Dear modellers,
I have searched the archives of the discussion forum for this and found a few questions, but no answers. I wish to impart the template structure to a sequence exactly, and select only the loops for refinement. I am using modeller 7v7, and have not switched to 8 yet. I would be interested in ways to do this for both versions, but most interested in 7v7. Currently, I use a snippet of top script taken from one of the examples that looks like this:
SUBROUTINE ROUTINE = 'select_atoms' PICK_ATOMS SELECTION_SEGMENT='100:' '105:', SELECTION_SEARCH='segment', ; PICK_ATOMS_SET=1, RES_TYPES='all', ATOM_TYPES='all', ; SELECTION_FROM='all', SELECTION_STATUS='initialize' RETURN END_SUBROUTINE
but I have no idea how to extend this to do multiple loop modelling all at the same time... i.e. segment 100-105, 158-164, 200-206, etc... all at once. I've tried lots of things, but I have not been able to figure out the syntax of top too well. I saw an email in the archive by Obdulia Rabal, who used something like
SUBROUTINE ROUTINE = 'select_loop_atoms' READ_ALIGNMENT FILE = ALNFILE, ALIGN_CODES = KNOWNS SEQUENCE, ADD_SEQUENCE = off SET SELECTION_STATUS = 'INITIALIZE' SET SELECTION_MODE = 'ATOM' SET SELECTION_SEARCH = 'SEGMENT' SET SELECTION_FROM = 'ALL' SET GAP_EXTENSION = 2 2 # gaps are extended for two residues in both # directions to get loops (insertions & deletions) PICK_ATOMS SELECTION_SEGMENT = 'LOOPS' RETURN END_SUBROUTINE
but he complained that the program still selected all the atoms for refinement, instead of only the loop portions. I've finally resorted to modelling every single loop separately, but this is no fun at all, and is time consuming. Can someone help please?
Thanks, David
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= David Bostick Office: 262 Venable Hall Dept. of Physics and Astronomy Phone: (919)962-0165 Program in Molecular and Cellular Biophysics UNC-Chapel Hill CB #3255 Phillips Hall dbostick@physics.unc.edu Chapel Hill, NC 27599 http://www.unc.edu/~dbostick =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
David L. Bostick wrote: > I have searched the archives of the discussion forum for this and found a > few questions, but no answers. I wish to impart the template structure to > a sequence exactly, and select only the loops for refinement. I am using > modeller 7v7, and have not switched to 8 yet. I would be interested in > ways to do this for both versions, but most interested in 7v7.
You can use TOP scripts with both Modeller 7v7 and Modeller 8.
> SUBROUTINE ROUTINE = 'select_atoms' > PICK_ATOMS SELECTION_SEGMENT='100:' '105:', SELECTION_SEARCH='segment', ; > PICK_ATOMS_SET=1, RES_TYPES='all', ATOM_TYPES='all', ; > SELECTION_FROM='all', SELECTION_STATUS='initialize' > RETURN > END_SUBROUTINE > > but I have no idea how to extend this to do multiple loop modelling all at > the same time... i.e. segment 100-105, 158-164, 200-206, etc... all at > once.
First, you use select_loop_atoms, not select_atoms, for loop refinement.
You can simply call PICK_ATOMS multiple times; every time after the first, set SELECTION_STATUS to 'add':
SUBROUTINE ROUTINE = 'select_loop_atoms' PICK_ATOMS SELECTION_SEGMENT='100:' '105:', ; SELECTION_SEARCH='segment', PICK_ATOMS_SET=1, ; RES_TYPES='all', ATOM_TYPES='all', ; SELECTION_FROM='all', SELECTION_STATUS='initialize' PICK_ATOMS SELECTION_SEGMENT='158:' '164:', ; SELECTION_SEARCH='segment', PICK_ATOMS_SET=1, ; RES_TYPES='all', ATOM_TYPES='all', ; SELECTION_FROM='all', SELECTION_STATUS='add' PICK_ATOMS SELECTION_SEGMENT='200:' '206:', ; SELECTION_SEARCH='segment', PICK_ATOMS_SET=1, ; RES_TYPES='all', ATOM_TYPES='all', ; SELECTION_FROM='all', SELECTION_STATUS='add' RETURN END_SUBROUTINE
...and so on. For Python with Modeller 8v1, use the loop.py example at http://salilab.org/modeller/manual/node33.html as your base and use a select_loop_atoms routine similar to the following:
def select_loop_atoms(self): stat = 'INITIALIZE' for segs in (('100:', '105:'), ('158:', '164:'), ('200:', '206:')): self.pick_atoms(selection_segment=segs, selection_search='segment', pick_atoms_set=1, res_types='all', atom_types='all', selection_from='all', selection_status=stat) stat = 'ADD'
Ben Webb, Modeller Caretaker
Thank you Ben,
I have a few follow-up questions. I will enumerate them for clarity:
1) does select_loop_atoms use template-derived restraints if they are available?
2) does it also use secondary structure restraints if they are defined.. i.e.:
SUBROUTINE ROUTINE = 'special_restraints' SET ADD_RESTRAINTS = on MAKE_RESTRAINTS RESTRAINT_TYPE = 'alpha', RESIDUE_IDS = '153' '162' RETURN END_SUBROUTINE
3) when I have used select_loopo_atoms in the past, often the loops look "over-refined" as if they don't fit into the context of the protein. Is it possible that I'm using it wrong? ... or is it that the answer to (1) and (2) above is, "no."
4) is it possible to pick multiple segments to refine with select_atoms as well as select_loop_atoms? I have been able to do this with select_loop_atoms, but haven't figured it out with select_atoms.
5) This is probably trivial and I'm a bit embarrassed to ask, but in the python scripts, is is possible to have multiple def statements? For example if I wanted to select atoms for loop modelling, AND impose secondary structure constraints at the same time? For example, I tried something to the effect below, but I don't believe it's correct, because I get complaints about "indentation" or something::
from modeller.automodel import * log.verbose() # Override the 'select_atoms' routine in the 'automodel' class class mymodel(automodel): # select only certain segments for refinement def select_atoms(self): self.pick_atoms(selection_segment=('100:', '103:'), selection_search='segment', pick_atoms_set=1, res_types='all', atom_types='all', selection_from='all', selection_status='initialize') self.pick_atoms(selection_segment=('154:', '177:'), selection_search='segment', pick_atoms_set=1, res_types='all', atom_types='all', selection_from='all', selection_status='add') # special restraints def special_restraints(self, aln): rsr = self.restraints # An alpha-helix: rsr.make(aln, restraint_type='ALPHA', residue_ids=('99', '102'), spline_on_site=False) rsr.make(aln, restraint_type='ALPHA', residue_ids=('153', '179'), spline_on_site=False)
env = environ() env.io.atom_files_directory = './'
# Choose 1 for env.edat.nonbonded_sel_atoms if you want the selected atoms to "feel" # the unselected atoms. If you want them not to "feel" them, choose 2. env.edat.nonbonded_sel_atoms = 1
m = mymodel(env, alnfile='alignment.ali', knowns=('temp','kvap'), sequence='1out') m.starting_model = 1 m.ending_model = 2 m.make()#
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= David Bostick Office: 262 Venable Hall Dept. of Physics and Astronomy Phone: (919)962-0165 Program in Molecular and Cellular Biophysics UNC-Chapel Hill CB #3255 Phillips Hall dbostick@physics.unc.edu Chapel Hill, NC 27599 http://www.unc.edu/~dbostick =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
On Tue, 26 Jul 2005, Modeller Caretaker wrote:
> David L. Bostick wrote: > > I have searched the archives of the discussion forum for this and found a > > few questions, but no answers. I wish to impart the template structure to > > a sequence exactly, and select only the loops for refinement. I am using > > modeller 7v7, and have not switched to 8 yet. I would be interested in > > ways to do this for both versions, but most interested in 7v7. > > You can use TOP scripts with both Modeller 7v7 and Modeller 8. > > > SUBROUTINE ROUTINE = 'select_atoms' > > PICK_ATOMS SELECTION_SEGMENT='100:' '105:', SELECTION_SEARCH='segment', ; > > PICK_ATOMS_SET=1, RES_TYPES='all', ATOM_TYPES='all', ; > > SELECTION_FROM='all', SELECTION_STATUS='initialize' > > RETURN > > END_SUBROUTINE > > > > but I have no idea how to extend this to do multiple loop modelling all at > > the same time... i.e. segment 100-105, 158-164, 200-206, etc... all at > > once. > > First, you use select_loop_atoms, not select_atoms, for loop refinement. > > You can simply call PICK_ATOMS multiple times; every time after the > first, set SELECTION_STATUS to 'add': > > SUBROUTINE ROUTINE = 'select_loop_atoms' > PICK_ATOMS SELECTION_SEGMENT='100:' '105:', ; > SELECTION_SEARCH='segment', PICK_ATOMS_SET=1, ; > RES_TYPES='all', ATOM_TYPES='all', ; > SELECTION_FROM='all', SELECTION_STATUS='initialize' > PICK_ATOMS SELECTION_SEGMENT='158:' '164:', ; > SELECTION_SEARCH='segment', PICK_ATOMS_SET=1, ; > RES_TYPES='all', ATOM_TYPES='all', ; > SELECTION_FROM='all', SELECTION_STATUS='add' > PICK_ATOMS SELECTION_SEGMENT='200:' '206:', ; > SELECTION_SEARCH='segment', PICK_ATOMS_SET=1, ; > RES_TYPES='all', ATOM_TYPES='all', ; > SELECTION_FROM='all', SELECTION_STATUS='add' > RETURN > END_SUBROUTINE > > ...and so on. For Python with Modeller 8v1, use the loop.py example at > http://salilab.org/modeller/manual/node33.html as your base and use a > select_loop_atoms routine similar to the following: > > def select_loop_atoms(self): > stat = 'INITIALIZE' > for segs in (('100:', '105:'), ('158:', '164:'), > ('200:', '206:')): > self.pick_atoms(selection_segment=segs, > selection_search='segment', > pick_atoms_set=1, > res_types='all', > atom_types='all', > selection_from='all', > selection_status=stat) > stat = 'ADD' > > Ben Webb, Modeller Caretaker > -- > modeller-care@salilab.org http://www.salilab.org/modeller/ > Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage >
David L. Bostick wrote: > I have a few follow-up questions. I will enumerate them for clarity: > > 1) does select_loop_atoms use template-derived restraints if they are > available?
No - only the statistical potential is used. Loops generally have no usable template information anyway.
> 2) does it also use secondary structure restraints if they are defined.. > i.e.: > > SUBROUTINE ROUTINE = 'special_restraints' > SET ADD_RESTRAINTS = on > MAKE_RESTRAINTS RESTRAINT_TYPE = 'alpha', RESIDUE_IDS = '153' '162' > RETURN > END_SUBROUTINE
Yes. special_patches is also called, if defined.
> 3) when I have used select_loopo_atoms in the past, often the loops look > "over-refined" as if they don't fit into the context of the protein. Is it > possible that I'm using it wrong? ... or is it that the answer to (1) and > (2) above is, "no."
You should build multiple loop models, and pick the 'best' one, as the procedure will often give you loops that don't look that great.
> 4) is it possible to pick multiple segments to refine with select_atoms as > well as select_loop_atoms? I have been able to do this with > select_loop_atoms, but haven't figured it out with select_atoms.
Sure; the procedure is exactly the same as shown in my previous email.
> 5) This is probably trivial and I'm a bit embarrassed to ask, but in the > python scripts, is is possible to have multiple def statements? For example > if I wanted to select atoms for loop modelling, AND impose secondary > structure constraints at the same time? For example, I tried something to > the effect below, but I don't believe it's correct, because I get > complaints about "indentation" or something:: > > from modeller.automodel import * > log.verbose() > # Override the 'select_atoms' routine in the 'automodel' class > class mymodel(automodel): > # select only certain segments for refinement > def select_atoms(self): > self.pick_atoms(selection_segment=('100:', '103:'), > selection_search='segment', pick_atoms_set=1, > res_types='all', atom_types='all', > selection_from='all', selection_status='initialize') > self.pick_atoms(selection_segment=('154:', '177:'), > selection_search='segment', pick_atoms_set=1, > res_types='all', atom_types='all', > selection_from='all', selection_status='add') > # special restraints > def special_restraints(self, aln): > rsr = self.restraints > # An alpha-helix: > rsr.make(aln, restraint_type='ALPHA', residue_ids=('99', '102'), > spline_on_site=False) > rsr.make(aln, restraint_type='ALPHA', residue_ids=('153', '179'), > spline_on_site=False)
Of course you can - but Python is sensitive to indentation, so you have to make sure that the 'def select_atoms' and 'def special_restraints' lines are both at the same indentation level (i.e. they have the same number of spaces before the 'def').
Ben Webb, Modeller Caretaker
-
David L. Bostick -
Modeller Caretaker