def _pre_exec(self):
if self.occupancy_one:
occ_file = "occupancy_one.pdb"
npdb = NumPdb(self.pdb_input)
npdb['occupancy'] = 1.0
npdb.write(occ_file)
self.pdb_input = self.outpath(occ_file)
def make_dry_pdb(self):
envelope = self.msms0.envelope_msms.get_vert(filt=False)
envelope = [(x['x'], x['y'], x['z']) for x in envelope]
envelope_tree = scipy.spatial.KDTree(envelope)
npdb_non = NumPdb(self.no_water_file, features=self.npdb_features)
non_tree = scipy.spatial.KDTree(npdb_non['xyz'])
pr2 = self.probe_radius * 2
pr3 = self.probe_radius * 3
wet_pdb = [(self.dowser.dowser_file, self.dowser_dry_pdb),
(self.processed_pdb, self.original_dry_pdb)]
for pdb_file, out_file in wet_pdb:
npdb = NumPdb(pdb_file, features=self.npdb_features)
sele_het = npdb.sele(record="HETATM")
sele_not_wat = npdb.sele(resname="HOH", invert=True)
hetero_tree = get_tree(
npdb.get('xyz', sele=(sele_het & sele_not_wat)))
sele = npdb.sele()
i = 0
for numa in npdb.iter_resno(incomplete=True):
flag = True
if numa[0]['resname'] == 'HOH':
# assume the first water atom is the oxygen
dist_env = envelope_tree.query(numa['xyz'][0])[0]
dist_het = hetero_tree.query(numa['xyz'][0])[0]
dist_non = non_tree.query(numa['xyz'][0])[0]
# TODO het cutoff only for dowser_file
if dist_env < pr2 or dist_het < 3.9 or dist_non > pr3:
flag = False
for a in numa._atoms:
sele[i] = flag
i += 1
npdb.copy(sele=sele).write2(out_file)
def _get_ca (self,pdb_file,chain, residue):
npdb=NumPdb( pdb_file )
sele1={"resno": self.residue, "chain": self.chain}
resname1=npdb.get('resname', **sele1)[0]
sele={"resno": residue, "chain": chain, "resname": resname1 ,"atomname":'CA'}
cacoord=npdb.get('xyz',**sele)
return cacoord
def get_npdb_dicts(self):
mplanes = np.array(self.opm.get_planes())
dist = abs(point_plane_dist(mplanes[1][0], mplanes[0]))
npdb_dict = {}
npdb_tm_dict = {}
npdb_sol_dict = {}
for suffix, pdb_file in self.water_variants:
npdb = NumPdb(pdb_file, features=self.npdb_features)
npdb_dict[suffix] = npdb
sele = npdb.sele()
i = 0
# create transmembrane region selection
for numa in npdb.iter_resno(incomplete=True):
flag = True
for c in numa._coords:
d1 = abs(point_plane_dist(c, mplanes[0]))
d2 = abs(point_plane_dist(c, mplanes[1]))
if d1 < dist and d2 < dist:
flag = False
break
for a in numa._atoms:
sele[i] = flag
i += 1
npdb_sol_dict[suffix] = npdb.copy(sele=sele)
npdb.copy(sele=sele).write(self.outpath("sol_region.pdb"))
np.logical_not(sele, sele)
npdb_tm_dict[suffix] = npdb.copy(sele=sele)
npdb.copy(sele=sele).write(self.outpath("tm_region.pdb"))
return npdb_dict, npdb_tm_dict, npdb_sol_dict
def _post_exec(self):
print('TEST 1')
print(self.relpath(self.dowser_file))
if '++' in self.relpath(self.dowser_file):
print('TEST !!!')
print('Dowser++ not looked at!')
return None
self._make_provi_file(input_file=self.relpath(self.input_file),
wat_file=self.relpath(self.wat_file),
watall_file=self.relpath(self.watall_file),
intsurf_file=self.relpath(self.intsurf_file))
# make sure all file exist
for f in [self.wat_file, self.watall_file, self.intsurf_file]:
with open(f, "a"):
pass
# write a pdb with all dowser waters but no others
with open(self.dowser_file, "w") as fp:
with open(self.pdb_file, "r") as fp_pdb:
for line in fp_pdb:
if (line[0:6] in ["ATOM ", "HETATM"]
and line[17:20] != "HOH"):
fp.write(line)
with open(self.wat_file, "r") as fp_wat:
for line in fp_wat:
if line[0:6] in ["HETATM"]:
fp.write(line)
# rename HOH.OW atoms to HOH.O and
# repair atom index
npdb = NumPdb(self.dowser_file)
for i, a in enumerate(npdb._atoms, start=1):
if a["atomname"] == " OW " and a["resname"] == "HOH":
a["atomname"] = " O "
a["atomno"] = i
npdb.write(self.dowser_file)
def _crop_pdb( self ):
npdb = NumPdb( self.pdb_file )
npdb.write(
self.cropped_pdb,
chain=self.res1["chain"],
resno=[ self.res1["resno"] + 1, self.res2["resno"] - 1 ],
invert=True
)
def _pre_exec(self):
self.pdb_assembly()
# extract first model
npdb = NumPdb(self.pdb_assembly.assembly_file,
features={
"phi_psi": False,
"info": False,
})
npdb.write2(self.assembly_file)
def _pre_exec( self ):
backbone = ( ' N ',' C ', ' CA ',' O ' )
npdb=NumPdb( self.pdb_file )
back=npdb.sele(atomname=backbone)
npdb.copy(sele=back).write('nobb.pdb')
self._make_script_file(
map_name=self.relpath(self.map_file, no_ext=True) ,
pixelsize=self.pixelsize,
pdb_file='nobb.pdb'
)
def func( self ):
self.delete_backbone()
npdb=NumPdb( self.pdb_file )
for i, numa in enumerate( npdb.iter_resno() ):
sele={'resno':i+1}
resname1=numa.get('resname') [0]
resno1=numa.get('resno') [0]
chain1=numa.get('chain') [0]
dirg="%s_%s_%i" %(chain1,resname1,resno1)
print resname1, resno1, chain1
if numa.get('resname') [0] not in ('ALA', 'GLY'):
if resname1 in ('SER','CYS','PRO','ASP','THR','ASN','VAL','GLU','GLN'):
SpiderSidechainCorrelation(self.map_file, self.pdb_file, self.pixelsize,self.resolution, resno1, chain1,output_dir=self.subdir(dirg))
else:
SpiderSidechainCorrelation('deletebb.cpv', self.pdb_file, self.pixelsize,self.resolution, resno1, chain1,output_dir=self.subdir(dirg))
aaname="%s.%s" %(dirg,'pdb')
print aaname
aa=npdb.sele(resname=resname1,resno=resno1)
dire="%s/%s" %(dirg,aaname)
npdb.copy(sele=aa).write(dire)
OriSidechainCorrel('deletebb.cpv',dire,self.pixelsize,self.resolution,resno1,chain1,output_dir=self.subdir(dirg))
else:
aaname="%s.%s" %(dirg,'pdb')
print aaname
aa=npdb.sele(resname=resname1,resno=resno1)
dire="%s/%s" %(dirg,aaname)
if not os.path.exists(self.subdir(dirg)): os.makedirs(self.subdir(dirg))
npdb.copy(sele=aa).write(dire)
def _post_exec(self):
npdb = NumPdb(self.pdb_file,
features={
"sstruc": False,
"phi_psi": False,
"extra": [("dssp", "|S1", " ")]
})
self.get_numpdb_sstruc(npdb)
def _pre_exec( self ):
npdb=NumPdb( self.pdb_file )
sele={"resno": self.residue, "chain": self.chain}
resname1=npdb.get('resname', **sele)[0]
num_rota=get_rotno(resname1)
coords=self._get_ca(self.pdb_file,self.chain, self.residue)
print "huh" ,coords [0] [0]
x_ca= coords[0] [0]
y_ca=coords [0][1]
z_ca=coords [0] [2]
self._make_script_file(
x_ca= x_ca,
y_ca=y_ca,
z_ca=z_ca,
map_name=self.relpath(self.map_file, no_ext=True) ,
pixelsize=self.pixelsize,
resolution=self.resolution,
pdb_file=self.relpath(self.pdb_file, no_ext=False) ,
)
def get_npdb_dicts(self):
npdb_dict = {}
npdb_tm_dict = {}
npdb_sol_dict = {}
for suffix, pdb_file in self.water_variants:
npdb = NumPdb(pdb_file, features=self.npdb_features)
npdb_dict[suffix] = npdb
sele = npdb.sele()
i = 0
# create transmembrane region selection
for numa in npdb.iter_resno(incomplete=True):
flag = True
for a in numa._atoms:
sele[i] = flag
i += 1
npdb_sol_dict[suffix] = npdb.copy(sele=sele)
npdb.copy(sele=sele).write(self.outpath("sol_region.pdb"))
np.logical_not(sele, sele)
npdb_tm_dict[suffix] = npdb.copy(sele=sele)
npdb.copy(sele=sele).write(self.outpath("tm_region.pdb"))
return npdb_dict, npdb_tm_dict, npdb_sol_dict
def func (self):
self.bestbuild()
npdb=NumPdb( "bestrotamers.pdb" )
a=True
z=2
#while a==True:
clashes,a=find_all_clashes(npdb)
gtree=get_tree(npdb['xyz'])
print clashes
for i in clashes:
localresi=[]
#hole CA atom vom ersten clashpartner
r=npdb.sele(resno=i[0],atomname='CA')
p=npdb.get('xyz',sele=r)
#indices im 6 A umkreis um den clash
l=gtree.query_ball_point(p,6)
print l[0]
#aus den indices eine liste mit residuname und nummer machen
for x in l[0]:
a=npdb.get('resname')[x]
b=npdb.get('resno')[x]
c=npdb.get('chain')[x]
localresi.append([b,a,c])
localresi=sorted(localresi)
localresi=list(localresi for localresi,_ in itertools.groupby(localresi))
#print localresi
for index,i in enumerate (localresi):
if i[1] not in ('ALA','GLY'):
rotadir="%s_%s_%i" % (i[2],i[1],i[0])
ccsortpath=os.path.join(self.result_direc,rotadir)
ccsort="%s/%s" % (ccsortpath,'ccsort.cpv')
print ccsortpath
break
def func(self):
if self.check_only:
return
def do(name):
if not self.tools:
return True
if self.tools[0] == "!":
return name not in self.tools[1:]
else:
return name in self.tools
if not self.analyze_only:
if do("pdb_info"):
self.pdb_info()
if do("info"):
self.make_info()
if do("provi"):
npdb = NumPdb(self.pdb_input, features=self.npdb_features)
# npdb = NumPdb( self.final_pdb, features=self.npdb_features )
self._make_provi_file(
pdb_id=self.pdb_id,
pdb_title=self.pdb_info.get_info()["title"],
pdb_file=self.relpath(self.pdb_input),
# pdb_file=self.relpath( self.final_pdb ),
# pdb_org_file=self.relpath( self.original_dry_pdb ),
# mplane_file=self.relpath( self.opm.mplane_file ),
## hbx_file=self.relpath( self.hbexplore_fin.hbx_file ),
vol_file=self.relpath(self.voronoia_ori.vol_file),
# vol_file=self.relpath( self.voronoia_fin.vol_file ),
## msms_components=self.msms_vdw_fin.components_provi(
## color="lightgreen", translucent=0.5,
## relpath=self.relpath, max_atomno=len( npdb )
## ),
)
for suffix, pdb_file in self.water_variants:
for prefix, tool, tool_kwargs in self.do_tool_list:
name = "%s_%s" % (prefix, suffix)
self.__dict__[name]()
if do("stats"):
self.make_stats()
#if do( "stats2" ):
# self.make_stats2()
self.records = self.make_records()
self.write()
def make_final_pdb(self):
npdb_dow = NumPdb(self.dowser_dry_pdb, features=self.npdb_features)
npdb_org = NumPdb(self.original_dry_pdb, features=self.npdb_features)
dow_tree = get_tree(npdb_dow.get('xyz', resname="HOH"))
sele = npdb_org.sele()
i = 0
for numa in npdb_org.iter_resno(incomplete=True):
flag = False
if numa[0]['resname'] == 'HOH':
dist = dow_tree.query(numa['xyz'][0])[0]
if dist > 2.7:
flag = True
for a in numa._atoms:
sele[i] = flag
i += 1
coords_dow, atoms_dow = npdb_dow._select()
coords_org, atoms_org = npdb_org._select(sele=sele)
npdb_final = NumAtoms(np.hstack((atoms_dow, atoms_org)),
np.vstack((coords_dow, coords_org)))
npdb_final.write2(self.final_pdb)
def _pre_exec( self ):
npdb=NumPdb( self.pdb_file )
sele={"resno": self.residue, "chain": self.chain}
resname1=npdb.get('resname', **sele)[0]
num_rota=get_rotno(resname1)
coords=self._get_ca(self.pdb_file,self.chain, self.residue)
print "huh" ,coords [0] [0]
x_ca= coords[0] [0]
y_ca=coords [0][1]
z_ca=coords [0] [2]
self._make_script_file(
resolution=self.resolution,
pixelsize=self.pixelsize,
map_name=self.relpath( self.map_file, no_ext=True ),
x_ca=x_ca ,
y_ca=y_ca,
z_ca=z_ca,
cc_dir=self.relpath( self.sidechain_dir ) + os.sep,
num_rota=num_rota,
resname1=self.relpath (self.sidechain_dir)+ os.sep+resname1,
residue=self.residue
)
MakeAllRotameres(self.pdb_file, self.chain,self.residue,zfill=2,output_dir=self.sidechain_dir)
def _pre_exec (self):
npdb = NumPdb( self.pdb_file )
psf=1/self.pixelsize
schrumpf=npdb['xyz']*psf
#te=
ma=np.amax(schrumpf, axis=0)
mi=np.amin(schrumpf,axis=0)
middle=[(mi[0]+ma[0])/2,(mi[1]+ma[1])/2,(mi[2]+ma[2])/2]
abstand=round(((mi[0]-ma[0])**2+(mi[1]-ma[1])**2+(mi[2]-ma[2])**2)**0.5,0)
print ma,mi,middle,abstand
self._make_script_file(
x1=middle[0],
y1=middle[1],
z1=middle[2],
map_name=self.relpath(self.map_file, no_ext=True),
abstand=abstand,
ps=self.pixelsize,
)
def func(self):
if self.check_only:
return
def do(name):
if not self.tools:
return True
if self.tools[0] == "!":
return name not in self.tools[1:]
else:
return name in self.tools
if not self.analyze_only:
if do("pdb_info"):
self.pdb_info()
if do("info"):
self.make_info()
if do("provi"):
npdb = NumPdb(self.pdb_input, features=self.npdb_features)
self._make_provi_file(
pdb_id=self.pdb_id,
pdb_title=self.pdb_info.get_info()["title"],
pdb_file=self.relpath(self.pdb_input),
vol_file=self.relpath(self.voronoia_ori.vol_file),
)
for suffix, pdb_file in self.water_variants:
for prefix, tool, tool_kwargs in self.do_tool_list:
name = "%s_%s" % (prefix, suffix)
self.__dict__[name]()
if do("stats"):
self.make_stats()
self.records = self.make_records()
self.write()
def func (self):
npdb=NumPdb( self.pdb_file )
clashes,a=find_all_clashes(npdb)
gtree=get_tree(npdb['xyz'])
lclashes=[]
print clashes
resi=range(self.residue_1,self.residue_2)
print resi
for i in clashes:
print i[0], i[1]
if (i[0] or i[1]) in resi:
lclashes.append(i)
else:
continue
for x in lclashes:
print x
as1=x[0]
as2=x[1]
resname100=npdb.get('resname',resno=as1)[0]
resname101=npdb.get('resname',resno=as2)[0]
# #print resname100, resname101, i
if (resname100 and resname101) not in ('ALA','GLY'):
chain100=npdb.get('chain',resno=as1)[0]
chain101=npdb.get('chain',resno=as2)[0]
#print resname101
resno1=get_rotno(resname100)
resno2=get_rotno(resname101)
if resno1 >= resno2 :
# #print "anzahl rota" ,resno1,resno2
sele={'resno':as1}
atom_no= npdb.index(**sele)
# #erster clashpartner
rotadir="%s_%s_%i" % (chain100,resname100,as1)
ccsortpath=os.path.join(self.result_direc,rotadir)
ccsort="%s/%s" % (ccsortpath,'ccsort.cpv')
else:
sele={'resno':as2}
atom_no= npdb.index(**sele)
rotadir="%s_%s_%i" % (chain101,resname101,as2)
ccsortpath=os.path.join(self.result_direc,rotadir)
ccsort="%s/%s" % (ccsortpath,'ccsort.cpv')
def make_processed_pdb(self):
npdb = NumPdb(self.opm.processed_file, features=self.npdb_features)
sele = npdb.sele()
coords_dict = {}
i = 0
tree = scipy.spatial.KDTree(npdb['xyz'])
for numa in npdb._iter_resno():
for c in numa._coords:
c = tuple(c)
# remove atoms with identical coords
if c in coords_dict:
print npdb._atoms[i]
sele[i] = False
else:
coords_dict[c] = True
sele[i] = True
# remove atoms with almost identical coords
rslt = tree.query(c, k=10, distance_upper_bound=0.2)
rslt[1].sort()
if rslt[1][0] != i and rslt[0][1] != np.inf:
print i, npdb._atoms[i]
sele[i] = False
i += 1
npdb.copy(sele=sele).write2(self.processed_pdb)
def func(self):
if self.check_only:
return
def do(name):
if not self.tools:
return True
if self.tools[0] == "!":
return name not in self.tools[1:]
else:
return name in self.tools
if not self.analyze_only:
if do("opm_info"):
self.opm_info()
if do("opm"):
self.opm()
if do("proc"):
self.make_processed_pdb()
self.make_nowat_pdb()
if do("dowser"):
self.dowser()
if (self.dowserplus2):
self.dowserPlus2()
if do("msms0"):
self.msms0()
if do("dry"):
self.make_dry_pdb()
if do("final"):
self.make_final_pdb()
if do("dssp"):
self.dssp()
if do("pdb_info"):
self.pdb_info()
if do("mpstruc_info"):
self.mpstruc_download()
self.mpstruc_info()
if do("info"):
self.make_info()
if do("provi"):
npdb = NumPdb(self.final_pdb, features=self.npdb_features)
self._make_provi_file(
pdb_id=self.pdb_id,
pdb_title=self.pdb_info.get_info()["title"],
pdb_file=self.relpath(self.final_pdb),
pdb_org_file=self.relpath(self.original_dry_pdb),
mplane_file=self.relpath(self.opm.mplane_file),
hbx_file=self.relpath(self.hbexplore_fin.hbx_file),
vol_file=self.relpath(self.voronoia_fin.vol_file),
msms_components=""
'''self.msms_vdw_fin.components_provi(
color="lightgreen", translucent=0.5,
relpath=self.relpath, max_atomno=len( npdb )
),''')
for suffix, pdb_file in self.water_variants:
for prefix, tool, tool_kwargs in self.do_tool_list:
name = "%s_%s" % (prefix, suffix)
self.__dict__[name]()
if do("stats"):
self.make_stats()
if do("stats2"):
self.make_stats2()
self.records = self.make_records()
self.write()
def func( self ):
tmdet_xml = tmdet( self.pdb_file )
if not tmdet_xml:
tmdet_xml=" "
with open( self.tmdet_file, "w" ) as fp:
fp.write( tmdet_xml )
with open(self.tmdet_file, 'r') as fp:
text = fp.read()
if text != " ":
O = []
N = []
s = text
s = s.replace("\n", " ")
result = re.findall('<CHAIN(.*?)TYPE="alpha"', s)
if result:
chain = re.findall('CHAINID="(.*?)"', result[0])
regiontext = re.findall('TYPE="alpha"(.*?)</CHAIN>',s)
regions = re.findall('<REGION (.*?)type="H"/>', regiontext[0])
found = []
for reg in regions:
regio = reg.split("REGION")[-1]
pdb_beg = re.findall('pdb_beg="(.*?)"', regio)
pdb_end = re.findall('pdb_end="(.*?)"', regio)
found.append([ int(pdb_beg[0])-1, int(pdb_end[0])+1 ])
for index, elem in enumerate(found):
N.append(elem[index%2])
O.append(elem[(index+1)%2])
npdb = NumPdb( self.pdb_file, features={
"phi_psi": False, "sstruc": False, "backbone_only": True
})
sele = {"chain": chain[-1]}
with open(self.ply_file, 'w') as fp2:
fp2.write( "ply\n"+\
"format ascii 1.0\n"+\
"element vertex 8\n"+\
"property float x\n"+\
"property float y\n"+\
"property float z\n"+\
"property uchar red\n"+\
"property uchar green\n"+\
"property uchar blue\n"+\
"element face 4\n"+\
"property list uchar int vertex_indices\n"+\
"property uchar red\n"+\
"property uchar green\n"+\
"property uchar blue\n"+\
"end_header\n")
for val in [N, O]:
alln = []
if val == N:
sele["atomname"] = "C"
elif val == O:
sele["atomname"] = "N"
else:
print 'error'
for elem in val:
sele["resno"] = elem
coords = npdb.get( 'xyz', **sele )
if coords.size != 0:
alln.append(coords[0].tolist())
point, normall = planeFit(np.array(alln).transpose())
distance = scipy.spatial.distance.cdist(alln, [point])
if val == N:
normal = normall
distances = distance
d = -point.dot(normal)
maxdist = max(distances)[0]+2
xx = np.array([point[0]+maxdist, point[0]+maxdist, point[0]-maxdist, point[0]-maxdist])
yy = np.array([point[1]+maxdist, point[1]-maxdist, point[1]+maxdist, point[1]-maxdist])
z = (-normal[0] * xx - normal[1] * yy - d) * 1. /normal[2]
fp2.write( str( xx[0] )+" "+str( yy[0] )+" "+str( z[0] )+" 255\t0\t0\n")
fp2.write( str( xx[1] )+" "+str( yy[1] )+" "+str( z[1] )+" 255\t0\t0\n")
fp2.write( str( xx[2] )+" "+str( yy[2] )+" "+str( z[2] )+" 255\t0\t0\n")
fp2.write( str( xx[3] )+" "+str( yy[3] )+" "+str( z[3] )+" 255\t0\t0\n")
fp2.write("3 0 1 2 255\t0\t0\n")
fp2.write("3 1 2 3 255\t0\t0\n")
fp2.write("3 4 5 6 255\t0\t0\n")
fp2.write("3 5 6 7 255\t0\t0\n")
def _get_coords( self, pdb_file, res1, res2 ):
npdb = NumPdb( pdb_file )
return npdb.center( **res1 ), npdb.center( **res2 )
