covalently
bonded to any atoms in the center selection
"""
# define shift range of selected atoms
maxf3, minf3 = -10000.0, 10000.0
# get the chemshifts of the atoms in the selection sphere
# make a copy of the chemshifts file first
shutil.copyfile(cs_fn, cs_fn+'.bak')
cs_dict = readChemshifts(cs_fn)
cs_list = []
stored.list = []
cmd.iterate(displayed, "stored.list.append((resi,resn,name))")
for resi,resn,name in stored.list:
if name in ['HN','HT1','HT2','HT3']: name = 'H' # rename proton collecting to bb N
k = "%5s %s %4s" % (resi, resn, name)
try:
cs_list.append(cs_dict[k])
csv = float(cs_dict[k][18:26])
if csv > maxf3: maxf3 = csv
if csv < minf3: minf3 = csv
except KeyError:
#pass # TODO: check the output.txt after back-calc
print "INFO: Could not find chemshift for", k
#for cs in cs_list: print cs # debug print
f3shift = maxf3 + 0.2 # left edge of the spectrum
f3sweep = maxf3 - minf3 + 0.4 # direct 1H sweep width
if f3width is not None: f3sweep = f3width
if f3left is not None: f3shift = f3left
return f3sweep, f3shift
def write_seq_positions(groupdict, selection, csvfile, object_name = 'molecule'):
for name in groupdict:
# Get the selection:
seq_positions = get_seq_positions(name + '.' + object_name,
name + '.' + selection)
# The first position of each line holds the pdbid
to_be_written = [name]
# The second position holds the sequence of the selection
# Programs loading these selections will see if they can replicate
# the sequence. If they can't, they may be using different indices or
# something else bad.
stored.sequence = ''
cmd.iterate(name + '.' + object_name + ' & polymer & n. CA',
'stored.sequence += one_letter[resn]')
filtered_sequence = ''
for n in seq_positions:
filtered_sequence += stored.sequence[n]
to_be_written.append(filtered_sequence)
# All the positions after the second hold indices.
to_be_written += seq_positions
csvfile.writerow(to_be_written)
def collapse_resi(selection='(sele)', quiet=1):
'''
DESCRIPTION
Returns a compact selection macro for the given selection on
residue number (resi) level.
Rewrite of http://pymolwiki.org/index.php/CollapseSel
ARGUMENTS
selection = string: atom selection {default: (sele)}
'''
from collections import defaultdict
s_dict = defaultdict(set)
cmd.iterate(selection, 's_dict[model,segi,chain].add(resv)', space=locals())
r_all = []
for key, s in s_dict.items():
s = sorted(s)
r = [[s[0], s[0]]]
for i in s[1:]:
if i <= r[-1][1] + 1:
r[-1][1] = i
else:
r.append([i,i])
resi = '+'.join(('%d-%d' % (f,t) if f != t else '%d' % (f)) for (f,t) in r)
r_all.append('/%s/%s/%s/' % key + resi)
if not int(quiet):
for r in r_all:
print(' collapse_resi: ' + str(r))
return ' '.join(r_all)
def measure_all_torsions(sele,prot_dict):
id_list = cmd.identify("((%s) and name ca)"%sele)
for a in id_list:
res_sele = "(alt '' and (byres id %d) and (%s))"%(a,sele)
lst = find_torsions(res_sele)
if len(lst):
cmd.iterate("(%s and n;ca)"%res_sele,"stored.resn=resn")
resn = pymol.stored.resn
print resn
name_list = []
for a in lst:
pymol.stored.names = []
for b in a:
cmd.iterate("(%s and id %d)"%(res_sele,b),"stored.names.append(name)")
name_list.append(tuple(pymol.stored.names))
res_dict = {}
for a in name_list:
dihe = cmd.get_dihedral("(%s and name %s)"%(res_sele,a[0]),
"(%s and name %s)"%(res_sele,a[1]),
"(%s and name %s)"%(res_sele,a[2]),
"(%s and name %s)"%(res_sele,a[3]))
res_dict[a]=dihe
if not prot_dict.has_key(resn):
prot_dict[resn]=[]
prot_dict[resn].append(res_dict)
def testMAEchempy(self, molfilename):
cmd.load(self.datafile(molfilename), 'test', object_props='*', atom_props='*')
objs = cmd.get_object_list()
for obj in objs:
idxToVal = {}
natoms = cmd.count_atoms(obj)
for i in range(natoms):
idxToVal[i+1] = i*10
cmd.set_atom_property('test_prop', i*10, "index %d and %s" % (i+1, obj))
model = cmd.get_model(obj)
# test to make sure the properties that exist are the same
prop_list= cmd.get_property_list(obj)
mol_prop_list = [x[0] for x in model.molecule_properties]
self.assertEqual(set(prop_list), set(mol_prop_list))
# need to test whether the values are the same
for prop, val in model.molecule_properties:
propval = cmd.get_property(prop, obj)
self.assertEqual(propval, val)
self.assertEqual(natoms, len(model.atom))
# need to test values of all atom properties, including the ones that were set above
stored.prop_lookup = {}
cmd.iterate(obj, "stored.prop_lookup[index-1] = properties.all")
idx = 0
for at in model.atom:
for prop in at.atom_properties.keys():
val = at.atom_properties[prop]
self.assertEqual(val, stored.prop_lookup[idx][prop])
idx += 1
def ss(selection):
"""A command to list a summary of the secondary structure for a selection.
Use like "ss my_protein" where "my_protein" is the name of the chain or
structure in view.
"""
class SSE(object):
def __init__(self, start, typ):
self.start, self.typ = start, typ
self.end = -1
def __repr__(self):
return "%s-%s %s" % (self.start, self.end, self.typ)
stored.pairs = []
cmd.iterate("%s and n. ca" % selection, "stored.pairs.append((resi, ss))")
num, currentType = stored.pairs[0]
sses = [SSE(num, currentType)]
currentSSE = sses[0]
for resi, ssType in stored.pairs:
if ssType == currentType:
currentSSE.end = resi
else:
sses.append(SSE(resi, ssType))
currentSSE = sses[-1]
currentType = ssType
for sse in sses:
print sse
def preProcessPDB(fname):
stored.residues = []
stored.chains = []
cmd.iterate('(name ca)', 'stored.residues.append(resi)')
cmd.iterate('(name ca)', 'stored.chains.append(chain)')
cmd.show_as("cartoon", '(all)')
cmd.color("white", '(all)')
def sele2Color(self, sele):
globalShader = self.optionMenu_shader.getvalue()
newShader = self.seleShaderDict[sele]
if newShader != globalShader: # should be different from global shader, otherwise do nothing
newColorInc = ShaderFactory.seleSlot[newShader]
stored.idcolor_list = set()
cmd.iterate(sele, 'stored.idcolor_list.add(int(color))')
if(len(stored.idcolor_list)>1):
self.selectionConsole.set('Warning: Selection [%s] contains more than one color.' % sele)
color = stored.idcolor_list.pop()
rgb_color = cmd.get_color_tuple(color)
if rgb_color[2] >= 0.990:
color_id = '%s %s %s' % (str(rgb_color[0])[0:5].ljust(5, '0'), str(rgb_color[1])[0:5].ljust(5, '0'), str(rgb_color[2]-newColorInc)[0:5])
else:
color_id = '%s %s %s' % (str(rgb_color[0])[0:5].ljust(5, '0'), str(rgb_color[1])[0:5].ljust(5, '0'), str(rgb_color[2]+newColorInc)[0:5])
# color_id from all the sele:shader dictionary
# determine color_id:shader pair
if color_id not in self.spColorShaderDict:
self.spColorShaderDict[color_id] = [newShader, sele]
# apply new color to atom set
newRGB = color_id.split(' ')
newColor = [float(newRGB[0]), float(newRGB[1]), float(newRGB[2])]
cmd.set_color(color_id, newColor)
cmd.color(color_id, sele)
print 'apply shader [%s] to selection [%s].' % (newShader, sele)
def _testLabelByAtomType(self):
cmd.load(self.datafile('1oky.pdb.gz'))
cmd.remove("alt 'B'")
cmd.label('resn STU', 'text_type')
c3labels = []
cmd.iterate("resn STU", "c3labels.append(label == 'C.3')", space=locals())
self.assertEquals(sum(c3labels), 8)
def get_b_limits(input='[0,100]', selection='all'):
from pymol import cmd, stored
import math
try:
limits=str(input)
if limits.startswith('['):
# list
limits=eval(limits)[:2]
else:
#integer
selection=str(selection)
limits=abs(float(limits))
if limits>50: return False
stored.temp=[]
cmd.iterate(selection, 'stored.temp.append(b)')
stored.temp.sort()
kl=(len(stored.temp)-1)*limits/100
fl=math.floor(kl)
cl=math.ceil(kl)
kr=(len(stored.temp)-1)*(100-limits)/100
fr=math.floor(kr)
cr=math.ceil(kr)
limits=[0,0]
if fl==cl:
limits[0]=stored.temp[int(kl)]
else:
limits[0]=stored.temp[int(kl)]*(cl-kl)+stored.temp[int(kl)]*(kl-fl)
if fr==cr:
limits[1]=stored.temp[int(kr)]
else:
limits[1]=stored.temp[int(kr)]*(cr-kr)+stored.temp[int(kr)]*(kr-fr)
except:
return False
return limits
def testMMTF(self):
'''Styled MMTF export/import'''
S = 0b10 # 1 << 1 spheres
D = 0b1000000000 # 1 << 9 dots
B = 2 # blue
R = 4 # red
cmd.fragment('gly')
cmd.color(B)
cmd.color(R, 'elem C')
cmd.show_as('spheres')
cmd.show_as('dots', 'elem C')
with testing.mktemp('.mmtf') as filename:
cmd.save(filename)
cmd.delete('*')
cmd.load(filename)
color_list = []
reps_list = []
cmd.iterate('*', 'color_list.append(color)', space=locals())
cmd.iterate('*', 'reps_list.append(reps)', space=locals())
self.assertEqual(color_list, [B, R, R, B, B, B, B])
self.assertEqual(reps_list, [S, D, D, S, S, S, S])
def testExportStyle(self):
cmd.fab('ACDEF', 'm1')
cmd.hide()
cmd.show('cartoon', 'resi 1-3')
cmd.show('lines', 'resn CYS')
cmd.show('sticks', 'resn ASP+PHE')
cmd.show('spheres', 'resn GLU')
cmd.set('stick_ball', 1, 'resn PHE')
cmd.set('stick_ball_ratio', 1.5, 'm1')
testlabel = 'Hello "World"'
cmd.label('name SG', repr(testlabel))
with testing.mktemp('.mae') as filename:
cmd.save(filename)
cmd.delete('*')
cmd.load(filename, 'm2')
g_labels = []
cmd.iterate('name SG', 'g_labels.append(label)', space=locals())
cmd.alter('*', 'b = 1 if s.stick_ball else 0')
self._assertCountEqual('rep cartoon & guide', 'resi 1-3 & guide')
self._assertCountEqual('rep lines', 'resn CYS', delta=1)
self._assertCountEqual('rep sticks', 'resn ASP+PHE')
self._assertCountEqual('rep spheres', 'resn GLU')
self._assertCountEqual('b > 0.5', 'resn PHE')
self.assertTrue(cmd.get_setting_float('stick_ball_ratio', 'm2') > 1.1)
self.assertEqual(g_labels[0], testlabel)
def iterate_plot(selection, expr_y, expr_x=None, scatter=0, filename=None,
space=None, quiet=1):
'''
DESCRIPTION
Plot atomic properties.
ARGUMENTS
selection = string: atom selection
expr_y = string: python expression for y values
expr_x = string: python expression for x values {default: None}
scatter = 0/1: make line plot or scatter plot {default: 0, line plot}
EXAMPLE
# C-alpha b-factors
iterate_plot name CA, b, resv
'''
from . import matplotlib_fix
from matplotlib.pyplot import figure
scatter, quiet = int(scatter), int(quiet)
if space is None:
space = {'cmd': cmd, 'stored': cmd.pymol.stored}
if cmd.is_string(selection):
if selection.startswith('['):
sele_list = selection[1:-1].split(',')
else:
sele_list = ['(%s) and (%s)' % (model, selection) for model in
cmd.get_object_list('(' + selection + ')')]
else:
sele_list = selection
fig = figure()
plt = fig.add_subplot(111)
for selection in sele_list:
space['_values'] = y_values = []
cmd.iterate(selection, '_values.append(' + expr_y + ')', space=space)
if expr_x is None:
x_values = range(len(y_values))
else:
space['_values'] = x_values = []
cmd.iterate(selection, '_values.append(' + expr_x + ')', space=space)
color = get_model_color(selection)
if scatter:
plt.scatter(x_values, y_values, c=color)
else:
plt.plot(x_values, y_values, c=color)
_showfigure(fig, filename, quiet)
def findSurfaceResidues(objSel="(all)", cutoff=2.5, doShow=False, verbose=True):
"""
findSurfaceResidues
finds those residues on the surface of a protein
that have at least 'cutoff' exposed A**2 surface area.
PARAMS
objSel (string)
the object or selection in which to find
exposed residues
DEFAULT: (all)
cutoff (float)
your cutoff of what is exposed or not.
DEFAULT: 2.5 Ang**2
asSel (boolean)
make a selection out of the residues found
RETURNS
(list: (chain, resv ) )
A Python list of residue numbers corresponding
to those residues w/more exposure than the cutoff.
"""
tmpObj="__tmp"
cmd.create( tmpObj, objSel + " and polymer");
if verbose!=False:
print "WARNING: I'm setting dot_solvent. You may not care for this."
cmd.set("dot_solvent");
cmd.get_area(selection=tmpObj, load_b=1)
# threshold on what one considers an "exposed" atom (in A**2):
cmd.remove( tmpObj + " and b < " + str(cutoff) )
stored.tmp_dict = {}
cmd.iterate(tmpObj, "stored.tmp_dict[(chain,resv)]=1")
exposed = stored.tmp_dict.keys()
exposed.sort()
randstr = str(random.randint(0,10000))
selName = "exposed_atm_" + randstr
if verbose!=False:
print "Exposed residues are selected in: " + selName
cmd.select(selName, objSel + " in " + tmpObj )
selNameRes = "exposed_res_" + randstr
cmd.select(selNameRes, "byres " + selName )
if doShow!=False:
cmd.show_as("spheres", objSel + " and poly")
cmd.color("white", objSel)
cmd.color("red", selName)
cmd.delete(tmpObj)
print exposed
return exposed
def testAdvanced(self):
cmd.fab('A/123/ ADC B/234/ AFCD')
v = cmd.get_chains()
self.assertEqual(v, ['A', 'B'])
cmd.iterate('last chain B', 'stored.v = (resv, resn)')
self.assertEqual(stored.v, (237, 'ASP'))
def _test_cba(self, fun, expected_color):
cmd.fragment('gly')
cmd.color('white')
fun('*')
stored.colors = set()
cmd.iterate('elem C', 'stored.colors.add(color)')
colors = [get_color_tuple(c, 3) for c in stored.colors]
self.assertEqual(colors, [expected_color])
def start(self,sel):
self.lock = 1
cmd.iterate('(%s) and name CA' % sel,'idx2resn[model,index] = (resn, color, ss)',
space={'idx2resn': self.canvas.idx2resn})
for model_index, (phi,psi) in cmd.get_phipsi(sel, self.state).iteritems():
print " Plotting Phi,Psi: %8.2f,%8.2f" % (phi, psi)
self.canvas.plot(phi, psi, model_index)
self.lock = 0
def colorByProp(self, prop, palette="rainbow"):
stored.propUniqVals = set()
cmd.iterate(self.objName, "stored.propUniqVals.add(%s)" % prop)
v = sorted(stored.propUniqVals)
b = n.arange(1, len(v) + 1, dtype=float) # / len(v)
stored.b = dict(zip(v, b))
cmd.alter(self.objName, "b=stored.b[%s]" % prop)
cmd.spectrum("b", palette, self.objName)
def test_stub2ala():
cmd.reinitialize()
cmd.fab('EFG', 'm1')
cmd.remove('not (backbone or name CB)')
psico.editing.stub2ala()
my_values = []
cmd.iterate('guide', 'my_values.append(resn)', space=locals())
assert my_values == ['ALA', 'ALA', 'GLY']
def renumber(selection='all', start=1, startsele=None, quiet=1):
'''
DESCRIPTION
Set residue numbering (resi) based on connectivity.
ARGUMENTS
selection = string: atom selection to renumber {default: all}
start = integer: counting start {default: 1}
startsele = string: residue to start counting from {default: first in
selection}
'''
start, quiet = int(start), int(quiet)
model = cmd.get_model(selection)
cmd.iterate(selection, 'atom_it.next().model = model',
space={'atom_it': iter(model.atom)})
if startsele is not None:
startidx = cmd.index('first (' + startsele + ')')[0]
for atom in model.atom:
if (atom.model, atom.index) == startidx:
startatom = atom
break
else:
print(' Error: startsele not in selection')
raise CmdException
else:
startatom = model.atom[0]
for atom in model.atom:
atom.adjacent = []
atom.visited = False
for bond in model.bond:
atoms = [model.atom[i] for i in bond.index]
atoms[0].adjacent.append(atoms[1])
atoms[1].adjacent.append(atoms[0])
minmax = [start, start]
def traverse(atom, resi):
atom.resi = resi
atom.visited = True
for other in atom.adjacent:
if other.visited:
continue
if (atom.name, other.name) in [('C', 'N'), ("O3'", 'P')]:
minmax[1] = resi + 1
traverse(other, resi + 1)
elif (atom.name, other.name) in [('N', 'C'), ('P', "O3'")]:
minmax[0] = resi - 1
traverse(other, resi - 1)
elif (atom.name, other.name) not in [('SG', 'SG')]:
traverse(other, resi)
traverse(startatom, start)
cmd.alter(selection, 'resi = atom_it.next().resi',
space={'atom_it': iter(model.atom)})
if not quiet:
print(' Renumber: range (%d to %d)' % tuple(minmax))
def spectrumproperty(propname, color_list, selection='(all)', minimum=None, maximum=None, quiet=1):
'''
DESCRIPTION
Define a color spectrum with as many color-stops as you like (at least 2).
ARGUMENTS
propname = string: property name which has a float value.
color_list = string: Space separated list of colors
... all other arguments like with `spectrum` command
'''
quiet = int(quiet)
colors = color_list.split()
if len(colors) < 2:
print 'failed! please provide at least 2 colors'
return
colvec = [cmd.get_color_tuple(i) for i in colors]
parts = len(colvec) - 1
value_list = []
cmd.iterate(selection, 'value_list.append(properties[propname])', space=locals())
if len(value_list) == 0:
print 'empty selection'
return
if minimum is None:
minimum = min(value_list)
if maximum is None:
maximum = max(value_list)
minimum, maximum = float(minimum), float(maximum)
val_range = (maximum - minimum) * (1 + 1e-6)
if not quiet:
print ' Spectrum: range (%.5f to %.5f)' % (minimum, maximum)
if maximum == minimum:
print 'no spectrum possible, only equal values'
return
rgb = lambda i, p, j: int(255 * (colvec[i+1][j] * p + colvec[i][j] * (1.0 - p)))
col_list = []
for value in value_list:
p = (value - minimum) / val_range * parts
i = int(p)
p -= i
col_list.append(0x40000000 +
rgb(i, p, 0) * 0x10000 +
rgb(i, p, 1) * 0x100 +
rgb(i, p, 2))
cmd.alter(selection, 'color = col_iter.next()', space={'col_iter': iter(col_list)})
cmd.recolor()
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 getAllAtoms(sele):
'''Get the names of all the atoms in the selection.
'''
myspace = {'names':[]}
cmd.iterate(sele, "names.append(name)", space=myspace)
return myspace['names']
from pymol import cmd, stored
stored.list = []
cmd.iterate("(wachin)", "stored.list.append(index)")
print stored.list
def add_missing_atoms(selection='all', cycles=200, quiet=1):
'''
DESCRIPTION
Mutate those residues to themselves which have missing atoms
SEE ALSO
stub2ala
'''
from collections import defaultdict
from chempy import fragments
cycles, quiet = int(cycles), int(quiet)
reference = {
'ALA':
set(['CB']),
'ARG':
set(['CB', 'CG', 'NE', 'CZ', 'NH1', 'NH2', 'CD']),
'ASN':
set(['CB', 'CG', 'OD1', 'ND2']),
'ASP':
set(['CB', 'CG', 'OD1', 'OD2']),
'CYS':
set(['CB', 'SG']),
'GLN':
set(['CB', 'CG', 'CD', 'NE2', 'OE1']),
'GLU':
set(['CB', 'CG', 'OE2', 'CD', 'OE1']),
'GLY':
set([]),
'HIS':
set(['CE1', 'CB', 'CG', 'CD2', 'ND1', 'NE2']),
'ILE':
set(['CB', 'CD1', 'CG1', 'CG2']),
'LEU':
set(['CB', 'CG', 'CD1', 'CD2']),
'LYS':
set(['CB', 'CG', 'NZ', 'CE', 'CD']),
'MET':
set(['CB', 'CG', 'CE', 'SD']),
'PHE':
set(['CE1', 'CB', 'CG', 'CZ', 'CD1', 'CD2', 'CE2']),
'PRO':
set(['CB', 'CG', 'CD']),
'SER':
set(['OG', 'CB']),
'THR':
set(['CB', 'OG1', 'CG2']),
'TRP':
set([
'CZ2', 'CB', 'CG', 'CH2', 'CE3', 'CD1', 'CD2', 'CZ3', 'NE1', 'CE2'
]),
'TYR':
set(['CE1', 'OH', 'CB', 'CG', 'CZ', 'CD1', 'CD2', 'CE2']),
'VAL':
set(['CB', 'CG1', 'CG2']),
}
namedsele = cmd.get_unused_name('_')
cmd.select(namedsele, selection, 0)
namelists = defaultdict(list)
cmd.iterate('(%s) and polymer' % namedsele,
'namelists[model,segi,chain,resn,resi,resv].append(name)',
space=locals())
sele_dict = defaultdict(list)
tmp_name = cmd.get_unused_name('_')
for key, namelist in namelists.items():
resn = key[3]
if resn not in reference:
if not quiet:
print(' Unknown residue: + ' + resn)
continue
if not reference[resn].issubset(namelist):
try:
frag = fragments.get(resn.lower())
for a in frag.atom:
a.segi = key[1]
a.chain = key[2]
a.resi = key[4]
a.resi_number = key[5]
cmd.load_model(frag, tmp_name, 1, zoom=0)
skey = '/%s/%s/%s/%s`%s' % key[:5]
cmd.remove(skey + ' and not name N+C+O+OXT+CA')
cmd.align(tmp_name, skey + ' and name N+C+CA', cycles=0)
cmd.remove(tmp_name + ' and (name N+C+O+CA or hydro)')
cmd.fuse('name CB and ' + tmp_name,
'name CA and ' + skey,
move=0)
if resn == 'PRO':
cmd.bond(skey + '/N', skey + '/CD')
cmd.unpick()
cmd.delete(tmp_name)
sele_dict[key[0]].append(skey)
if not quiet:
print(' Mutated ' + skey)
except:
print(' Mutating ' + skey + ' failed')
for model in sele_dict:
cmd.sort(model)
sculpt_relax(' '.join(sele_dict[model]), 0, 0, model, cycles)
cmd.delete(namedsele)
def rand_sequence(selector):
"""
DESCRIPTION
The "rand_sequence" takes a selection and randomizes the amino acid identities in it. Random rotamers are selected.
USAGE
rand_sequence selector
ARGUMENTS
selector = string: name of the selection to "design"
NOTES
Any residues that do not contain a CA atom will be ignored.
"""
#Create lists to contain object, chain and resi info for each residue in selection.
stored.atomlist_object = []
stored.atomlist_chain = []
stored.atomlist_resi = []
#Populate the lists with all CA atoms in the selector.
cmd.iterate(
'(' + selector + ') and name ca',
'stored.atomlist_object.append(model); stored.atomlist_chain.append(chain); stored.atomlist_resi.append(resi)'
)
#Make a list of residue types
res_types = [
'ALA', 'CYS', 'ASP', 'GLU', 'PHE', 'GLY', 'HIS', 'ILE', 'LYS', 'LEU',
'MET', 'ASN', 'PRO', 'GLN', 'ARG', 'SER', 'THR', 'VAL', 'TRP', 'TYR'
]
#For each residue:
#1. Select a random residue to mutate to.
#2. Start the mutagenesis wizard interface.
#3. Figure out how many backbone-dependent rotamers exist for that residue.
#4. Select a random rotamer to mutate to.
#5. Mutate to the desired residue.
#6. Clean-up the wizard and make the mutation.
for atomidx in range(len(stored.atomlist_resi)):
curres = stored.atomlist_object[
atomidx] + ' and chain ' + stored.atomlist_chain[
atomidx] + ' and resi ' + stored.atomlist_resi[atomidx]
#Pick a random residue type.
newres = random.choice(res_types)
#Set up the mutagenesis wizard.
cmd.wizard('mutagenesis')
cmd.refresh_wizard
#Select the desired residue to mutate as (sele)
cmd.select('sele', 'none')
cmd.select('sele', curres)
#Generate the library of backbone-independent rotamers.
lib = cmd.get_wizard().ind_library.get(newres)
print(newres)
#Change the residue to the selected residue type
cmd.get_wizard().do_select('''sele''')
cmd.get_wizard().set_mode(newres)
#If the residue is not GLY, pick a random rotamer and change to it.
if (newres not in ('GLY', 'ALA')):
newrot = random.randint(1, len(lib))
cmd.get_wizard().do_state(newrot)
#Apply the changes
cmd.get_wizard().apply()
#Clear the wizard
cmd.get_wizard().clear()
cmd.set_wizard()
def tmFRET(obj, chain, r1, r2, dist, force=False, strict=False):
"""
DESCRIPTION
Finds all residues within dist of the midpoint between n1 and n2, and calculates their solvent accessible radii.
USAGE
tmFRET obj, chain, r1, r2, dist[, force=False[, strict=False]]
EXAMPLES
tmFRET 3j5p, c, 378, 380, 20 # Finds calphas w/in 20 ang. of the MBS between 378 and 380 on chain c of object 3j5p
"""
# Identify target metal binding site residues
res1str = "chain " + chain + " and resi " + r1
res2str = "chain " + chain + " and resi " + r2
cmd.set('dot_solvent', 1)
sasa1 = cmd.get_area(res1str)
sasa2 = cmd.get_area(res2str)
if ((sasa1 < 10.0) or (sasa2 < 10.0)):
if (not force):
print "WARNING: One or both of the residues you selected are likely buried."
print res1str + " SASA: " + str(sasa1)
print res2str + " SASA: " + str(sasa2)
print "If you wish to run the tmFRET helper on this pair anyway, please use:"
print "tmFRET object, chain, residue1, residue2, distance, force=True"
return
else:
print "WARNING: being forced to run the tmFRET helper on a poorly accessible site:"
print res1str + " SASA: " + str(sasa1)
print res2str + " SASA: " + str(sasa2)
else:
print res1str + " SASA: " + str(sasa1)
print res2str + " SASA: " + str(sasa2)
print "Beginning mutagenesis. Please be patient, this may take some time..."
# mutate residues to his
cmd.wizard("mutagenesis")
cmd.do("refresh_wizard")
cmd.get_wizard().set_mode("HIS")
# Set up lists of nitrogen positions
# also track CE1 posns to prevent odd orientations
ND1_1list = []
NE2_1list = []
CE1_1list = []
ND1_2list = []
NE2_2list = []
CE1_2list = []
# find all posns for nitrogens for first histidine
cmd.get_wizard().do_select(res1str)
nRot = cmd.count_states() # Find number of possible rotomers
for i in range(1,nRot+1):
cmd.get_wizard().do_select(res1str)
cmd.frame(i)
cmd.get_wizard().apply()
ND1pos = cmd.get_coords(res1str + " and name ND1")[0]
NE2pos = cmd.get_coords(res1str + " and name NE2")[0]
CE1pos = cmd.get_coords(res1str + " and name CE1")[0]
pos1String = "[%3.2f,%3.2f,%3.2f]" % (ND1pos[0], ND1pos[1], ND1pos[2])
#print res1str + " rotamer " + str(i) + " ND1 at " + pos1String
pos2String = "[%3.2f,%3.2f,%3.2f]" % (NE2pos[0], NE2pos[1], NE2pos[2])
#print res1str + " rotamer " + str(i) + " NE2 at " + pos2String
ND1_1list.append(ND1pos)
NE2_1list.append(NE2pos)
CE1_1list.append(CE1pos)
# find all posns for nitrogens for second histidine
cmd.get_wizard().do_select(res2str)
nRot = cmd.count_states() # Find number of possible rotomers
for i in range(1,nRot+1):
cmd.get_wizard().do_select(res2str)
cmd.frame(i)
cmd.get_wizard().apply()
ND1pos = cmd.get_coords(res2str + " and name ND1")[0]
NE2pos = cmd.get_coords(res2str + " and name NE2")[0]
CE1pos = cmd.get_coords(res2str + " and name CE1")[0]
pos1String = "[%3.2f,%3.2f,%3.2f]" % (ND1pos[0], ND1pos[1], ND1pos[2])
#print res2str + " rotamer " + str(i) + " ND1 at " + pos1String
pos2String = "[%3.2f,%3.2f,%3.2f]" % (NE2pos[0], NE2pos[1], NE2pos[2])
#print res2str + " rotamer " + str(i) + " NE2 at " + pos2String
ND1_2list.append(ND1pos)
NE2_2list.append(NE2pos)
CE1_2list.append(CE1pos)
# iterate over all pairs of nitrogens, determining distance for each
# Compare to distance for an "ideal" metal binding site, and save best pair
idealDistance = 3.0
bestDistance = 3.6 # use default bestDist to gate what range of values we'll accept
delta = math.fabs(idealDistance - bestDistance)
besti = -1
bestj = -1
r1N = ""
r2N = ""
print "r1rot,r2rot,ND1-ND1,ND1-NE2,NE2-ND1,NE2-NE2"
for i in range(len(ND1_1list)): # iterate over 1st H rotamers
for j in range(len(ND1_2list)): # iterate over 2nd H rotamers
# Debug diagnostic, or can be used to generate a printout to make a histogram in Excel
dND11_ND12 = distance(ND1_1list[i],ND1_2list[j])
dND11_NE22 = distance(ND1_1list[i],NE2_2list[j])
dNE21_ND12 = distance(NE2_1list[i],ND1_2list[j])
dNE21_NE22 = distance(NE2_1list[i],NE2_2list[j])
print "%i,%i,%3.2f,%3.2f,%3.2f,%3.2f" % (i, j, dND11_ND12, dND11_NE22, dNE21_ND12, dNE21_NE22)
tmpDist = distance(ND1_1list[i],ND1_2list[j])
tmpDelta = math.fabs(tmpDist-idealDistance)
if (tmpDelta < delta):
#check other nitrogen pair, to make sure they're facing away from MBS (ie: not clashing)
tmpDist2 = distance(NE2_1list[i],NE2_2list[j])
if (tmpDist2 > tmpDist):
# check apical carbons, to make sure they're not pointing into MBS
tmpDist3 = distance(CE1_1list[i],CE1_2list[j])
if strict and (tmpDist3 > tmpDist):
delta = tmpDelta
bestDistance = tmpDist
besti = i
bestj = j
r1N = "ND1"
r2N = "ND1"
tmpDist = distance(ND1_1list[i],NE2_2list[j])
tmpDelta = math.fabs(tmpDist-idealDistance)
if (tmpDelta < delta):
tmpDist2 = distance(NE2_1list[i],ND1_2list[j])
if (tmpDist2 > tmpDist):
tmpDist3 = distance(CE1_1list[i],CE1_2list[j])
if strict and (tmpDist3 > tmpDist):
delta = tmpDelta
bestDistance = tmpDist
besti = i
bestj = j
r1N = "ND1"
r2N = "NE2"
tmpDist = distance(NE2_1list[i],ND1_2list[j])
tmpDelta = math.fabs(tmpDist-idealDistance)
if (tmpDelta < delta):
tmpDist2 = distance(ND1_1list[i],NE2_2list[j])
if (tmpDist2 > tmpDist):
tmpDist3 = distance(CE1_1list[i],CE1_2list[j])
if strict and (tmpDist3 > tmpDist):
delta = tmpDelta
bestDistance = tmpDist
besti = i
bestj = j
r1N = "NE2"
r2N = "ND1"
tmpDist = distance(NE2_1list[i],NE2_2list[j])
tmpDelta = math.fabs(tmpDist-idealDistance)
if (tmpDelta < delta):
tmpDist2 = distance(ND1_1list[i],ND1_2list[j])
if (tmpDist2 > tmpDist):
tmpDist3 = distance(CE1_1list[i],CE1_2list[j])
if strict and (tmpDist3 > tmpDist):
delta = tmpDelta
bestDistance = tmpDist
besti = i
bestj = j
r1N = "NE2"
r2N = "NE2"
if (besti == -1 or bestj == -1): #couldn't find a rotamer pair in the accepted distance range
print "No appropriate rotamers could be found for the selected metal binding site residue pair. Please try again with different residues."
# recover original residues here somehow, should go back and build this in
cmd.set_wizard("done") #cleanup
return
# set up the MBS
cmd.get_wizard().do_select(res1str)
cmd.frame(besti+1)
cmd.get_wizard().apply()
cmd.get_wizard().do_select(res2str)
cmd.frame(bestj+1)
cmd.get_wizard().apply()
cmd.set_wizard("done") #cleanup
# make the metal as a pseudoatom
posn1 = cmd.get_coords(res1str + " and name " + r1N)
posn2 = cmd.get_coords(res2str + " and name " + r2N)
# midpoint between 2 coordinating nitrogens is the MBS
MBS = (posn1 + posn2) / 2
MBS = MBS[0] # necessary b/c get_coords returns a coord list: [[x,y,z]]
# make a pseudoatom at the MBS
posString = "[%3.2f,%3.2f,%3.2f]" % (MBS[0], MBS[1], MBS[2])
print "Making metal pseudoatom at" + posString
cmd.pseudoatom("XitionMetal", pos=posString)
cmd.show("spheres", "XitionMetal")
# find the residues within dist of the MBS
selString="XitionMetal around " + dist + " and name cb"
cmd.select("potentialSites", selString)
# iterate over the residues in the selection, coloring Calpha by the solvent accessible surface area
myspace={'markAccessible': markAccessible}
cmd.iterate('(potentialSites)', 'markAccessible(model, chain, resi)', space=myspace)
'hip', 'hie', 'hid', 'ile', 'leu', 'lys', 'met', 'phe', 'pro',
'ser', 'thr', 'trp', 'tyr', 'val'):
stored.fc_dict = {}
print "'" + string.upper(grp) + "': ["
for res in (grp, "nt_" + grp, "ct_" + grp):
ch = Champ()
print " ("
if grp == 'wat' and len(res) > 3:
continue
cmd.fragment(res, "tmp")
stored.pc_dict = {}
stored.at_dict = {}
cmd.iterate(
"tmp",
"stored.pc_dict[name]=partial_charge;stored.at_dict[name]=text_type",
quiet=1)
if res[0:3] == "ct_":
cmd.alter("name OXT", "formal_charge=-1")
elif res[0:3] == "nt_":
cmd.alter("name N", "formal_charge=1")
# some PyMOL fragments are missing formal charges...
cmd.alter("GLU/OE2|ASP/OD2", "formal_charge=-1")
cmd.alter("HIP/ND1|ARG/NH1|LYS/NZ", "formal_charge=1")
model = cmd.get_model("tmp")
cmd.delete("tmp")
m1 = ch.insert_model(model)
ch.pattern_detect_chirality(m1)
ch.pattern_orient_bonds(m1)
pat1 = ch.pattern_get_string_with_names(m1)
def test_dss(self):
ss_list = []
cmd.fab('A' * 6, ss=1)
cmd.dss()
cmd.iterate('2-5/CA', 'ss_list.append(ss)', space=locals())
self.assertEqual(ss_list, ['H', 'H', 'H', 'H'])
def spectrumany(expression,
color_list,
selection='(all)',
minimum=None,
maximum=None,
quiet=1):
'''
DESCRIPTION
Define a color spectrum with as many color-stops as you like (at least 2).
USAGE
spectrumany expression, color_list [, selection [, minimum [, maximum ]]]
ARGUMENTS
expression = count, resi, b, q, or pc: respectively, atom count, residue
index, temperature factor, occupancy, or partial charge {default: count}
color_list = string: Space separated list of colors
... all other arguments like with `spectrum` command
EXAMPLE
spectrumany count, forest green yellow white
spectrumany b, red yellow white, (polymer), maximum=100.0
SEE ALSO
spectrum
'''
quiet = int(quiet)
colors = color_list.split()
if len(colors) < 2:
print('failed! please provide at least 2 colors')
return
colvec = [cmd.get_color_tuple(i) for i in colors]
parts = len(colvec) - 1
expression = {
'pc': 'partial_charge',
'fc': 'formal_charge',
'count': 'index'
}.get(expression, expression)
minmax_expr = {'resi': 'resv'}.get(expression, expression)
discrete_expr = ['index', 'resi']
if cmd.count_atoms(selection) == 0:
print('empty selection')
return
if None in [minimum, maximum]:
stored.e = list()
cmd.iterate(selection, 'stored.e.append(%s)' % (minmax_expr))
if minimum is None:
minimum = min(stored.e)
if maximum is None:
maximum = max(stored.e)
minimum, maximum = float(minimum), float(maximum)
if not quiet:
print(' Spectrum: range (%.5f to %.5f)' % (minimum, maximum))
if maximum == minimum:
print('no spectrum possible, only equal %s values' % (expression))
return
if expression in discrete_expr:
val_range = int(maximum - minimum + 1)
else:
val_range = maximum - minimum
cmd.color(colors[0], selection)
steps = 60 / parts
steps_total = steps * parts
val_start = minimum
for p in range(parts):
for i in range(steps):
ii = float(i) / steps
col_list = [
colvec[p + 1][j] * ii + colvec[p][j] * (1.0 - ii)
for j in range(3)
]
col_name = '0x%02x%02x%02x' % (col_list[0] * 255, col_list[1] *
255, col_list[2] * 255)
val_end = val_range * (i + 1 + p * steps) / steps_total + minimum
if expression in discrete_expr:
cmd.color(
col_name, '(%s) and %s %d-%d' %
(selection, expression, val_start, val_end))
else:
cmd.color(
col_name,
'(%s) and %s > %f' % (selection, expression, val_start))
val_start = val_end
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 tmalign(mobile, target, mobile_state=1, target_state=1, args='',
exe='TMalign', ter=0, transform=1, object=None, quiet=0):
'''
DESCRIPTION
TMalign wrapper. You may also use this as a TMscore or MMalign wrapper
if you privide the corresponding executable with the "exe" argument.
Reference: Y. Zhang and J. Skolnick, Nucl. Acids Res. 2005 33, 2302-9
http://zhanglab.ccmb.med.umich.edu/TM-align/
ARGUMENTS
mobile, target = string: atom selections
mobile_state, target_state = int: object states {default: 1}
args = string: Extra arguments like -d0 5 -L 100
exe = string: Path to TMalign (or TMscore, MMalign) executable
{default: TMalign}
ter = 0/1: If ter=0, then ignore chain breaks because TMalign will stop
at first TER record {default: 0}
'''
import subprocess, tempfile, os, re
from .exporting import save_pdb_without_ter
ter, quiet = int(ter), int(quiet)
mobile_filename = tempfile.mktemp('.pdb', 'mobile')
target_filename = tempfile.mktemp('.pdb', 'target')
matrix_filename = tempfile.mktemp('.txt', 'matrix')
mobile_ca_sele = '(%s) and (not hetatm) and name CA and alt +A' % (mobile)
target_ca_sele = '(%s) and (not hetatm) and name CA and alt +A' % (target)
if ter:
save = cmd.save
else:
save = save_pdb_without_ter
save(mobile_filename, mobile_ca_sele, state=mobile_state)
save(target_filename, target_ca_sele, state=target_state)
exe = cmd.exp_path(exe)
args = [exe, mobile_filename, target_filename, '-m', matrix_filename] + args.split()
try:
process = subprocess.Popen(args, stdout=subprocess.PIPE)
lines = process.stdout.readlines()
except OSError:
print('Cannot execute "%s", please provide full path to TMscore or TMalign executable' % (exe))
raise CmdException
finally:
os.remove(mobile_filename)
os.remove(target_filename)
# TMalign >= 2012/04/17
if os.path.exists(matrix_filename):
lines += open(matrix_filename).readlines()
os.remove(matrix_filename)
r = None
re_score = re.compile(r'TM-score\s*=\s*(\d*\.\d*)')
rowcount = 0
matrix = []
line_it = iter(lines)
headercheck = False
alignment = []
for line in line_it:
if 4 >= rowcount > 0:
if rowcount >= 2:
a = list(map(float, line.split()))
matrix.extend(a[2:5])
matrix.append(a[1])
rowcount += 1
elif not headercheck and line.startswith(' * '):
a = line.split(None, 2)
if len(a) == 3:
headercheck = a[1]
elif line.lower().startswith(' -------- rotation matrix'):
rowcount = 1
elif line.startswith('(":" denotes'):
alignment = [line_it.next().rstrip() for i in range(3)]
else:
match = re_score.search(line)
if match is not None:
r = float(match.group(1))
if not quiet:
print(line.rstrip())
if not quiet:
for i in range(0, len(alignment[0])-1, 78):
for line in alignment:
print(line[i:i+78])
print('')
assert len(matrix) == 3*4
matrix.extend([0,0,0,1])
if int(transform):
for model in cmd.get_object_list('(' + mobile + ')'):
cmd.transform_object(model, matrix, state=0, homogenous=1)
# alignment object
if object is not None:
mobile_idx, target_idx = [], []
space = {'mobile_idx': mobile_idx, 'target_idx': target_idx}
cmd.iterate(mobile_ca_sele, 'mobile_idx.append("%s`%d" % (model, index))', space=space)
cmd.iterate(target_ca_sele, 'target_idx.append("%s`%d" % (model, index))', space=space)
for i, aa in enumerate(alignment[0]):
if aa == '-':
mobile_idx.insert(i, None)
for i, aa in enumerate(alignment[2]):
if aa == '-':
target_idx.insert(i, None)
if (len(mobile_idx) == len(target_idx) == len(alignment[2])):
cmd.rms_cur(
' '.join(idx for (idx, m) in zip(mobile_idx, alignment[1]) if m in ':.'),
' '.join(idx for (idx, m) in zip(target_idx, alignment[1]) if m in ':.'),
cycles=0, matchmaker=4, object=object)
else:
print('Could not load alignment object')
if not quiet:
if headercheck:
print('Finished Program:', headercheck)
if r is not None:
print('Found in output TM-score = %.4f' % (r))
return r
# CD4 = chains C, E, F
# 17b = chains H, K, M and L, N, O
cmd.remove('c;C,E,F,H,K,M,L,N,O')
# Tweak initial display and color of Env monomers
cmd.hide('everything')
cmd.bg_color('white')
cmd.show('cartoon')
cmd.color('grey40')
#cmd.color('grey20', structure)
cmd.set('cartoon_transparency', '0.5')
cmd.set('cartoon_transparency', '0', structure)
# Identify unique sites in structure
sites_in_structure = []
cmd.iterate("(name ca)", "sites_in_structure.append(resi)")
unique_sites_in_structure = []
for site in sites_in_structure:
if site not in unique_sites_in_structure:
unique_sites_in_structure.append(site)
# Get site-specific RMSDcorrected values and whether the shift at a site is
# significant
RMSD_dict = {}
sig_dict = {}
with open('../BG505_to_BF520_prefs_dist.csv') as f:
lines = f.readlines()[1:]
for line in lines:
elements = line.split(',')
site = elements[0]
RMSDcorrected = elements[1]
def list_hb(selection,
selection2=None,
cutoff=3.2,
angle=55,
mode=1,
output='hbonds'):
"""
USAGE
list_hb selection, [selection2 (default=None)], [cutoff (default=3.2)],
[angle (default=55)], [mode (default=1)],
[hb_list_name (default='hbonds')]
The script automatically adds a requirement that atoms in the
selection (and selection2 if used) must be either of the elements N or
O.
If mode is set to 0 instead of the default value 1, then no angle
cutoff is used, otherwise the angle cutoff is used and defaults to 55
degrees.
e.g.
To get a list of all H-bonds within chain A of an object
list_hb 1abc & c. a &! r. hoh, cutoff=3.2, hb_list_name=abc-hbonds
To get a list of H-bonds between chain B and everything else:
list_hb 1tl9 & c. b, 1tl9 &! c. b
"""
cutoff = float(cutoff)
angle = float(angle)
mode = float(mode)
selection = selection + " & e. n+o"
if not selection2:
hb = cmd.find_pairs(selection,
selection,
mode=mode,
cutoff=cutoff,
angle=angle)
else:
selection2 = selection2 + " & e. n+o"
hb = cmd.find_pairs(selection,
selection2,
mode=mode,
cutoff=cutoff,
angle=angle)
# sort the list for easier reading
hb.sort(lambda x, y: (cmp(x[0][1], y[0][1])))
stored.hb_list = []
for pairs in hb:
cmd.iterate(
"%s and index %s" % (pairs[0][0], pairs[0][1]),
'stored.hb_list.append("%1s,%3s,%s,%-4s," % (chain,resn,resi,name)),'
)
cmd.iterate(
"%s and index %s" % (pairs[1][0], pairs[1][1]),
'stored.hb_list.append("%1s,%3s,%s,%-4s," % (chain,resn,resi,name)),'
)
stored.hb_list.append(
"%.2f" %
cmd.distance(output, "%s and index %s" %
(pairs[0][0], pairs[0][1]), "%s and index %s" %
(pairs[1][0], pairs[1][1]) + ',') + '\n')
def mutate(
self, pdbid: str,
replace_with: Dict[int,
Optional[str]]) -> Union[List[str], rdchem.Mol]:
"""Modify amino acid residues at the defined positions.
If the locations indexes exceed the amount in data, then they
will be ignored and a warning will be announced.
Params:
-------
pdbid: str
PDB ID associated with the structure.
replace_with: dict
The index location(s) within the full protein to replace
certain residue(s) with. If a residue associated with a
index location is None, then the modified residue is chosen
randomly. If a certain index exceeds the number of available
residues in the protein, then those enteries are simply
ignored and the user is notified.
Returns:
--------
protein: list of str or rdkit.Chem.rdchem.Mol
Modified protein with residues. If fmt="primary", then list
of string (peptide names) is returned. If fmt="tertiary",
then 3D molecule structure is returned.
"""
# Load PDB structure (download, if necessary)
pdb_dir = maybe_create_dir(os.path.join(self.rootdir, pdbid))
pdb_file = os.path.join(pdb_dir, f"{pdbid}.pdb")
if not os.path.exists(pdb_file):
is_successful = cmd.fetch(pdbid,
name=pdbid,
state=1,
type="pdb",
path=pdb_dir)
if is_successful == -1:
raise DownloadError(f"Unable to download '{pdbid}'.")
else:
cmd.load(pdb_file, object=pdbid, state=1, format="pdb")
# Get all residue names, see: https://pymolwiki.org/index.php/List_Selection
resnames_dict = {"names": []}
cmd.iterate("(name ca)", "names.append(resn)", space=resnames_dict)
residue_names = resnames_dict["names"]
num_residues = len(residue_names)
# Cleanup idxs: remove indicies that exceed number of available residues
nonvalid_idxs = [
idx for idx in replace_with.keys() if idx > num_residues
]
for idx in nonvalid_idxs:
print(
f"OutOfRange: Removing idx {idx} (only {num_residues} residues)."
)
replace_with.pop(idx)
# Randomly choose an amino acid (AA) to replace residue, if None is provided.
# Additionally, format string such that it is a valid 3 letter amino acid.
for idx, residue in replace_with.items():
if residue is None:
replace_with[idx] = np.random.choice(aa3)
elif is_aa(residue):
residue = residue.upper()
if len(residue) == 1:
replace_with[idx] = one_to_three.get(residue)
elif len(residue) == 3:
replace_with[idx] = residue
else:
raise ValueError(
f"Invalid residue '{residue}'. Choose one from "
f"the following {aa1+aa3}.")
# Determine save filepath name
modified_res_str = ":".join(
[f"{k}{three_to_one.get(v)}" for k, v in replace_with.items()])
filename = f"{self.fmt}_{modified_res_str}"
filename += ".pdb" if self.fmt == "tertiary" else ".json"
save_filepath = os.path.join(self.rootdir, pdbid, filename)
# Replace primary structure, i.e. residue names (str)
if self.fmt == "primary":
# Load data from cache, if it exists
protein = None
if os.path.exists(save_filepath):
with open(save_filepath) as json_file:
protein = json.load(json_file)
if protein is None:
for idx, residue in replace_with.items():
residue_names[
idx - 1] = residue # since PDB starts with idx of 1
protein = [three_to_one.get(name) for name in residue_names]
# Save sequence temporarily
_ = maybe_create_dir(save_filepath)
with open(save_filepath, "w") as outfile:
json.dump(protein, outfile)
# Replace tertiary structure, i.e. residue's 3D coordinates
elif self.fmt == "tertiary":
if not os.path.exists(save_filepath):
# Split states so that we can optimize only on specific state(s).
# NOTE: Might be useful to choose lowest energy state to mutate,
# OR mutate rotamers for all positions, then choose one with
# lowest energy.
cmd.split_states(object=pdbid)
# Delete all other objects other than one we want to mutate
# NOTE: For now, keep only first object. This might change
# depending on which state needs to be kept.
objs = cmd.get_object_list() # aka states
keep_objs = [pdbid + "_0001"]
for obj in objs:
if obj not in keep_objs:
cmd.delete(obj)
assert keep_objs == cmd.get_object_list()
# Mutate residues
cmd.wizard("mutagenesis")
wizard: Mutagenesis = cmd.get_wizard()
for idx, res in replace_with.items():
selection = "{0:s}//A/{1:d}/".format(keep_objs[0], idx)
wizard.do_select(
selection) # select which residue index to replace
wizard.set_mode(
res) # choose name of residue to replace with
wizard.do_state(
1
) # select rotamer with least strain (aka conflicts w/ other atoms)
wizard.apply() # apply point mutation
cmd.set_wizard(None) # close wizard
# Save PDB temporarily
_ = maybe_create_dir(save_filepath)
cmd.save(save_filepath, selection=pdbid, format="pdb")
cmd.delete("all") # remove all objects, clears workspace
# Load + choose model/structure with lowest energy
# NOTE: If sanitize=True, the function checks if Mol has the correct
# hybridization/valance structure (aka is it chemically reasonable).
# When converting from the PDB block, this sometimes results in
# improper parsing. Instead, for now, we just check if the Mol is
# syntactically valid (i.e. all rings/branches closed, no illegal
# atom types, etc).
protein = rdmolfiles.MolFromPDBFile(save_filepath,
sanitize=False,
removeHs=False)
if protein.GetNumConformers() > 1:
protein = _get_conformer(protein, conformer="min", algo="MMFF")
else:
raise NotImplementedError
# Remove file, if not needed
if not self.cache:
os.remove(save_filepath)
return protein
def interfaceResidues(cmpx,
cA='c. A',
cB='c. B',
cutoff=1.0,
selName="interface"):
"""
interfaceResidues -- finds 'interface' residues between two chains in a complex.
PARAMS
cmpx
The complex containing cA and cB
cA
The first chain in which we search for residues at an interface
with cB
cB
The second chain in which we search for residues at an interface
with cA
cutoff
The difference in area OVER which residues are considered
interface residues. Residues whose dASA from the complex to
a single chain is greater than this cutoff are kept. Zero
keeps all residues.
selName
The name of the selection to return.
RETURNS
* A selection of interface residues is created and named
depending on what you passed into selName
* An array of values is returned where each value is:
( modelName, residueNumber, dASA )
NOTES
If you have two chains that are not from the same PDB that you want
to complex together, use the create command like:
create myComplex, pdb1WithChainA or pdb2withChainX
then pass myComplex to this script like:
interfaceResidues myComlpex, c. A, c. X
This script calculates the area of the complex as a whole. Then,
it separates the two chains that you pass in through the arguments
cA and cB, alone. Once it has this, it calculates the difference
and any residues ABOVE the cutoff are called interface residues.
AUTHOR:
Jason Vertrees, 2009.
"""
# Save user's settings, before setting dot_solvent
oldDS = cmd.get("dot_solvent")
cmd.set("dot_solvent", 1)
# set some string names for temporary objects/selections
tempC, selName1 = "tempComplex", selName + "1"
chA, chB = "chA", "chB"
# operate on a new object & turn off the original
cmd.create(tempC, cmpx)
cmd.disable(cmpx)
# remove cruft and inrrelevant chains
cmd.remove(tempC + " and not (polymer and (%s or %s))" % (cA, cB))
# get the area of the complete complex
cmd.get_area(tempC, load_b=1)
# copy the areas from the loaded b to the q, field.
cmd.alter(tempC, 'q=b')
# extract the two chains and calc. the new area
# note: the q fields are copied to the new objects
# chA and chB
cmd.extract(chA, tempC + " and (" + cA + ")")
cmd.extract(chB, tempC + " and (" + cB + ")")
cmd.get_area(chA, load_b=1)
cmd.get_area(chB, load_b=1)
# update the chain-only objects w/the difference
cmd.alter("%s or %s" % (chA, chB), "b=b-q")
# The calculations are done. Now, all we need to
# do is to determine which residues are over the cutoff
# and save them.
stored.r, rVal, seen = [], [], []
cmd.iterate('%s or %s' % (chA, chB), 'stored.r.append((model,resi,b))')
cmd.enable(cmpx)
cmd.select(selName1, 'none')
for (model, resi, diff) in stored.r:
key = resi + "-" + model
if abs(diff) >= float(cutoff):
if key in seen: continue
else: seen.append(key)
rVal.append((model, resi, diff))
# expand the selection here; I chose to iterate over stored.r instead of
# creating one large selection b/c if there are too many residues PyMOL
# might crash on a very large selection. This is pretty much guaranteed
# not to kill PyMOL; but, it might take a little longer to run.
cmd.select(selName1,
selName1 + " or (%s and i. %s)" % (model, resi))
# this is how you transfer a selection to another object.
cmd.select(selName, cmpx + " in " + selName1)
# clean up after ourselves
cmd.delete(selName1)
cmd.delete(chA)
cmd.delete(chB)
cmd.delete(tempC)
# show the selection
cmd.enable(selName)
# reset users settings
cmd.set("dot_solvent", oldDS)
return rVal
def map_new_apbs(name, selection='all', grid=0.5, buffer=10.0, state=1,
preserve=0, exe='', assign=-1, focus='', quiet=1):
'''
DESCRIPTION
Create electrostatic potential map with APBS.
For more control over parameters and a graphical user interface I
recommend to use the APBS Tools Plugin instead.
In case of missing atoms or residues I recommend to remodel the input
structure with modeller before calculating the electrostatic potential.
If selection has no charges and radii, they will be automatically assigned
with PyMOL (not with pdb2pqr).
SEE ALSO
apbs_surface, map_new (coulomb), APBS Tools Plugin
'''
import tempfile, os, shutil, glob, subprocess
from pymol.util import protein_assign_charges_and_radii
from .modelling import add_missing_atoms
selection = '(%s) and not solvent' % (selection)
grid, buffer, state = float(grid), float(buffer), int(state)
preserve, assign, quiet = int(preserve), int(assign), int(quiet)
if exe:
exe = cmd.exp_path(exe)
else:
try:
import freemol.apbs
exe = freemol.apbs.get_exe_path()
except:
pass
if not exe:
exe = "apbs"
try:
r = subprocess.call([exe, "--version"],
stdout=open(os.devnull, "w"), stderr=subprocess.STDOUT)
if r < 0:
raise CmdException("Broken executable: " + exe)
except OSError:
raise CmdException("Cannot execute: " + exe)
# temporary directory
tempdir = tempfile.mkdtemp()
if not quiet:
print(' Tempdir:', tempdir)
# filenames
pqrfile = os.path.join(tempdir, 'mol.pqr')
infile = os.path.join(tempdir, 'apbs.in')
stem = os.path.join(tempdir, 'map')
# temporary object
tmpname = cmd.get_unused_name('mol' if preserve else '_')
cmd.create(tmpname, selection, state, 1)
# partial charges
assign = [assign]
if assign[0] == -1:
# auto detect if selection has charges and radii
cmd.iterate('first ((%s) and elem O)' % (tmpname),
'assign[0] = (elec_radius * partial_charge) == 0.0',
space=locals())
if assign[0]:
cmd.remove('hydro and model ' + tmpname)
add_missing_atoms(tmpname, quiet=quiet)
protein_assign_charges_and_radii(tmpname)
elif not quiet:
print(' Notice: using exsiting charges and radii')
cmd.save(pqrfile, tmpname, 1, format='pqr', quiet=quiet)
# grid dimensions
extent = cmd.get_extent(tmpname)
extentfocus = cmd.get_extent(focus) if focus else extent
fglen = [(emax-emin + 2*buffer) for (emin, emax) in zip(*extentfocus)]
cglen = [(emax-emin + 4*buffer) for (emin, emax) in zip(*extent)]
if not preserve:
cmd.delete(tmpname)
apbs_in = defaults_apbs_in.copy()
apbs_in['fglen'] = '%f %f %f' % tuple(fglen)
apbs_in['cglen'] = '%f %f %f' % tuple(cglen)
apbs_in['srad'] = cmd.get('solvent_radius')
apbs_in['write'] = 'pot dx "%s"' % (stem)
if focus:
apbs_in['fgcent'] = '%f %f %f' % tuple((emax + emin) / 2.0
for (emin, emax) in zip(*extentfocus))
# apbs input file
def write_input_file():
f = open(infile, 'w')
print('''
read
mol pqr "%s"
end
elec
mg-auto
''' % (pqrfile), file=f)
for (k,v) in apbs_in.items():
if v is None:
print(k, file=f)
elif isinstance(v, list):
for vi in v:
print(k, vi, file=f)
else:
print(k, v, file=f)
print('''
end
quit
''', file=f)
f.close()
try:
# apbs will fail if grid does not fit into memory
# -> on failure repeat with larger grid spacing
for _ in range(3):
dime = [1 + max(64, n / grid) for n in fglen]
apbs_in['dime'] = '%d %d %d' % tuple(dime)
write_input_file()
# run apbs
r = subprocess.call([exe, infile], cwd=tempdir)
if r == 0:
break
if r in (-6, -9):
grid *= 2.0
if not quiet:
print(' Warning: retry with grid =', grid)
continue
print(' Error: apbs failed with code', r)
raise CmdException
dx_list = glob.glob(stem + '*.dx')
if len(dx_list) != 1:
print(' Error: dx file missing')
raise CmdException
# load map
cmd.load(dx_list[0], name, quiet=quiet)
except OSError:
print(' Error: Cannot execute "%s"' % (exe))
raise CmdException
finally:
if not preserve:
shutil.rmtree(tempdir)
elif not quiet:
print(' Notice: not deleting %s' % (tempdir))
def test_alter_list(self):
cmd.fragment('gly')
cmd.alter_list('gly', [[i + 1, 'name = "X%d"' % i] for i in range(7)])
name_list = []
cmd.iterate('gly', 'name_list.append(name)', space=locals())
self.assertEqual(name_list, ['X%d' % i for i in range(7)])
import __main__
__main__.pymol_argv = ['pymol', '-qc']
import pymol
from pymol import cmd, stored
pymol.finish_launching()
cmd.set('dot_solvent', 1)
cmd.set('dot_density', 4) # Lowest resolution
pdb = '1y57_human_numbering'
cmd.load('../pdbs/' + pdb + '.pdb') # use the name of your pdb file
stored.residues = []
cmd.iterate('name ca', 'stored.residues.append(resi)')
sasa_per_residue = []
idx = []
for i in stored.residues:
sasa_per_residue.append(cmd.get_area('resi %s' % i))
idx.append(i)
with open(pdb + '_SASA.txt', 'w') as f:
for i, item in zip(idx, sasa_per_residue):
#f.write("%s\n" % item)
line = "{}\t{}\n".format(i, item)
f.write(line)
print sum(sasa_per_residue)
print cmd.get_area(
'all'
) # just to check that the sum of sasa per residue equals the total area
def format_bonds(
selection='all',
bonds=4,
):
'''
DESCRIPTION
Formats bonds in aromatic or charged residues
EXAMPLE
frag PHE
format_bonds
USAGE
format_bonds [ selection [, bonds ]]
ARGUMENTS
selection: <str> input selection {default: 'all'}
bonds: <int> toogles format of bonds
1: single bonds (deactivates valence display)
2: regular double bonds (activates valence display)
>=3: delocalized (activates valence display)
'''
# Selection
try:
# group selection with bracketing and select complete residues
selection = '(byres (' + str(selection) + '))'
# checks functional selection
cmd.count_atoms(selection)
except:
print("invalid selection")
return False
# PARAMETERS
try:
bonds = int(bonds)
except:
pass
if (not (bonds in [1, 2])):
bonds = 4
if bonds == 1:
cmd.set('valence', 0)
print("Valence display disabled!")
return bonds
else:
cmd.set('valence', 1)
print("Valence display enabled!")
# proceed
##### SELECTION BY OBJECT AND CHAIN #####
# variable for the selections
# get the names of the proteins in the selection
objects = cmd.get_object_list(selection)
# include chains
# subselect chains
names = []
for p in objects:
for chain in cmd.get_chains('model ' + p) or ['']:
names.append("(model %s and chain '%s')" % (p, chain))
##### SELECTION LISTS #####
# get TRP
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn TRP+NIW) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
# the integer is to ensure unique keys
TRP_tuple = (1, ) + tuple(stored.temp)
# get PHETYR
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn PHE+TYR+PTR+NIY+PNIY) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
PHETYR_tuple = (2, ) + tuple(stored.temp)
# get HIS
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn HIS) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
HIS_tuple = (3, ) + tuple(stored.temp)
# get NITRO
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn NIY+PNIY+NIW) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
NITRO_tuple = (4, ) + tuple(stored.temp)
# get GLU
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn GLU) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
GLU_tuple = (5, ) + tuple(stored.temp)
# get ASP
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn ASP) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
ASP_tuple = (6, ) + tuple(stored.temp)
# get CTERM
stored.temp = []
for p in names:
cmd.iterate(
'(byres (last %s)) and (not (hetatm)) '
'and (name OXT)' % (p),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
CTERM_tuple = (7, ) + tuple(stored.temp)
# get ARG
stored.temp = []
for p in names:
cmd.iterate(('(%s) and (resn ARG) '
'and (name CA)' % (p)),
'stored.temp.append("(%s and resi "+str(resi)+")")' % p)
ARG_tuple = (8, ) + tuple(stored.temp)
##### SELECTION TUPLES DONE #####
##### ATOM LISTS #####
TRP_bonds_all = [['CG', 'CD1'], ['CD1', 'NE1'], ['NE1', 'CE2'],
['CE2', 'CD2'], ['CD2', 'CG'], ['CD2', 'CE3'],
['CE3', 'CZ3'], ['CZ3', 'CH2'], ['CH2', 'CZ2'],
['CZ2', 'CE2']]
TRP_bonds_double = [['CG', 'CD1'], ['CE2', 'CD2'], ['CE3', 'CZ3'],
['CH2', 'CZ2']]
PHETYR_bonds_all = [['CG', 'CD1'], ['CD1', 'CE1'], ['CE1', 'CZ'],
['CZ', 'CE2'], ['CE2', 'CD2'], ['CD2', 'CG']]
PHETYR_bonds_double = [['CG', 'CD1'], ['CE1', 'CZ'], ['CE2', 'CD2']]
HIS_bonds_all = [
['CG', 'CD2'],
['CD2', 'NE2'],
['NE2', 'CE1'],
['CE1', 'ND1'],
['ND1', 'CG'],
]
HIS_bonds_double = [['CG', 'CD2'], ['CE1', 'ND1']]
NITRO_bonds_all = [['NN', 'O1'], ['NN', 'O2']]
NITRO_bonds_double = [['NN', 'O1']]
GLU_bonds_all = [['CD', 'OE1'], ['CD', 'OE2']]
GLU_bonds_double = [['CD', 'OE1']]
ASP_bonds_all = [['CG', 'OD1'], ['CG', 'OD2']]
ASP_bonds_double = [['CG', 'OD1']]
CTERM_bonds_all = [['C', 'O'], ['C', 'OXT']]
CTERM_bonds_double = [['C', 'O']]
ARG_bonds_all = [['CZ', 'NH1'], ['CZ', 'NH2']]
ARG_bonds_double = [['CZ', 'NH1']]
##### FORMATING #####
# dictionary: entries:atoms
format_dict = {
TRP_tuple: [TRP_bonds_all, TRP_bonds_double],
PHETYR_tuple: [PHETYR_bonds_all, PHETYR_bonds_double],
HIS_tuple: [HIS_bonds_all, HIS_bonds_double],
NITRO_tuple: [NITRO_bonds_all, NITRO_bonds_double],
GLU_tuple: [GLU_bonds_all, GLU_bonds_double],
ASP_tuple: [ASP_bonds_all, ASP_bonds_double],
CTERM_tuple: [CTERM_bonds_all, CTERM_bonds_double],
ARG_tuple: [ARG_bonds_all, ARG_bonds_double]
}
if bonds != 2:
lines = 4
print("Formating as delocalized bonds")
else:
lines = 1
print("Formating as double bonds")
# for all tuples (i.e format_dict.keys())
for p in list(format_dict.keys()):
# go through list except ID at pos 1
for q in p[1:]:
# format bonds
for r in format_dict[p][0]:
cmd.unbond('%s and name %s' % (q, r[0]),
'%s and name %s' % (q, r[1]))
cmd.bond('%s and name %s' % (q, r[0]),
'%s and name %s' % (q, r[1]), lines)
if lines == 1:
# add double bonds
for r in format_dict[p][1]:
cmd.unbond('%s and name %s' % (q, r[0]),
'%s and name %s' % (q, r[1]))
cmd.bond('%s and name %s' % (q, r[0]),
'%s and name %s' % (q, r[1]), 2)
return bonds
def drawNetwork(path,
userSelection='all',
r=1,
edge_norm=None,
alpha=0.5,
node_color=(0.6, 0.6, 0.6),
edge_color1=(1, 0, 0),
edge_color2=(0, 0, 1),
labelling='0',
threshold=0):
'''
Draws a NetworkX network on the PyMol structure
'''
cmd.delete('nodes *edges')
cmd.label(selection=userSelection, expression="")
# Building position -- name correspondance
stored.posCA = []
stored.names = []
userSelection = userSelection + " and n. CA or n. C"
cmd.iterate_state(1, selector.process(userSelection),
"stored.posCA.append([x,y,z])")
cmd.iterate(userSelection, 'stored.names.append(resn+resi+chain)')
stored.labels = list(map(relabel, stored.names))
stored.resid = list(map(selection, stored.names))
node2id = dict(zip(stored.labels, stored.resid))
node2CA = dict(zip(stored.labels, stored.posCA))
# Getting graph
net = nx.read_gpickle(path)
cmd.set('auto_zoom', 0)
cmd.set("cgo_sphere_quality", 4)
#Drawing nodes
# selnodes = ""
# for u in net.nodes():
# x, y, z = node2CA[u]
# obj+=[SPHERE, x, y, z, r]
# selnodes+=node2id[u]
# selnodes = selnodes[4:]
# if labelling=='1':
# cmd.label(selection=selnodes, expression="t2o(resn)+resi")
# if labelling=='3':
# cmd.label(selection=selnodes, expression="resn+resi")
# cmd.load_cgo(obj, 'nodes')
if edge_norm == None:
edge_norm = max(
[net.edges()[(u, v)]['weight'] for u, v in net.edges()]) / r
obj1, obj2, nodelist = [], [], []
for u, v in net.edges():
radius = net[u][v]['weight'] / edge_norm
if abs(net[u][v]['weight']) >= threshold:
if 'color' in net[u][v]:
if net[u][v]['color'] == 'r':
obj1 += [
CYLINDER, *node2CA[u], *node2CA[v], radius,
*edge_color1, *edge_color1
]
else:
obj2 += [
CYLINDER, *node2CA[u], *node2CA[v], radius,
*edge_color2, *edge_color2
]
else:
if net[u][v]['weight'] <= 0:
obj1 += [
CYLINDER, *node2CA[u], *node2CA[v], radius,
*edge_color1, *edge_color1
]
else:
obj2 += [
CYLINDER, *node2CA[u], *node2CA[v], radius,
*edge_color2, *edge_color2
]
nodelist += [u, v]
#Drawing nodes
obj = [COLOR, *node_color]
nodelist = set(nodelist)
selnodes = ''.join([node2id[u] for u in nodelist])[4:]
for u in nodelist:
x, y, z = node2CA[u]
obj += [SPHERE, x, y, z, r]
if labelling == '1':
cmd.label(selection=selnodes, expression="t2o(resn)+resi")
if labelling == '3':
cmd.label(selection=selnodes, expression="resn+resi")
cmd.load_cgo(obj, 'nodes')
cmd.load_cgo(obj1, 'holo_edges')
cmd.load_cgo(obj2, 'apo_edges')
firstChainToCompare.id.upper()),
'%s & chain %s' % (structureName,
chainToCompare.id.upper()))
# Make sure the residues in each chain are zero-based.
stored.first = None
structureList = ['(model %s and (%s))' % (obj, structureName)
for obj in cmd.get_object_list(
'(' + structureName + ')')]
structureChainList = ['(%s and chain %s)' % (structure, chain)
for structure in structureList for chain in
cmd.get_chains(structure)]
for structureChain in structureChainList:
cmd.iterate('first %s and polymer and n. CA' % structureChain,
'stored.first=resv')
# Reassign the residue numbers.
cmd.alter(structureChain,
'resi=str(int(resi)-%d)' % int(stored.first))
# Alter the coordinates if necessary.
if first:
shift = 100
else:
shift = -100
if not setGridMode:
print('altering', structureName, shift)
cmd.alter_state(1, structureName, 'x+=%d' % shift)
cmd.rebuild()
import os
import os.path
import glob
from pymol import cmd
from pymol import stored
from pymol import selector
files = glob.glob("*.pdb")
print(files)
for file in files:
pdbName = os.path.basename(file).split(".")[0]
cmd.load(file, pdbName)
outFile = open(pdbName + '.ss', 'w+')
stored.ss = ""
cmd.iterate('(n. CA)', 'stored.ss = stored.ss + ("%1s"%ss)')
for c in stored.ss:
if c == " ":
outFile.write('.')
else:
outFile.write(c)
cmd.delete(pdbName)
outFile.close()
def select_sspick(selection, name=None, caonly=0, quiet=0):
'''
DESCRIPTION
Extend selection by connected secondary structure elements.
Also available as wizard (type: "wizard sspick").
USAGE
select_sspick selection [, name [, caonly [, quiet ]]]
ARGUMENTS
selection = string: selection-expression
name = string: create a named atom selection if not None {default: None}
'''
caonly, quiet = int(caonly), int(quiet)
qkeys = set()
cmd.iterate('bycalpha (%s)' % (selection),
'qkeys.add(((model,segi,chain,ss), resv))', space={'qkeys': qkeys})
def in_intervals(i, intervals):
for interval in intervals:
if interval[0] <= i <= interval[1]:
return True
return False
elements = dict()
for key, resv in qkeys:
element = elements.setdefault(key, [])
if in_intervals(resv, element):
continue
resv_set = set()
cmd.iterate('/%s/%s/%s//CA and ss "%s"' % key, 'resv_set.add(resv)',
space={'resv_set': resv_set})
resv_min = resv
resv_max = resv
while (resv_min - 1) in resv_set:
resv_min -= 1
while (resv_max + 1) in resv_set:
resv_max += 1
element.append((resv_min, resv_max))
sele_list = []
ss_names = {'S': 'Strand', 'H': 'Helix', '': 'Loop', 'L': 'Loop'}
for key in elements:
model,segi,chain,ss = key
for resv_min,resv_max in elements[key]:
sele = '/%s/%s/%s/%d-%d' % (model, segi, chain, resv_min, resv_max)
if caonly:
sele += '/CA'
sele_list.append(sele)
if not quiet:
print("%-6s %s" % (ss_names.get(ss, ss), sele))
sele = ' '.join(sele_list)
if name is not None:
cmd.select(name, sele)
elif not quiet:
print(' Selection: ' + sele)
return sele
def hydro_pairs(selection, cut_off):
"""
Find hydrogen bonds for a given selection
"""
states = cmd.count_states(selection)
if states:
hb = []
for i in range(1, states + 1):
p = cmd.find_pairs("%s and (name O* and neighbor elem H) or"
"(name N* and neighbor elem H)" % selection,
"%s and name N* or name O*" % selection,
cutoff=3.5,
mode=1,
angle=70,
state1=i,
state2=i)
hb.append(p)
if any([j for i in hb for j in i]):
seen = {}
for state in hb:
for pairs in state:
if pairs in seen:
seen[pairs] = seen[pairs] + 1
else:
seen[pairs] = 1
occurrence = seen.items()
occurrence = sorted(occurrence, key=lambda x: x[1], reverse=True)
fd = open("Hydrogen_bonds.dat", "w")
fd.write("--------------------------------------------------------"
"------------\n")
fd.write("-----------------------------Results--------------------"
"------------\n")
fd.write("--------------------------------------------------------"
"------------\n")
fd.write(" Donor |"
" Aceptor |\n")
fd.write(" Object Residue Atom Index|"
" Object Residue Atom Index| % occurrence\n")
stored.donors = []
stored.aceptors = []
sub = []
occurrence_list = []
for i in range(len(occurrence)):
cmd.iterate("index %s" % (occurrence[i][0][0][1]),
"stored.donors.append((resn, elem))")
cmd.iterate("index %s" % (occurrence[i][0][1][1]),
"stored.aceptors.append((resn, elem))")
if (occurrence[i][1] * 100 / states) >= cut_off:
fd.write("%8s%8s%6s%8s|%8s%8s%6s%8s|%5s\n" %
(occurrence[i][0][0][0],
stored.donors[i][0],
stored.donors[i][1],
occurrence[i][0][0][1],
occurrence[i][0][1][0],
stored.aceptors[i][0],
stored.aceptors[i][1],
occurrence[i][0][1][1],
"%.2f" % (occurrence[i][1] * 100 /states)
))
occurrence_list.append(occurrence[i][0])
fd.close()
cmd.set('suspend_updates', 'on')
for state, bonds in enumerate(hb):
for bond in bonds:
if bond in occurrence_list:
cmd.distance('HB',
'index %s' % bond[0][1],
'index %s' % bond[1][1],
state=state,
cutoff=3.5)
cmd.hide("labels", "HB")
cmd.set('suspend_updates', 'off')
print("Check working directory for hydrogen bonds text file report")
else:
print("No hydrogen bonds detected for this selection")
def test(self):
cmd.load("PYMOL-1514.mae")
b_list = []
cmd.iterate("all", "b_list.append(b)", space=locals())
self.assertArrayEqual(b_list, [-1.0, -1.0, -2.3, 2.86], 0.001)
def loadcd(
recid,
chain,
ligid,
ligname,
pocket,
flexdist="3",
crossdock="~/Documents/datasets/crossdocking/CrossDocked",
cdpath="",
):
# Clear everything
cmd.reinitialize("everything")
# Convert flexdist to float
flexdist = float(flexdist)
# Print informations
print(f"Loading {recid}_{chain}:{ligid}_{ligname}")
print(f"flexdist = {flexdist}")
# Build paths
name = f"{recid}_{chain}_rec_{ligid}_{ligname}"
path = os.path.join(cdpath, pocket)
ligname = f"{name}_lig.pdb"
ligandpath = os.path.join(path, ligname) # Docked ligand
flexname = f"{name}_flex.pdb"
flexpath = os.path.join(path, flexname) # Flexible residues
#receptorname = f"{recid}_{chain}__rec.pdb"
#receptorpath = os.path.join(crossdock, receptorname)
crystalname = f"{recid}_{chain}_rec.pdb"
crystalpath = os.path.join(crossdock, pocket,
crystalname) # Crystal receptor
# Load ligand and receptor
cmd.load(ligandpath, "ligand") # Selection name: ligand
cmd.load(flexpath, "flex") # Selection name: flex
#cmd.load(receptorpath, "receptor") # Selection name: receptor
cmd.load(crystalpath, "crystal") # Selection name: receptor
# Hide everything
cmd.hide("all")
# Show receptor
#cmd.show("cartoon", "receptor")
#cmd.color("grey", "receptor")
cmd.show("cartoon", "crystal")
cmd.color("grey", "crystal")
# Show docked ligand
cmd.show("sticks", "ligand")
cmd.show("spheres", "ligand")
cmd.set("sphere_scale", 0.2, "ligand")
cmd.color("marine", "ligand and name C*")
# Show flexible residues
cmd.show("licorice", "flex")
cmd.set("stick_radius", 0.2, "flex")
cmd.color("teal", "flex and name C*")
# Center and zoom to ligand
cmd.center("ligand")
cmd.zoom("ligand", 10)
# Remove solvent
#cmd.remove("solvent")
# Get residues of crystal close to flexible residues
stored.list = []
cmd.iterate(
"crystal within 0.1 of flex", # Selection
"stored.list.append((resn, resi, chain))", # Action
)
# Remove redundancies
flexres = set(stored.list) # Set of flexible residues
# Outline flexible residues of the receptor
#for _, resi, chain in flexres:
# if chain != "": # ???
# sel = f"receptor and (resi {resi} in chain {chain})"
# cmd.show("sphere", sel)
# cmd.set("sphere_scale", 0.2, sel)
# cmd.color("yellow", sel)
# cmd.remove(f"hydro in ({sel})")
# Outline flexible residues of the crystal
for _, resi, chain in flexres:
if chain != "": # ???
sel = f"crystal and (resi {resi} in chain {chain})"
cmd.show("licorice", sel)
cmd.set("stick_radius", 0.15, sel)
cmd.color("grey", sel)
cmd.remove(f"hydro in ({sel})")
# Show metal atoms
cmd.show("spheres", "metals")
cmd.set("sphere_scale", 0.5, "metals")
def mutate(selection,
new_resn,
inplace=0,
sculpt=0,
hydrogens='auto',
mode=0,
quiet=1):
'''
DESCRIPTION
Mutate a single residue. Does call the mutagenesis wizard non-interactively
and tries to select the best rotamer. Can do some sculpting in the end to
the best rotamer.
USAGE
mutate selection, new_resn [, inplace [, sculpt [, hydrogens]]]
ARGUMENTS
selection = string: atom selection of single residue
new_resn = string: new residue name (3-letter or 1-letter)
inplace = 0 or 1: work on copy of input object if 0 {default: 0}
sculpt = 0: no sculpting {default}
sculpt = 1: do sculpting on best rotamer
sculpt = 2: do sculpting with neighbors
hydrogens = string: keep, auto or none {default: auto}
mode = 0: select rotamer with best clash score {default}
mode = 1: take chi angles from original residue
mode = 2: first rotamer
EXAMPLE
fetch 2x19, async=0
select x, A/CYS`122/
zoom x
mutate x, LYS
'''
from pymol.wizard import mutagenesis
_assert_package_import()
from . import three_letter
inplace, sculpt = int(inplace), int(sculpt)
mode = int(mode)
quiet = int(quiet)
org = cmd.get_object_list(selection)[0]
tmp = cmd.get_unused_name()
new_resn = new_resn.upper()
new_resn = three_letter.get(new_resn, new_resn)
if inplace:
cpy = org
else:
cpy = cmd.get_unused_name(org + '_cpy')
cmd.create(cpy, org, -1, 1, zoom=0)
scr = []
cmd.iterate('first (%s)' % selection,
'scr[:] = (segi,chain,resi,resn)',
space={'scr': scr})
res = '/%s/%s/%s/%s' % tuple([cpy] + scr[:3])
if mode == 1:
old_resn = scr[3]
chi_atoms = {
'ALA': [],
'ARG': ['C', 'CA', 'CB', 'CG', 'CD', 'NE', 'CZ'],
'ASN': ['C', 'CA', 'CB', 'CG', 'OD1'],
'ASP': ['C', 'CA', 'CB', 'CG', 'OD1'],
'CYS': ['C', 'CA', 'CB', 'SG'],
'GLN': ['C', 'CA', 'CB', 'CG', 'CD', 'OE1'],
'GLU': ['C', 'CA', 'CB', 'CG', 'CD', 'OE1'],
'GLY': [],
'HIS': ['C', 'CA', 'CB', 'CG', 'ND1'],
'ILE': ['C', 'CA', 'CB', 'CG1', 'CD1'],
'LEU': ['C', 'CA', 'CB', 'CG', 'CD1'],
'LYS': ['C', 'CA', 'CB', 'CG', 'CD', 'CE', 'NZ'],
'MET': ['C', 'CA', 'CB', 'CG', 'SD', 'CE'],
'MSE': ['C', 'CA', 'CB', 'CG', 'SE', 'CE'],
'PHE': ['C', 'CA', 'CB', 'CG', 'CD1'],
'PRO': [],
'SER': ['C', 'CA', 'CB', 'OG'],
'THR': ['C', 'CA', 'CB', 'OG1'],
'TRP': ['C', 'CA', 'CB', 'CG', 'CD2'],
'TYR': ['C', 'CA', 'CB', 'CG', 'CD1'],
'VAL': ['C', 'CA', 'CB', 'CG2'],
}
atoms = [res + '/' + name for name in chi_atoms.get(old_resn, [])]
old_chi = []
for args in zip(atoms, atoms[1:], atoms[2:], atoms[3:]):
try:
old_chi.append(cmd.get_dihedral(*args))
except:
break
cmd.remove('%s and not name CA+C+N+O+OXT' % (res))
# start the wizard to count the number of rotamers for this residue
cmd.wizard("mutagenesis")
cmd.get_wizard().set_mode(new_resn)
cmd.get_wizard().set_hyd(hydrogens)
cmd.get_wizard().do_select("(" + res + ")")
def get_best_state_bump():
best_state = (1, 1e9)
cmd.create(tmp, '%s and not name CA+C+N+O or (%s within 8.0 of (%s and name CB))' % \
(mutagenesis.obj_name, cpy, mutagenesis.obj_name), zoom=0, singletons=1)
cmd.bond('name CB and %s in %s' % (tmp, mutagenesis.obj_name),
'name CA and %s in %s' % (tmp, res))
cmd.sculpt_activate(tmp)
for i in range(1, cmd.count_states(tmp) + 1):
score = cmd.sculpt_iterate(tmp, state=i)
if not quiet:
print('Frame %d Score %.2f' % (i, score))
if score < best_state[1]:
best_state = (i, score)
cmd.delete(tmp)
if not quiet:
print(' Best: Frame %d Score %.2f' % best_state)
return best_state
if cmd.count_states(mutagenesis.obj_name) < 2 or mode > 0:
best_state = (1, -1.0)
else:
best_state = get_best_state_bump()
cmd.frame(best_state[0])
cmd.get_wizard().apply()
cmd.wizard()
if mode == 1:
atoms = [res + '/' + name for name in chi_atoms.get(new_resn, [])]
for args in zip(atoms, atoms[1:], atoms[2:], atoms[3:], old_chi):
cmd.set_dihedral(*args)
cmd.unpick()
if sculpt > 0:
sculpt_relax(res, 0, sculpt == 2, cpy)
return cpy
with open(filename, 'r') as f:
lines = f.readlines()
lines = list(filter(filter_func, lines))
ang_pairs.extend(list(map(map_func, lines)))
restraints.extend(list(map(map_func2, lines)))
out_lines = []
for ang_pair, restraint in zip(ang_pairs, restraints):
group1 = "(id %d)" % ang_pair[0]
group2 = "(id %d)" % ang_pair[1]
group3 = "(id %d)" % ang_pair[2]
cmd.angle("%d-%d-%d" % ang_pair, group1, group2, group3)
resns = []
resis = []
atoms = []
cmd.iterate("id %d+%d+%d" % ang_pair, "resns.append(resn)")
cmd.iterate("id %d+%d+%d" % ang_pair, "resis.append(resi)")
cmd.iterate("id %d+%d+%d" % ang_pair, "atoms.append(name)")
angle = cmd.get_angle(group1, group2, group3)
out_lines.append(
"[ %s%s(%s)-%s%s(%s)-%s%s(%s) ~ %.2f degree (%s, %s, %s) ]\n" %
((resns[0], resis[0], atoms[0], resns[1], resis[1], atoms[1], resns[2],
resis[2], atoms[2], angle) + restraint))
out_lines.append("%d %d %d\n" % ang_pair)
# write dis_pairs to distance_pairs.ndx for analysis
with open("angle_out.ndx", 'w') as f:
for line in out_lines:
f.write(line)
print(line)
def muta():
'''
DESCRIPTION
Creates an alignment of two proteins and superimposes them.
Aligned residues that are different in the two (i.e. mutations) are highlighted and
colored according to their difference in the BLOSUM90 matrix.
Is meant to be used for similar proteins, e.g. close homologs or point mutants,
to visualize their differences.
USAGE
color_by_mutation selection1, selection2 [,waters [,labels ]]
ARGUMENTS
obj1: object or selection
obj2: object or selection
waters: bool (0 or 1). If 1, waters are included in the view, colored
differently for the both input structures.
default = 0
labels: bool (0 or 1). If 1, the possibly mutated sidechains are
labeled by their chain, name and id
default = 0
EXAMPLE
color_by_mutation protein1, protein2
SEE ALSO
super
'''
from pymol import stored, CmdException
obj1 = "native_na"
obj2 = "design_na"
waters = 0
labels = 0
if cmd.count_atoms(obj1) == 0:
print '%s is empty' % obj1
return
if cmd.count_atoms(obj2) == 0:
print '%s is empty' % obj2
return
waters = int(waters)
labels = int(labels)
# align the two proteins
aln = '__aln'
# first, an alignment with 0 cycles (no atoms are rejected, which maximized the number of aligned residues)
# for some mutations in the same protein this works fine). This is essentially done to get a
# sequence alignment
cmd.super(obj2, obj1, object=aln, cycles=0)
# superimpose the the object using the default parameters to get a slightly better superimposition,
# i.e. get the best structural alignment
cmd.super(obj2, obj1)
stored.resn1, stored.resn2 = [], []
stored.resi1, stored.resi2 = [], []
stored.chain1, stored.chain2 = [], []
# store residue ids, residue names and chains of aligned residues
cmd.iterate(obj1 + ' and name CA and ' + aln, 'stored.resn1.append(resn)')
cmd.iterate(obj2 + ' and name CA and ' + aln, 'stored.resn2.append(resn)')
cmd.iterate(obj1 + ' and name CA and ' + aln, 'stored.resi1.append(resi)')
cmd.iterate(obj2 + ' and name CA and ' + aln, 'stored.resi2.append(resi)')
cmd.iterate(obj1 + ' and name CA and ' + aln,
'stored.chain1.append(chain)')
cmd.iterate(obj2 + ' and name CA and ' + aln,
'stored.chain2.append(chain)')
mutant_selection = ''
non_mutant_selection = 'none or '
colors = []
# loop over the aligned residues
for n1, n2, i1, i2, c1, c2 in zip(stored.resn1, stored.resn2, stored.resi1,
stored.resi2, stored.chain1,
stored.chain2):
# take care of 'empty' chain names
if c1 == '':
c1 = '""'
if c2 == '':
c2 = '""'
if n1 == n2:
non_mutant_selection += '((%s and resi %s and chain %s) or (%s and resi %s and chain %s)) or ' % (
obj1, i1, c1, obj2, i2, c2)
else:
mutant_selection += '((%s and resi %s and chain %s) or (%s and resi %s and chain %s)) or ' % (
obj1, i1, c1, obj2, i2, c2)
# get the similarity (according to the blosum matrix) of the two residues and
c = getBlosum90ColorName(n1, n2)
colors.append(
(c, '%s and resi %s and chain %s and elem C' % (obj2, i2, c2)))
if mutant_selection == '':
print ' Error: No mutations found'
raise CmdException
# create selections
cmd.select('mutations', mutant_selection[:-4])
cmd.select('non_mutations', non_mutant_selection[:-4])
cmd.select(
'not_aligned',
'(%s or %s) and not mutations and not non_mutations' % (obj1, obj2))
# create the view and coloring
cmd.hide('everything', '%s or %s' % (obj1, obj1))
cmd.show('cartoon', '%s or %s' % (obj2, obj1))
cmd.show(
'lines',
'(%s or %s) and ((non_mutations or not_aligned) and not name c+o+n)' %
(obj2, obj1))
cmd.show('sticks',
'(%s or %s) and mutations and not name c+o+n' % (obj2, obj1))
cmd.color('white', 'elem C and non_mutations')
cmd.color('cyan', 'elem C and mutations and %s' % obj1)
cmd.color('gray', 'elem C and not_aligned')
cmd.color('limon', 'chainA')
cmd.util.cnc("all")
for (col, sel) in colors:
cmd.color(col, sel)
cmd.hide('everything', '(hydro) and (%s or %s)' % (obj2, obj1))
cmd.center('%s or %s' % (obj2, obj1))
if labels:
cmd.label('mutations and name CA', '"(%s-%s-%s)"%(chain, resi, resn)')
if waters:
cmd.set('sphere_scale', '0.1')
cmd.show('spheres', 'resn HOH and (%s or %s)' % (obj2, obj1))
cmd.color('red', 'resn HOH and %s' % obj1)
cmd.color('salmon', 'resn HOH and %s' % obj2)
print '''
Mutations are highlighted in blue and red.
All mutated sidechains of %s are colored blue, the corresponding ones from %s are
colored on a spectrum from blue to red according to how similar the two amino acids are
(as measured by the BLOSUM90 substitution matrix).
Aligned regions without mutations are colored white.
Regions not used for the alignment are gray.
NOTE: There could be mutations in the gray regions that were not detected.''' % (
obj2, obj1)
cmd.delete(aln)
cmd.deselect()
cmd.zoom("interface")
def peptide_rebuild(name, selection='all', cycles=1000, state=1, quiet=1):
'''
DESCRIPTION
Rebuild the peptide from selection. All atoms which are present in
selection will be kept fixed, while atoms missing in selection are
placed by sculpting.
USAGE
peptide_rebuild name [, selection [, cycles [, state ]]]
SEE ALSO
stub2ala, add_missing_atoms, peptide_rebuild_modeller
'''
from chempy import fragments, feedback, models
cycles, state, quiet = int(cycles), int(state), int(quiet)
# suppress feedback for model merging
feedback['actions'] = False
# work with named selection
namedsele = cmd.get_unused_name('_')
cmd.select(namedsele, '{} & present'.format(selection), 0)
identifiers = []
cmd.iterate(namedsele + ' and polymer and guide and alt +A',
'identifiers.append([segi,chain,resi,resv,resn])',
space=locals())
model = models.Indexed()
for (segi, chain, resi, resv, resn) in identifiers:
try:
fname = resn.lower() if resn != 'MSE' else 'met'
frag = fragments.get(fname)
except IOError:
print(' Warning: unknown residue: ' + resn)
continue
for a in frag.atom:
a.segi = segi
a.chain = chain
a.resi = resi
a.resi_number = resv
a.resn = resn
model.merge(frag)
if not quiet:
print(' Loading model...')
cmd.load_model(model, name, 1, zoom=0)
if cmd.get_setting_boolean('auto_remove_hydrogens'):
cmd.remove(name + ' and hydro')
cmd.protect(name + ' in ' + namedsele)
cmd.sculpt_activate(name)
cmd.update(name, namedsele, 1, state)
cmd.delete(namedsele)
if not quiet:
print(' Sculpting...')
cmd.set('sculpt_field_mask', 0x003, name) # bonds and angles only
cmd.sculpt_iterate(name, 1, int(cycles / 4))
cmd.set('sculpt_field_mask', 0x09F, name) # local + torsions
cmd.sculpt_iterate(name, 1, int(cycles / 4))
cmd.set('sculpt_field_mask', 0x0FF, name) # ... + vdw
cmd.sculpt_iterate(name, 1, int(cycles / 2))
cmd.sculpt_deactivate(name)
cmd.deprotect(name)
cmd.unset('sculpt_field_mask', name)
if not quiet:
print(' Connecting peptide...')
pairs = cmd.find_pairs(name + ' and name C', name + ' and name N', 1, 1,
2.0)
for pair in pairs:
cmd.bond(*pair)
cmd.h_fix(name)
if not quiet:
print(' peptide_rebuild: done')
def loadflexdock(
system,
dataset,
idx="1",
flexdist="3",
pdbbindpath="../../PDBbind18",
dockingpath="",
):
# Clear everything
cmd.reinitialize("everything")
# Convert string of indices to numbers
idx = int(idx)
# Convert flexdist to float
flexdist = float(flexdist)
# Print informations
print(f"Loading {dataset}/{system} (rank {idx})")
print(f"flexdist = {flexdist}")
# Build paths
ligname = f"{system}_ligand-{idx}.pdb"
ligandpath = os.path.join(dockingpath, dataset, system, ligname) # Docked ligand
flexname = f"{system}_flex-{idx}.pdb"
flexpath = os.path.join(dockingpath, dataset, system, flexname) # Flexible residues
receptorname = f"{system}_protein-{idx}.pdb"
receptorpath = os.path.join(dockingpath, dataset, system, receptorname)
crystalname = f"{system}_protein.pdb"
crystalpath = os.path.join(
pdbbindpath, dataset, system, crystalname
) # Crystal receptor
# Load ligand and receptor
cmd.load(ligandpath, "ligand") # Selection name: ligand
cmd.load(flexpath, "flex") # Selection name: flex
cmd.load(receptorpath, "receptor") # Selection name: receptor
cmd.load(crystalpath, "crystal") # Selection name: receptor
# Hide everything
cmd.hide("all")
# Show receptor
cmd.show("cartoon", "receptor")
cmd.color("grey", "receptor")
# Show docked ligand
cmd.show("sticks", "ligand")
cmd.show("spheres", "ligand")
cmd.set("sphere_scale", 0.2, "ligand")
cmd.color("marine", "ligand and name C*")
# Show flexible residues
cmd.show("licorice", "flex")
cmd.set("stick_radius", 0.2, "flex")
cmd.color("teal", "flex and name C*")
# Center and zoom to ligand
cmd.center("ligand")
cmd.zoom("ligand", 10)
# Remove solvent
cmd.remove("solvent")
# Get residues of receptor close to flexible residues
stored.list = []
cmd.iterate(
"receptor within 0.1 of flex", # Selection
"stored.list.append((resn, resi, chain))", # Action
)
# Remove redundancies
flexres = set(stored.list) # Set of flexible residues
# Outline flexible residues of the receptor
for _, resi, chain in flexres:
if chain != "": # ???
sel = f"receptor and (resi {resi} in chain {chain})"
cmd.show("sphere", sel)
cmd.set("sphere_scale", 0.2, sel)
cmd.color("yellow", sel)
cmd.remove(f"hydro in ({sel})")
# Outline flexible residues of the crystal
for _, resi, chain in flexres:
if chain != "": # ???
sel = f"crystal and (resi {resi} in chain {chain})"
cmd.show("licorice", sel)
cmd.set("stick_radius", 0.15, sel)
cmd.color("deeppurple", sel)
cmd.remove(f"hydro in ({sel})")
# Show metal atoms
cmd.show("spheres", "metals")
cmd.set("sphere_scale", 0.5, "metals")
