- modeller_usage - salilab.org

is LOOP threading a common result?
by Starr Hazard 14 Dec '05

14 Dec '05

Folks, I am using Mod8v1 on a MAC G5 running OSX 10.3. In a protein with about 500 aa there was about a 100 amino acid region which did not match the single available template. I asked Modeller to make 1000 loops in the gap region. The happened and after a few fits I got the DOPE analysis of all 1000 loops complete. I picked the best 20 loops (lowest overall DOPE scores;minimal plots in the loop region) and examined these structures. In every one of the best twenty AND in the worst loop model, I see regions where the routines have threaded the loop though the nonloop regions of the original model. In words, suppose I have two antiparallel beta sheet strands. The backbone of the loop sometimes passes between these sheet stands or through the turn loop at the ends of the sheets before coursing away from the existing protein. Do the loop generator routines pay any attention at all to the nonloop regions? In older versions of Modeller, the routines would create knots where the insertions in the alignment were present. The alignment could be adjusted to prevent this. How can I persuade modeller not to thread regions of loop through "existing" non loop regions of the protein? Starr

2 1

beta-sheets and alpha-helixes
by Matthias Negri 14 Dec '05

14 Dec '05

Hello, my name is Matthias and I'm a PhD Student at University of Saarland. I'm trying to model a cypP450 with modeller7v7 by defining the secondary structure (alpha, sheet..) in the special_restraints ROUTINE, but I still get error messages and the job is broken up! That's the error message I got: "Dynamically allocated memory at amaxseq [B,kB,MB]: 39334681 38412.773 37.512 TOP_________> 169 174 CALL ROUTINE = 'special_restraints' TOP_________> 170 479 SET ADD_RESTRAINTS = ON TOP_________> 171 480 MAKE_RESTRAINTS RESTRAINT_TYPE = 'ALPHA', RESIDUE_IDS = '; 228:A' '253:A' assgn___215E> Dynamic memory allocation failed. Routine, variable, status: vars%val 1 recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1" While my top file (special_restraints) is: "SUBROUTINE ROUTINE = 'special_restraints' SET ADD_RESTRAINTS = on MAKE_RESTRAINTS RESTRAINT_TYPE = 'ALPHA', RESIDUE_IDS = '228:A' '253:A' MAKE_RESTRAINTS RESTRAINT_TYPE = 'ALPHA', RESIDUE_IDS = '140:A' '159:A' MAKE_RESTRAINTS RESTRAINT_TYPE= 'SHEET', RESIDUE_IDS = '464:A' '467:A' MAKE_RESTRAINTS RESTRAINT_TYPE= 'SHEET', RESIDUE_IDS = '480:A' '484:A' MAKE_RESTRAINTS RESTRAINT_TYPE= 'SHEET', RESIDUE_IDS = '490:A' '495:A' RETURN END_SUBROUTINE" 1.Can you tell me what the reason for failing is? 2.Further, does modeller8.1 resolve this problem (if there is a bug in mod7v7?)! 3.I even get the error disclaimer: "openf5__224_> Open 11 OLD SEQUENTIAL ${MODINSTALL7v7}/modlib/sched.lib TOP_________> 196 233 WRITE_SCHEDULE FILE = SCHFILE openf5__224_> Open 14 UNKNOWN SEQUENTIAL CYP17.sch TOP_________> 197 234 DO IREPEAT = 1 REPEAT_OPTIMIZATION 1 TOP_________> 198 235 CALL ROUTINE = 'single_model_pass' TOP_________> 199 253 READ_MODEL FILE = MODEL,MODEL_SEGMENT = '31:A' '496:A' openf5__224_> Open 11 OLD SEQUENTIAL Mixed1.ini openf5__224_> Open 11 OLD SEQUENTIAL Mixed1.ini rdatm___297_> Segments, residues, atoms: 1 466 7457 rdatm___298_> Segment: 1 31 A 496 A 7457 TOP_________> 200 254 DO SCHEDULE_STEP = 1 N_SCHEDULE 1 TOP_________> 201 255 OPERATE ARGUMENTS = ID1 1, OPERATION = 'SUM', RESULT = ID; 1 TOP_________> 202 256 PICK_RESTRAINTS ADD_RESTRAINTS = OFF nselat__596E> Atom index is out of range (iatm,natm): 7458 7457 recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1 Dynamically allocated memory at finish [B,kB,MB]: 39111561 " What's the reason for, where I do mistakes? Thanks in advance, Matthias Negri 8.2 Pharmaceutical and Medicinal Chemistry Saarland University P.O.Box 151150 66041 Saarbrücken, Germany phone: +49-681-302-2980 fax: +49-681-302-4386 Email: m.negri(a)mx.uni-saarland.de http://www.uni-saarland.de/fak8/hartmann Computational Medicinal Chemistry Group Dept. of Pharmaceutical Sciences, University of Bologna Via Belmeloro, 6; 40100 Bologna, Italy Email: thias(a)scfarm.unibo.it http://www.scfarm.unibo.it

2 1

(no subject)
by mahboobeh zarrabi 14 Dec '05

14 Dec '05

hi how can i model protein consist of heteroatoms(K ion and water)? thanks

1 0

Re: modeller_usage Digest, Vol 4, Issue 113
by rathankar rao 09 Dec '05

09 Dec '05

modeller_usage-request(a)salilab.org wrote: Send modeller_usage mailing list submissions to modeller_usage(a)salilab.org To subscribe or unsubscribe via the World Wide Web, visit http://salilab.org/mailman/listinfo/modeller_usage or, via email, send a message with subject or body 'help' to modeller_usage-request(a)salilab.org You can reach the person managing the list at modeller_usage-owner(a)salilab.org When replying, please edit your Subject line so it is more specific than "Re: Contents of modeller_usage digest..." Today's Topics: 1. file name problem (Starr Hazard) 2. Re: file name problem (Modeller Caretaker) 3. Re: To Model missing domains in the content of the rest (Shirley Li) 4. To use separate templates for different region of a protein (Shirley Li) ---------------------------------------------------------------------- Message: 1 Date: Thu, 08 Dec 2005 13:53:38 -0500 From: Starr Hazard Subject: [modeller_usage] file name problem To: modeller_usage(a)salilab.org Message-ID: Content-Type: text/plain; charset=us-ascii; format=flowed Folks, The following lines: mdl.assess_dope(output='ENERGY_PROFILE NO_REPORT', file=code+'.profile', normalize_profile=True, smoothing_window=15) are producing output files with a question mark in the name string. ABC31_HUMAN.BL09950001.pdb?.profile The question mark renders cp, mv and rename incapable of doing anything with the file. Is there a way to prevent modeller from appending the question mark to the created files? Thanks, Starr ------------------------------ Message: 2 Date: Thu, 08 Dec 2005 11:17:16 -0800 From: Modeller Caretaker Subject: Re: [modeller_usage] file name problem To: Starr Hazard Cc: modeller_usage(a)salilab.org Message-ID: <439886BC.6050700(a)salilab.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Starr Hazard wrote: > mdl.assess_dope(output='ENERGY_PROFILE NO_REPORT', > file=code+'.profile', > normalize_profile=True, smoothing_window=15) > > are producing output files with a question mark in the name string. > > ABC31_HUMAN.BL09950001.pdb?.profile If this is from your earlier script, you're getting 'code' from the following bit of Python: code = in_file.readline() Note that readline() does not remove the line-end character(s), and that's probably confusing Modeller. Try putting in code = code.rstrip('\r\n') after you read the line from your file. > The question mark renders cp, mv and rename incapable of doing anything > with the file. You can just quote the filename, or escape the ? with a \ character, to make cp, mv and friends work with it. Ben Webb, Modeller Caretaker -- modeller-care(a)salilab.org http://www.salilab.org/modeller/ Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage ------------------------------ Message: 3 Date: Thu, 8 Dec 2005 13:15:23 -0800 (PST) From: Shirley Li Subject: [modeller_usage] Re: To Model missing domains in the content of the rest To: modeller_usage(a)salilab.org Message-ID: <20051208211523.94416.qmail(a)web53801.mail.yahoo.com> Content-Type: text/plain; charset="iso-8859-1" Dear MODELLER experts, I have a target protein, whose X-ray structure is PARTIALLY available (the middle part, named domain B). There are two big regions missing (domain A & C) in the crystal structure, with more than 50 residues in each I would like to use MODELLER to model domain A & C, and there are templates available. I wonder if it is possible to include THE CONTENT OF domain B. in another word, to create a complete model of A-B-C while keep the coordinates of B domain fixed. Any help/information will be greatly appreciated. hi its possible, just include the sequence of domain B in the ali file along with the seqence of domains A and B . the program will definitely orient the known structure with the calculated structure, but still u have to minimize it later on to infer the proper alignment with regards rathankar Shirley --------------------------------- __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0

loops iterations and various DOPE scores
by Starr Hazard 08 Dec '05

08 Dec '05

Folks, I am having several difficulties with analysis of my loops. I built a model and selected a region to refine as a loop. I have successfully created 1000 loop files. Based on the model_energies.py file from the examples, I am attempting to scan the 1000 loop model files. So far the script manages to read only about 250 of the set. Here's the script. The file "loop.list" has the names of the 1000 files, one per line: ___________________________ from modeller.automodel import * # Load the automodel class log.verbose() # request verbose output env = environ() # create a new MODELLER environment to build this model in env.libs.topology.read(file='$(LIB)/top_heav.lib') # read topology env.libs.parameters.read(file='$(LIB)/par.lib') # read parameters # directories for input atom files env.io.atom_files_directory = './:../atom_files' def dope_profile(mdl, file): # DOPE energy parameters edat = energy_data(contact_shell=15.0, dynamic_modeller=True, dynamic_lennard=False, dynamic_sphere=False, excl_local=(False, False, False, False)) # DOPE group restraints oldgprsr = mdl.group_restraints if not model.dope_restraints: model.dope_restraints = \ group_restraints(classes='${LIB}/atmcls-mf.lib', parameters='${LIB}/dist-mf.lib') mdl.group_restraints = model.dope_restraints molpdf = mdl.energy(output='SHORT', file=file, edat=edat, residue_span_range=(1, 9999), normalize_profile=True, smoothing_window=10) mdl.group_restraints = oldgprsr return molpdf filename = 'loop.list' in_file = open(filename,"r") while 1: # read model file code = in_file.readline() mdl = model(env) mdl.read(file=code) aln = alignment(env) # generate topology aln.append_model(mdl, atom_files=code, align_codes=code) aln.append_model(mdl, atom_files=code, align_codes=code+'-ini') mdl.generate_topology(aln, sequence=code+'-ini') mdl.transfer_xyz(aln) if info.version_info == (8,0): dope_profile(mdl, code+'.profile') else: mdl.assess_dope(output='ENERGY_PROFILE NO_REPORT', file=code+'.profile', normalize_profile=True, smoothing_window=15) _____________________________________ Here are the first five lines of an output profile: __________________________________________________________ # Energy of each residue is written to: ABC31_HUMAN.BL00010001.pdb .profile # The profile IS normalized by the number of restraints. # The profiles are smoothed over a window of residues: 13 # The sum of all numbers in the file: -15.0456 ___________________________________________________ The actual profile file name looks like this "ABC31_HUMAN.BL00010001.pdb?.profile" 1a) So, Is there a limit to ability of Modeller or python to handle long file name strings? 1b) Why am I not able to digest all 1000 loop models? 2) In the log file that's produced I see a line like this for the first loop model: DOPE score :-51270.746094 What is the relationship between the summed per residue values from the profile output (ie -15.0456) and this DOPE score from the log file? Thanks, Starr

2 3

Re: To Model missing domains in the content of the rest
by Shirley Li 08 Dec '05

08 Dec '05

Dear MODELLER experts, I have a target protein, whose X-ray structure is PARTIALLY available (the middle part, named domain B). There are two big regions missing (domain A & C) in the crystal structure, with more than 50 residues in each I would like to use MODELLER to model domain A & C, and there are templates available. I wonder if it is possible to include THE CONTENT OF domain B. in another word, to create a complete model of A-B-C while keep the coordinates of B domain fixed. Any help/information will be greatly appreciated. Shirley --------------------------------- __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

To use separate templates for different region of a protein
by Shirley Li 08 Dec '05

08 Dec '05

Dear All, With MODELLER, I wonder if it is possible to use different template(s) for different region of one target protein. For example, One protein has domain 1, 2 and 3. I would like to construct a complete model of this protein, with templates A and B for domain 1, templates C,D,and E for domain 3, and domain 2 already has coordinates/structure, which need to be preserved in the complete model. Is there any way to do the above? If so, how? I tried to use the "inifile" feature and use the existing coordinate file of domain 2 as "inifile". while I got error message of " Atom index is out of range (iatm,natm)". Any help/information will be very much appreciated. Shirley --------------------------------- Yahoo! Shopping Find Great Deals on Holiday Gifts at Yahoo! Shopping

1 0

file name problem
by Starr Hazard 08 Dec '05

08 Dec '05

Folks, The following lines: mdl.assess_dope(output='ENERGY_PROFILE NO_REPORT', file=code+'.profile', normalize_profile=True, smoothing_window=15) are producing output files with a question mark in the name string. ABC31_HUMAN.BL09950001.pdb?.profile The question mark renders cp, mv and rename incapable of doing anything with the file. Is there a way to prevent modeller from appending the question mark to the created files? Thanks, Starr

2 1

Re: [modeller_usage] How to choose the best model?
by ��¨Ե��ط�� Ե��͹�� 07 Dec '05

07 Dec '05

Dear Modeller users For my research the parameter which i should consider for the best model is from the evaluated program such as PROSA PROCHECK or Verify3D. For Procheck , i use the model which give a quaility of model within >80% core regions. That i used in my project so if any recomments for my research please contact me. Thank Mr. JItrayut Jitonnom CSML Chiang Mai University Thailand -----Original Message----- From: carromeu(a)usp.br Sent: Fri, 02 Dec 2005 21:41:42 -0200 To: modeller_usage(a)salilab.org Subject: [modeller_usage] How to choose the best model? > Dear Modeller users, > > I'm generating diverses models to my protein (400 - is this to great?). >However I don't know how is the best parameter to look when choosing the best >model. Objective function from Modeller, ProCheck parameters, Energy >parameters... What parameters you use and why? If using Ramachandran plot as a >discriminative parameter, how great must be the parameters calculated (for >example: >90% in favored regions, ---------------------------------------------------------------------------------- http://www.chaiyo.com ไชโยเมล์ - ฟรีอีเมล์รูปแบบใหม่ สดใสกว่าเดิม...

3 2

How to include alignment to model_symmetry_define()
by bmbbl＠leeds.ac.uk 07 Dec '05

07 Dec '05

Hi Modellers, I need to obtain loop conformation distributions in a tetramer structure. So I used loopmodel() to redefine the 4 loops in each subunit. I found that the 4 loops in each tetramer is not symmetrical. I therefore applied model_symmetry_define() in order to fix them the same during the optimisation. Here I have a question. The code I have used is at http://salilab.org/modeller/manual/node160.html However, in define_symmetry-3.atm file, the structure looks very bizarre. It has lost its secondary structure elements and 4 loops are not symmetrical at all. So I tried to add the sequence alignment between the loop model and the template. I cannot work out the python code. Could some one kindly have a look what I have done wrongly please. Here is the code: # Example for: model.symmetry.define() # This will force two copies of RCK4 to have similar mainchain # conformation. log.level(1, 1, 1, 1, 0) env = environ() env.libs.topology.read(file='$(LIB)/top_heav.lib') env.libs.parameters.read(file='$(LIB)/par.lib') #env.io.atom_files_directory = 'xxx' env.io.atom_files_directory = '.' def defsym(lop, aln, seg1, seg2): for (set, seg) in [(2, seg1), (3, seg2)]: lop.pick_atoms(aln, pick_atoms_set=set, selection_segment=seg, atom_types='All', selection_status='INITIALIZE', selection_search='SEGMENT') lop.symmetry.define(symmetry_weight=1.0, add_symmetry=(True, False)) # Generate two copies of a segment: lop = model(env, file='model-nam', model_segment=('1:A', '110:D')) aln = alignment(env) aln.append_model(lop, align_codes='model-nam', atom_files='tmp-nam') aln.append_model(lop, align_codes='model-nam'+'_ini', atom_files='tmp-nam'+'_ini') lop.generate_topology(aln, sequence='model-nam'+'_ini') lop.transfer_xyz(aln) lop.build(initialize_xyz=False, build_method='INTERNAL_COORDINATES') #lop.rename_segments(segment_ids=('A', 'B')) #, renumber_residues=(1, 1)) myedat = energy_data(dynamic_sphere = False) lop.energy(edat=myedat) lop.randomize_xyz(deviation=6.0) # Define the two segments (chains in this case) to be identical: defsym(lop, aln, seg1=('39:A', '49:A'), seg2=('39:B', '49:B')) # Make them identical by optimizing the initial randomized structure # without any other restraints: lop.energy(edat=myedat) lop.write(file='define_symmetry-1.atm') lop.optimize(max_iterations=300, edat=myedat) lop.write(file='define_symmetry-2.atm') lop.energy(edat=myedat) # Now optimize with stereochemical restraints so that the # result is not so distorted a structure (still distorted # because optimization is not thorough): myedat.dynamic_sphere = True lop.restraints.make(aln, restraint_type='stereo', spline_on_site=False, edat=myedat) lop.randomize_xyz(deviation=3.0) for method in (1, 3, 1): # 1 = conjugate gradients, 3 = molecular dynamics lop.optimize(max_iterations=300, md_return='FINAL', edat=myedat, optimization_method=method, output='REPORT') lop.write(file='define_symmetry-3.atm') lop.energy(edat=myedat) # Create a blank alignment so that superpose uses its 1:1 default aln = alignment(env) lop = model(env, file='define_symmetry-3.atm', model_segment=('39:A', '49:A')) lop2 = model(env, file='define_symmetry-3.atm', model_segment=('39:B', '49:B')) lop.pick_atoms(aln, atom_types='All') lop.superpose(lop2, aln) Also, what does aln.append_model(lop, align_codes='model-nam'+'_ini', atom_files='tmp-nam'+'_ini') mean? The erro message by executing the above .py file is: "_modeller.error: transfe_682E> Structural information is not available for all sequences in the alignment. Aborting." Thank you very much. Binbin

2 1