def LoopBackrubRef(self):
"""
Backrubs the loop using the LoopMover_Refine_Backrub Mover
As far as I know, it does not work. Though it would be nice.
"""
movemap = MoveMap()
ft = self.pose.fold_tree()
ft_o = FoldTree()
ft_o.assign(ft)
ft.clear()
ft, movemap, loops_object = loop_tools.InitializeLoops(
self.pose, self.input_class.loops_as_strings, ft, movemap)
print ft
print "Fold Tree Correct? " + repr(ft.check_fold_tree())
self.pose.fold_tree(ft)
rounds = int(rounds)
if self.score_class.score == 0:
self.score_class.score = create_self.score_class.score_function_ws_patch(
'standard', 'self.score_class.score12')
self.score_class.score.set_weight(chainbreak, 100)
#Taking No Chances!
print loops_object
ref = LoopMover_Refine_Backrub(loops_object, self.score_class.score)
self.run_protocol(ref)
ft.assign(ft_o)
self.pose.fold_tree(ft)
self.score_class.score.set_weight(chainbreak, 0)
def default_KIC(self):
"""
LoopMover_Perturb_KIC for Low Resolution Loop Modeling
"""
print "Please cite: Mandell DJ, Coutsias EA, Kortemme T. (2009). Sub-angstrom accuracy in protein loop reconstruction by robotics-inspired conformational sampling. Nature Methods 6(8):551-2."
print "Additional options can be set using the options system. Symmetry is not supported at this time."
if self.score_class.ScoreType.get(
) != "cen_std" != self.score_class.ScorePatch.get() != "score4L":
result = tkMessageBox.askyesnocancel(
title="Set loop modeling scorefunction?",
message=
"Standard loop modeling scorefunction not set (cen_std/score4L patch). Temporarily set scorefunction?"
)
if result:
start_scorefxn = ScoreFunction()
start_scorefxn.assign(self.score_class.score)
self.score_class.score.assign(
create_score_function_ws_patch("cen_std", "score4L"))
elif result == None:
return
extend = tkMessageBox.askyesno(
title="Extended",
message=
"Start with idealized bond lenghths, angles, and discard original phi/psi?"
)
movemap = MoveMap()
ft = self.pose.fold_tree()
ft_o = FoldTree()
ft_o.assign(ft)
ft.clear()
#ft.simple_tree(self.pose.total_residue())
ft, movemap, loops_object = loop_tools.InitializeLoops(
self.pose, self.input_class.loops_as_strings, ft, movemap)
print "Fold Tree Correct? " + repr(ft.check_fold_tree())
self.pose.fold_tree(ft)
#Setup extended loop to throw away initial loop structure
if extend:
for loop in loops_object:
loop.set_extended(True)
kic_mover = LoopMover_Perturb_KIC(loops_object, self.score_class.score)
switch = SwitchResidueTypeSetMover("centroid")
recover = ReturnSidechainMover(self.pose)
switch.apply(self.pose)
self.run_protocol(kic_mover)
recover.apply(self.pose)
ft.assign(ft_o)
self.pose.fold_tree(ft)
if result:
self.score_class.score.assign(start_scorefxn)
def default_KIC(self):
"""
Uses LoopMover_Refine_KIC as High Res Mover
"""
print "Please cite: Mandell DJ, Coutsias EA, Kortemme T. (2009). Sub-angstrom accuracy in protein loop reconstruction by robotics-inspired conformational sampling. Nature Methods 6(8):551-2."
print "Additional options can be set using the options system. Symmetry is not supported at this time."
movemap = MoveMap()
ft = self.pose.fold_tree()
ft_o = FoldTree()
ft_o.assign(ft)
ft.clear()
#ft.simple_tree(self.posetotal_residue())
print len(self.input_class.loops_as_strings)
ft, movemap, loops_object = loop_tools.InitializeLoops(
self.pose, self.input_class.loops_as_strings, ft, movemap)
#print "Fold Tree Correct? " + repr(ft.check_fold_tree())
self.pose.fold_tree(ft)
loop_refine = LoopMover_Refine_KIC(loops_object)
loop_refine.set_scorefxn(self.score_class.score)
self.run_protocol(loop_refine)
ft.assign(ft_o)
self.pose.fold_tree(ft)
def default_CCD(self):
"""
Uses LoopMover_Refine_CCD as High Res Mover
"""
print "Please cite: Wang C, Bradley P, Baker D (2007). Protein-protein docking with backbone flexibility. J. Mol. Biol. 373, 503."
print "As well as: Canutescu A, Dunbrack R., Jr (2003) Cyclic coordinate descent: A robotics algorithm for protein loop closure. Protein Sci. 12, 963."
print "Additional options can be set using the options system. Symmetry is not supported at this time."
movemap = MoveMap()
ft = self.pose.fold_tree()
ft_o = FoldTree()
ft_o.assign(ft)
ft.clear()
#ft.simple_tree(self.posetotal_residue())
print len(self.input_class.loops_as_strings)
ft, movemap, loops_object = loop_tools.InitializeLoops(
self.pose, self.input_class.loops_as_strings, ft, movemap)
#print "Fold Tree Correct? " + repr(ft.check_fold_tree())
self.pose.fold_tree(ft)
loop_refine = LoopMover_Refine_CCD(loops_object)
loop_refine.set_scorefxn(self.score_class.score)
self.run_protocol(loop_refine)
ft.assign(ft_o)
self.pose.fold_tree(ft)
def default_CCD(self, fragset=False, fraglength=False):
"""
LoopMover_Perturb_CCD for Low Resolution Loop Modeling
"""
print "Please cite: Wang C, Bradley P, Baker D (2007). Protein-protein docking with backbone flexibility. J. Mol. Biol. 373, 503."
print "As well as: Canutescu A, Dunbrack R., Jr (2003) Cyclic coordinate descent: A robotics algorithm for protein loop closure. Protein Sci. 12, 963."
print "Additional options can be set using the options system. Symmetry is not supported at this time."
if self.score_class.ScoreType.get(
) != "cen_std" != self.score_class.ScorePatch.get() != "score4L":
result = tkMessageBox.askyesnocancel(
title="Set loop modeling scorefunction?",
message=
"Standard loop modeling scorefunction not set (cen_std/score4L patch). Temporarily set scorefunction?"
)
if result:
start_scorefxn = ScoreFunction()
start_scorefxn.assign(self.score_class.score)
self.score_class.score.assign(
create_score_function_ws_patch("cen_std", "score4L"))
elif result == None:
return
if not fragset:
fragset = tkFileDialog.askopenfilename(
initialdir=global_variables.current_directory,
title="Fragset File")
if not fragset: return
if not fraglength:
fraglength = tkSimpleDialog.askinteger(title="Fragment Length",
initialvalue=3)
extend = tkMessageBox.askyesno(
title="Extended",
message=
"Start with idealized bond lenghths, angles, and discard original phi/psi?"
)
movemap = MoveMap()
ft = self.pose.fold_tree()
ft_o = FoldTree()
ft_o.assign(ft)
ft.clear()
#ft.simple_tree(self.pose.total_residue())
print len(loops_as_string_array)
ft, movemap, loops_object = loop_tools.InitializeLoops(
p, loops_as_string_array, ft, movemap)
#print "Fold Tree Correct? " + repr(ft.check_fold_tree())
self.pose.fold_tree(ft)
Frag = ConstantLengthFragSet(fraglength, fragset)
switch = SwitchResidueTypeSetMover("centroid")
recover = ReturnSidechainMover(self.pose)
switch.apply(self.pose)
#Setup extended loop to throw away initial loop structure
if extend:
for loop in loops_object:
loop.set_extended(True)
x = LoopMover_Perturb_CCD(loops_object, self.score_class.score, Frag)
self.run_protocol(x)
recover.apply(self.pose)
ft.assign(ft_o)
if result:
self.score_class.score.assign(start_scorefxn)
self.pose.fold_tree(ft)