- modeller_usage - salilab.org

transfer of side chain conformations
by Daniel John Rigden 02 Jun '00

02 Jun '00

Hi everyone Is it possible to get MODELLER to transfer side chain conformations to a new backbone? I have some backbone traces derived from dynamic studies. I'd like to transfer the side chains from the original structure prior to refinement and further studies. Any help much appreciated. Thanks Daniel Rigden +-------------------------------------------------------------------------+ | Dr Daniel John Rigden | | CENARGEN/EMBRAPA | e-mail: daniel(a)cenargen.embrapa.br | | Parque Estacao Biologica | http://www.cenargen.embrapa.br | | PqEB - Final - Av. W3 Norte | Phone: +55 (61)448-4741 | | 70770-900, Brasilia-D.F.-BRAZIL | Fax: +55 (61)340-3624 | +-------------------------------------------------------------------------+

2 1

"STARTING_MODEL" and "ENDING_MODEL"
by Xiao-Ping Zhang 23 May '00

23 May '00

Hi, Sorry, I have noticed that someone already asked the following question two years ago and Azat Badretdinov had answered the it. But I still don't understand the meaning behind the STARTING_MODEL" and "ENDING_MODEL. > > > >What exactly is the meaning behind setting "STARTING_MODEL" and > >"ENDING_MODEL"? Does setting the former to 1 and the latter to, say, > >7 result in seven models being built with the 7th model being the most > >refined? >No, it does not. STARTING and ENDING will determine the range of indeces >in .B???? file names of the models I am modeling a very short peptide with modeller4. If I set STARTING_MODEL=1, ENDING_MODEL=10, I get 10 models which have different orientations and spatial location. If I would like to get a model which can overlap with the original model (maximum uses the template coordinates), how to do it? And is it possible to keep the backbone in the same conformation of original peptide ? All suggestions are highly appreciated. Thank you for your help. ****************************** Xiao-Ping Zhang Department of Biochemistry Arrhenius Laboratories of Natural Sciences Stockholm Universities 106 91 Stockholm Sweden Phone: 046-08-162472 /162582 Fax: 046-08-153679 e-mail: zhang(a)biokemi.su.se

2 3

Segmantation fault
by Humberto Muniz 23 May '00

23 May '00

I was running MODELLER on Atlhon with Mandrake 7.02. The benchmarks runs well, but when i try modelling my protein the program returns the following message: /usr/local/bin/modeller4/bin/mod4: line 39: 2793 Segmentation fault /usr/local/bin/modeller4/bin/${EXECUTABLE} $STEERF >$LOGF Whats wrong??? Thanks in advance Hmuniz.

1 0

Re: multiple and not overlapping templates
by Xiao-Ping Zhang 10 Apr '00

10 Apr '00

Dear Dr. Andras Fiser, Sorry for replying you late. Thank you for your top and alignment files. They are very helpful. To compare your top file with mine, I found that I used a comma between two knows files: SET KNOWNS='template1, template2' although I don't know if it really matter. Another problem was found by using the debug option: SET OUTPUT_CONTROL= 1 1 1 1. Modeller complained that one of my template was unacceptable, probably that was because its low resolution (2.6 or something) and too many missing atoms. After changing to another template, I got model. Thank you for your help. Sincerely, Xiao-Ping At 07:05 PM 3/28/00 -0500, you wrote: > > Dear Xiao-Ping , > > It is not very fruitful, if you are not showing what exactly you are dealing > with. > > As a last help I send one of mine last application for the very same topic, > hope it helps. > a top file and an alignment file. Structures were generataed without any > problem. > > best wishes, > > Andras > > ########################################### > > INCLUDE > SET OUTPUT_CONTROL= 1 1 1 1 > SET ALNFILE='tetu.pir' > SET KNOWNS='2abc3' '2abc2' > SET SEQUENCE= 'tetu' > SET STARTING_MODEL= 1 > SET ENDING_MODEL = 30 > > CALL ROUTINE = 'model' > > > SUBROUTINE ROUTINE = 'transfer_xyz' > > # SET ALIGNMENT_FORMAT = 'PIR' > ## READ_ALIGNMENT FILE = ALNFILE, ALIGN_CODES = KNOWNS > ## MALIGN3D FIT = off, GAP_PENALTIES_3D = 0 4 > READ_ALIGNMENT FILE = ALNFILE, ALIGN_CODES = KNOWNS SEQUENCE > > READ_TOPOLOGY FILE = TOPLIB > READ_PARAMETERS FILE = PARLIB > > CALL ROUTINE = 'create_topology' > > TRANSFER_XYZ CLUSTER_CUT = -1.0 > BUILD_MODEL INITIALIZE_XYZ = OFF > > WRITE_MODEL FILE = MODEL > > RETURN > END_SUBROUTINE > ################################################## > > >P1;2abc3 > structureX:2ezm3:55::99::::: > ----YPSDNEEFDPHEQEDKLADPHEQCEFDCDPYPSDNEEDYPSHEQC----------- > -----------------------------------------* > >P1;2abc2 > structureX:2ezm2:5::48::::: > -------------------------------------------------------SQTCY > LSDKEYHRTCNMKHGSDFEQAPLPWDFHGKLMNYTELKS--* > >P1;tetu > sequence:tetu:::::::: > DCDPKLADNEEDEFDCDPNEEFDPHEDKLADPHEQCEFDCDPYPSDNEEDYPSDNEEDKL > NSAIQGSVLTSTCERTNGGYNTSSIDLNSVIENVDGSLKYE* > > ################################################# > > > -- > , > Andras Fiser, PhD # phone: (212) 327 7206 > The Rockefeller University # fax: (212) 327 7540 > Box 270, 1230 York Avenue # e-mail:fisera@rockefeller.edu > New York, NY 10021-6399, USA # http://salilab.org/~andras ------------------------------------------------ Xiao-Ping Zhang Department of Biochemistry Stockholm University 106 91 Stockholm Sweden Phone: 46-08-162582 / 162472 Fax: 46-08-153679 e-mail: zhang(a)biokemi.su.se

1 0

[Fwd: Core dump with Modeller on Linux]
by Andras Fiser 07 Apr '00

07 Apr '00

1 0

Core dump with Modeller on Linux
by David Chalmers 06 Apr '00

06 Apr '00

Hi all, I am trying to run Modeller on a linux box but all of the executables I have downloaded core dump. Do I need a new password? I am running plain Redhat 6.1. I have installed the f2c libraries using f2c-19990728-1.i386.rpm. On another note. How do you get to the year 2000 archives of the modeller list? There are only links to 1999 and 1998 as far as I can see. Best wishes and thanks for the use of a very useful program! David _____________________________________________________________________________ David Chalmers Lab: 9903 9110 Victorian College of Pharmacy Fax: 9903 9582 381 Royal Pde, Parkville, Vic 3053 http://synapse.vcp.monash.edu.au Australia David.Chalmers(a)vcp.monash.edu.au _____________________________________________________________________________

1 0

None
by Bogdan Morgun 05 Apr '00

05 Apr '00

1 0

Sequence file problem
by Troy Wymore 04 Apr '00

04 Apr '00

Hello MODELLERs, I am new to MODELLER and would like to know if there is an easy way to accomplish this task. I have a raw sequence that I want to align with my 3D structural alignment that contains nine sequences. First, it seems like my raw sequence should be formatted so that MODELLER can read it. Since this is just a sequence with no corresponding structure shouldn't the second line of the sequence file just be: sequence: It seems strange to me that in the manual all the examples of sequences have a corresponding PDB code when the point of homology modelling is to create the model hence no PDB code. Or for this do I just make one up? If the sequence file for the target is then correct can I do the following: READ_ALIGNMENT FILE = 'structures3d.ali', ALIGN_CODES = 'all' # my 3D # structural alignment for nine structures SET ADD_SEQUENCE = on, ALIGN_BLOCK = NUMB_OF_SEQUENCES READ_ALIGNMENT FILE = 'sequence.ali', ALIGN_CODES = 'all' # my target # sequence SET OVERHANG = 0 ALIGN2D GAP_PENALTIES_1D= -400 -150, GAP_PENALTIES_2D = 0.5 0.5 0.5 0.5 0.5 6 Any advice would be much appreciated. General strategies that accomplish the same basic task but don't address my specific problem are also welcome. Troy Wymore Biomedical Initiatives Group Pittsburgh Supercomputing Center

3 2

Multiple heteroatoms
by Thorsten Schweikardt 31 Mar '00

31 Mar '00

Hi there, I am trying to model a protein containing 2 copper atoms. I took the advice you gave Michel Giorgi (5-2-99) and pretended that copper is zinc. But the models I receive only have one zinc atom, not two. I tried several changes in the command syntax (/z*) to get the second copper/zinc modelled, e.g. /zz* or / z-z*, but any change results in the error: forrtl: severe (61): format/variable-type mismatch, unit 6, file stdout. I did not find anything about it in the manual. Can anyone tell me how to model the second zinc/copper atom into the structure ? Are there updated libraries available, where copper is parametrized ? Thanks for your help, Thorsten Schweikardt Molekulare Biophysik Universitat Mainz Germany

2 1

Linux huge version and multiple and not overlapping templates
by Xiao-Ping Zhang 28 Mar '00

28 Mar '00

Hi, I would like to ask a few questions: 1. I wonder if somebody succeeded with modelling of a protein with "multiple and not overlapping templates" (sorry for asking an old question) with Modeller. I have two situations about this issue: (1) model a two domain protein from two separate pdb files, there is a gap about 10 residues between them; and (2) model a two subunits protein from one pdb file which contains many subunits. I would appreciate very much if I can share your experience (How the top file is and how the alignment file should be?). Which program is the best for these kinds of modeling? 2. Has anybody compiled a linux huge version? I would like to try on the above examples. 3. Does anybody know a modeling program in which you can do ab initio modeling for short peptide (30-60 residues)? Thank you very much for your kind attention. Sincerely yours, Xiao-Ping Zhang ------------------------------------------------ Xiao-Ping Zhang Department of Biochemistry Stockholm University 106 91 Stockholm Sweden Phone: 46-08-162582 / 162472 Fax: 46-08-153679 e-mail: zhang(a)biokemi.su.se

1 0