def getAromaticRes(receptor):
# TODO add acids and amides?
aromatic_res = { "PHE": ["A"],
"TYR": ["A"],
"HIS": ["A", "N", "NA"],
"TRP": ["A", "N", "NA"],
"ARG": ["N", "C" , "NA"],
"DA" : ["A", "NA", "N"], # DNA
"DC" : ["A", "N", "NA"],
"DT" : ["A", "N", "NA"],
"DG" : ["A", "N", "NA"],
"A" : ["A", "NA", "N"], # DNA/RNA
"C" : ["A", "N", "NA"],
"G" : ["A", "N", "NA"],
"T" : ["A", "N", "NA"],
"U" : ["A", "N", "NA"],
}
residues = {}
special = [] # must be treated as a ligand part
for a in receptor:
a = a.strip()
rtype = a[16:21].strip().upper() # NOTE: we can recognize only all caps!
if a.startswith("HETATM"):
special.append(a)
elif a.startswith('ATOM') and rtype in aromatic_res.keys():
if hf.getAtype(a) in aromatic_res[rtype]:
res_id = hf.pmvAtomStrip(a).rsplit(":",1)[0] # A:GLY48:N -> A:GLY48
if not res_id in residues.keys():
residues[res_id] = []
residues[res_id].append(a)
if special and DEBUG:
print "found potential co-factor: %d" % len(special)
for a in special: print a
return residues, special # residues = { "A:GLY48" : [ atom, atom, atom..]}
def normalizeWaterScore(self): #pose,entropyPenalty=False):
# keep track of the original energy values
if self.debug:
old_e = self.pose['energy']
old_leff = self.pose['leff']
# rec_displaced waters
rec_displaced = 0
for p in self.pose['water_over_rec_penalty']:
rec_displaced += p
# print p # XXX
#print "RECDISP", rec_displaced # XXX
# lig_displaced waters
lig_displaced = 0
for p in self.pose['water_over_lig_penalty']:
lig_displaced += p
# clustering-displaced
clust_displaced = 0
for p in self.pose['water_cluster_penalty']:
clust_displaced += p
# count heavy at for lig.efficiency
heavy_atoms = 0
for a in self.pose['text']:
if hf.isAtom(a) and (not hf.getAtype(a) in ['HD', 'W']):
heavy_atoms += 1
# energy correction
if self.conservedwaterentropy:
# XXX read the penalty from a variable!
entropy = (0.2) * len(self.pose['water_bridge'])
#print " ENTROPY PENALTY = ", entropy
else:
entropy = 0
self.pose[
'energy'] += lig_displaced + rec_displaced + clust_displaced + entropy
# lig.efficiency correction
self.pose['leff'] = self.pose['energy'] / float(heavy_atoms)
if self.debug:
#print "\n# PENALTY:\t +%2.3f [rec: %2.3f, lig: %2.3f, clust: %2.3f ]" % (rec_displaced+lig_displaced+clust_displaced,
# rec_displaced, lig_displaced, clust_displaced)
print " Penalties:"
print " - lig overlap : %2.3f" % lig_displaced
print " - rec overlap : %2.3f" % rec_displaced
print " - cluster penalty : %2.3f" % clust_displaced
print " - extra entropy : %2.3f" % entropy
print " ENERGY:\t %2.3f [ old: %2.3f ]" % (self.pose['energy'],
old_e)
print " L.EFF.:\t %2.3f [ old: %2.3f ]\n" % (self.pose['leff'],
old_leff)
def weakWaterCleanup(self): #
#pose, cutoff_weak = -0.35, cutoff_strong=-0.5, interact_dist=3.5 ):
""" clean-up extra_weak waters (increment ['water_rec_over_penalty'], if necessary"""
org_waters = len(self.pose['water_bridge'])
#ignore_types = ['HD']
lone_water = []
interact_dist = self.interact_dist**2
hb_types = ['OA', 'NA', 'SA', 'N', 'O', 'OA', 'NA', 'SA']
hb_types += hf.METALS
for idx in range(len(self.pose['water_bridge'])):
water = self.pose['water_bridge'][idx]
score = self.pose['water_bridge_scores'][idx]
# analyze only waters that are not ranked strong
if score >= self.cutoff_strong:
atom_pool = [ a for a in self.pose['water_bridge_contacts'][idx] if hf.getAtype(a) in hb_types]
if self.debug:
hf.writeList('ATOM_POOL_%d_HB_%d.pdb' % (idx, sess), atom_pool, addNewLine=True)
lone_water.append(water)
try:
for a in atom_pool:
if (dist(water, a, sq=0)<= interact_dist):
#print "FOUND MATE",
#print dist(water, a, sq=True)
#print a.strip()
#print "------------------"
raise self.qs
#print "NO MATE FOR THIS", score
#print water.strip()
except:
lone_water.remove(water)
for w in lone_water:
idx = self.pose['water_bridge'].index(w)
del self.pose['water_bridge'][idx]
del self.pose['water_bridge_contacts'][idx]
# EXPERIMENTAL
#w_score = self.pose['water_bridge_scores'][idx]
#if w_score >
# XXX THIS shouldn't happen:
# a weakly bound water could be hanging in the bulk, but it should be accounted for
# ...?
# how to describe the bulk solvent?
# combine desolv map + discrete waters?
#
self.pose['water_over_rec_penalty'].append( -self.pose['water_bridge_scores'].pop(idx) )
# update PDBQT text
#print pose['water_over_rec_penalty'] # XXX
text_idx = self.pose['text'].index(w)
del self.pose['text'][text_idx]
if self.debug:
print "LONE WATERS # REMOVED: ", (org_waters- len(self.pose['water_bridge']) )
def weakWaterCleanup(self): #
#pose, cutoff_weak = -0.35, cutoff_strong=-0.5, interact_dist=3.5 ):
""" clean-up extra_weak waters (increment ['water_rec_over_penalty'], if necessary"""
org_waters = len(self.pose['water_bridge'])
#ignore_types = ['HD']
lone_water = []
interact_dist = self.interact_dist**2
hb_types = ['OA', 'NA', 'SA', 'N', 'O', 'OA', 'NA', 'SA']
hb_types += hf.METALS
for idx in range(len(self.pose['water_bridge'])):
water = self.pose['water_bridge'][idx]
score = self.pose['water_bridge_scores'][idx]
# analyze only waters that are not ranked strong
if score >= self.cutoff_strong:
atom_pool = [ a for a in self.pose['water_bridge_contacts'][idx] if hf.getAtype(a) in hb_types]
if self.debug:
hf.writeList('ATOM_POOL_%d_HB_%d.pdb' % (idx, sess), atom_pool, addNewLine=True)
lone_water.append(water)
try:
for a in atom_pool:
if (dist(water, a, sq=0)<= interact_dist):
#print "FOUND MATE",
#print dist(water, a, sq=True)
#print a.strip()
#print "------------------"
raise self.qs
#print "NO MATE FOR THIS", score
#print water.strip()
except:
lone_water.remove(water)
for w in lone_water:
idx = self.pose['water_bridge'].index(w)
del self.pose['water_bridge'][idx]
del self.pose['water_bridge_contacts'][idx]
# EXPERIMENTAL
#w_score = self.pose['water_bridge_scores'][idx]
#if w_score >
# XXX THIS shouldn't happen:
# a weakly bound water could be hanging in the bulk, but it should be accounted for
# ...?
# how to describe the bulk solvent?
# combine desolv map + discrete waters?
#
self.pose['water_over_rec_penalty'].append( -self.pose['water_bridge_scores'].pop(idx) )
# update PDBQT text
#print pose['water_over_rec_penalty'] # XXX
text_idx = self.pose['text'].index(w)
del self.pose['text'][text_idx]
if self.debug:
print "LONE WATERS # REMOVED: ", (org_waters- len(self.pose['water_bridge']) )
def normalizeWaterScore(self): #pose,entropyPenalty=False):
# keep track of the original energy values
if self.debug:
old_e = self.pose['energy']
old_leff = self.pose['leff']
# rec_displaced waters
rec_displaced = 0
for p in self.pose['water_over_rec_penalty']:
rec_displaced += p
# print p # XXX
#print "RECDISP", rec_displaced # XXX
# lig_displaced waters
lig_displaced = 0
for p in self.pose['water_over_lig_penalty']:
lig_displaced += p
# clustering-displaced
clust_displaced = 0
for p in self.pose['water_cluster_penalty']:
clust_displaced += p
# count heavy at for lig.efficiency
heavy_atoms = 0
for a in self.pose['text']:
if hf.isAtom(a) and ( not hf.getAtype(a) in ['HD', 'W'] ):
heavy_atoms += 1
# energy correction
if self.conservedwaterentropy:
# XXX read the penalty from a variable!
entropy = (0.2)*len(self.pose['water_bridge'])
#print " ENTROPY PENALTY = ", entropy
else:
entropy = 0
self.pose['energy'] += lig_displaced + rec_displaced + clust_displaced + entropy
# lig.efficiency correction
self.pose['leff'] = self.pose['energy'] / float( heavy_atoms )
if self.debug:
#print "\n# PENALTY:\t +%2.3f [rec: %2.3f, lig: %2.3f, clust: %2.3f ]" % (rec_displaced+lig_displaced+clust_displaced,
# rec_displaced, lig_displaced, clust_displaced)
print " Penalties:"
print " - lig overlap : %2.3f" % lig_displaced
print " - rec overlap : %2.3f" % rec_displaced
print " - cluster penalty : %2.3f" % clust_displaced
print " - extra entropy : %2.3f" % entropy
print " ENERGY:\t %2.3f [ old: %2.3f ]" % (self.pose['energy'], old_e)
print " L.EFF.:\t %2.3f [ old: %2.3f ]\n" % (self.pose['leff'], old_leff)
def recOverlapWaters(self): # pose, cutoff = CLASH,
#ignore_types = ['HD', 'H', 'W'] ):
""" INPUT : pdb lig (multi)model and pdb target model
OUTPUT: cleaned pdb lig model
"""
#print "WARNING! missing interactions! Line B1"
overlapping = []
atom_pool = []
cutoff = self.clash**2
#cutoff += 0.5
#cutoff = 0
#for a in self.pose['vdw_contacts']: # PDBQT, no-HD # XXX TODO BUG Line B1
for aset in self.pose[
'water_bridge_contacts']: # PDBQT, no-HD # XXX TODO BUG Line B1
for a in aset:
if not hf.getAtype(a) in self.ignore_types:
atom_pool.append(a)
if self.debug:
fp = open('DEBUG_rec_overlap.pdb', 'w')
fp.write("REMARK WATERS REMOVED BY RECEPTOR CLASHES")
for w in self.pose['water_bridge']:
try: # try/except is faster
for a in atom_pool:
#if hf.dist(w, a, sq=False) < self.clash:
if hf.dist(w, a, sq=False) < cutoff:
overlapping.append(w)
self.pose['text'].remove(w)
#print "\n\n\n #### REMOVED", w
if self.debug:
fp.write(w + "\n")
fp.write(a + "\n")
raise self.qs
except:
pass
if self.debug:
fp.close()
print "Processed: %d | Rec-overlap %d" % (len(
self.pose['water_bridge']), len(overlapping))
for w in overlapping:
idx = self.pose['water_bridge'].index(w)
del self.pose['water_bridge_contacts'][idx]
self.pose['water_bridge'].remove(w)
self.pose['water_over_rec'] = overlapping
def recOverlapWaters(self): # pose, cutoff = CLASH,
#ignore_types = ['HD', 'H', 'W'] ):
""" INPUT : pdb lig (multi)model and pdb target model
OUTPUT: cleaned pdb lig model
"""
#print "WARNING! missing interactions! Line B1"
overlapping = []
atom_pool = []
cutoff = self.clash**2
#cutoff += 0.5
#cutoff = 0
#for a in self.pose['vdw_contacts']: # PDBQT, no-HD # XXX TODO BUG Line B1
for aset in self.pose['water_bridge_contacts']: # PDBQT, no-HD # XXX TODO BUG Line B1
for a in aset:
if not hf.getAtype(a) in self.ignore_types:
atom_pool.append(a)
if self.debug:
fp = open('DEBUG_rec_overlap.pdb','w')
fp.write("REMARK WATERS REMOVED BY RECEPTOR CLASHES")
for w in self.pose['water_bridge']:
try: # try/except is faster
for a in atom_pool:
#if hf.dist(w, a, sq=False) < self.clash:
if hf.dist(w, a, sq=False) < cutoff:
overlapping.append(w)
self.pose['text'].remove(w)
#print "\n\n\n #### REMOVED", w
if self.debug:
fp.write(w+"\n")
fp.write(a+"\n")
raise self.qs
except:
pass
if self.debug:
fp.close()
print "Processed: %d | Rec-overlap %d" % (len(self.pose['water_bridge']), len(overlapping) )
for w in overlapping:
idx = self.pose['water_bridge'].index(w)
del self.pose['water_bridge_contacts'][idx]
self.pose['water_bridge'].remove(w)
self.pose['water_over_rec'] = overlapping
