- modeller_usage - salilab.org

Ideal helices
by Dan Thomas Major 18 Jun '02

18 Jun '02

Hi, I'm modelling G-protein coupled receptors using rhodopsin as a template. When I compare the Ramachandran plot of the models vs that of rhodopsin, the models have much more ideal helices than the rhodopsin X-ray structure (2.8 A). Is this because: 1) The homology is relatively low (20-30%) and the distance constraints receive a low weight in the pdf function?? 2) The helices in rhodopsin are less regular than the globular proteins used in the pdf parametrization?? 3) None of the above... Sincerely, Dan -- Dan Thomas Major (at Dr. B. Fischer's lab) Bar-Ilan University Ramat-Gan, Israel Phone: 972-3-5317785 Fax: 972-3-5348730

2 1

user_distance, group restraints
by Karl Gruber 18 Jun '02

18 Jun '02

Dear Modeller team, I have got a couple of questions about the more uncommon restraint types that are available in Modeller (v62). There is a rather big loop in my target structure for which no homology derived restraints can be defined and which tends to protrude into the vacuum in the resulting models. I wanted to try 'user_distance' in order to get a more compact model but failed in producing a correct restraints file. test input file: ----- . . set atom_types = 'CA' pick_atoms pick_atoms_set=2, selection_segment='2:' '3:' pick_atoms pick_atoms_set=3, selection_segment='10:' '11:' set restraint_parameters = 3 1 1 27 2 2 0 0.0 0.010 make_restraints restraint_type='user_distance' . . ----- restraints file: ----- MODELLER5 VERSION: MODELLER FORMAT R 3 1 1 27 2 2 0 11 0 0.0000 0.0100 R 3 1 1 27 2 2 0 11 0 0.0000 0.0100 R 3 1 1 27 2 2 0 20 0 0.0000 0.0100 R 3 1 1 27 2 2 0 20 0 0.0000 0.0100 . . ----- It appears to be a problem of pick-set 3. The numbers of atoms and restraints are correct (also in more realistic examples) but the second set of atom indices gets lost somewhere along the way. My second bunch of questions concerns restraint feature 5 (distance between gravity centers). What is the exact format of this restraint? How are 'groups of atoms' defined in the rsr-file? Is it equivalent with defining the gravity centers as pseudo-atoms and restraining the distance between them? Thanks a lot in advance! Best wishes, Karl --------------------------------------------------------------------- Karl Gruber Institute of Chemistry KFU Graz Heinrichstrasse 28 e-mail: karl.gruber(a)uni-graz.at A-8010 Graz phone: (+43 316) 380-5483 AUSTRIA fax: (+43 316) 380-9850 ---------------------------------------------------------------------

1 0

array too small problem
by Tommy Wang 15 Jun '02

15 Jun '02

What do I do with the following problem? Thanks. Kind, OS, HostName, Kernel, Processor: 4, Linux orion.ima.org.sg 2.4.18-4custom i686 Date and time of compilation : 05/24/2002 09:27:23 Job starting time (YY/MM/DD HH:MM:SS): 2002/06/15 17:09:13.786 check_ali___> Checking the sequence-structure alignment. Implied target CA(i)-CA(i+1) distances longer than 8.0 angstroms: ALN_POS TMPL RID1 RID2 NAM1 NAM2 DIST ---------------------------------------------- 210 1 165 170 F K 8.143 270 1 229 245 V G 11.690 END OF TABLE delete__443_> Restraints marked for deletion were removed. Total number of restraints before, now: 22885 20618 pair2___223E> Array too small. Increase MPAIRS current maximum, current need: 7593 7593 recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1 Dynamically allocated memory at finish [B,kB,MB]: 15765545 15396.040 15.035 Starting time : 2002/06/15 17:09:13.786 Closing time : 2002/06/15 17:10:18.222 Total CPU time [seconds] : 31.71

1 0

align problem
by Gianluca Croce 14 Jun '02

14 Jun '02

Hi all, I'm tring to align an amino acid sequence with a protein (solved by X-ray) which conteins 6 chains in its PDB file. for my job I use the script: >SET ATOM_FILES_DIRECTORY = './../atom_files' >READ_MODEL FILE = '1icf.pdb' >SEQUENCE_TO_ALI ALIGN_CODES = '1icf', ATOM_FILES = '1icf' >READ_ALIGNMENT FILE = 't.ali', ALIGN_CODES = ALIGN_CODES 't', ADD_SEQUENCE = ON >ALIGN2D >WRITE_ALIGNMENT FILE='t-1fh0.ali', ALIGNMENT_FORMAT='PIR' >WRITE_ALIGNMENT FILE='t-1fh0.pap', ALIGNMENT_FORMAT='PAP but it don't work end the output is: >iatmcls_286W> MODEL atom not classified: THR:OT1 THR >iatmcls_286W> MODEL atom not classified: THR:OT2 THR >iatmcls_286W> MODEL atom not classified: VAL:OT1 VAL >iatmcls_286W> MODEL atom not classified: VAL:OT2 VAL >iatmcls_286W> MODEL atom not classified: THR:OT1 THR >iatmcls_286W> MODEL atom not classified: THR:OT2 THR >iatmcls_286W> MODEL atom not classified: VAL:OT1 VAL >iatmcls_286W> MODEL atom not classified: VAL:OT2 VAL >iatmcls_286W> MODEL atom not classified: SER:OT1 SER >iatmcls_286W> MODEL atom not classified: SER:OT2 SER >iatmcls_286W> MODEL atom not classified: SER:OT1 SER >iatmcls_286W> MODEL atom not classified: SER:OT2 SER >read_al_375E> Unknown residue type,position,sequence: 1 2 >recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1 any ideas? Thanks! G. ********************************** Dr. Gianluca Croce Ph.D. Student DISTA - University Amedeo Avogadro Corso Borsalino 54-15100 Alessandria-Italy Tel_office: +390131287414 Fax: +390131287416

1 0

Mol file...
by Bkcho 13 Jun '02

13 Jun '02

Hi all modellers. I want to draw 3D small molecules (Mol file) for docking to my enzyme active site. Is there any freeware to draw 3D molecules and minimize it, such as SYBYL? Best wishes, Byung-Kwan Cho Ph.D candidate, Laboratory of Molecular Biotechnology and Biomaterials, School of Chemical Engineering, Seoul National University.

1 0

Modelling post trans mods
by J Simms 11 Jun '02

11 Jun '02

Hi All, I have just started modelling a G Protein-Coupled Receptor, which has a number of post translational modifications. My problem is that my template, Rhodopsin, is glycosylated on the extracellular loops at a different position to my target sequence. Is there a way of adding the glycosylation to my target protein either at the alignment stage or before the model in refined. As always many thanks in advance. John.

2 1

Mod 6v2 : array too small
by Juliette Martin 10 Jun '02

10 Jun '02

I have installed mod 6v2, Execution of examples shows there is a problem : Scan_all_err results in : ******************* Test of last lines: ******************* **************** Test of E>: **************** FILE: ./commands/./debug_function.log pair1___223E> Array too small. Increase MPAIRS recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1 FILE: ./commands/./define_symmetry.log pair2___223E> Array too small. Increase MPAIRS recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1 And finally close my terminal.... Any idea about what should be done ? Thank you Juliette Martin

1 0

problem in modeller
by syed azimuddin 08 Jun '02

08 Jun '02

Dear modeller, iam facing strange problem. the modeller4 installed on my system linux 7.2 is not running correctly. the error message with mod command is /usr/local/bin/modeller4/bin/mod_i386-linux file not found, but with /usr/local/bin/modeller4/bin/mod_i386-linux this line or command it told that command not found. Infact before the crash of my system modeller4 was running smoothly on linux 7.2. After reinstallation of everything, it is not running. The surprising thing is, this modeller starts to run smoothlly on my friend's computer when I run it from my hard drive but with his linux 6.0. That means the files are complete there is some problem in my operating system. the env variabes are looking correct. To be in safe side I have copied the directory of modeller4 from his computer to my hard drive. both modeller were copied that is /usr/local/bin/modeller4 and the main modeller. Even then it is not running, what is the cause of the problem please suggest. Iqbal Azimuddin. Research Fellow, Protein Chemistry Group, H.E.J. Research Institute, University of Karachi, Karachi-75270. Pakistan. ------------------------------------------------------------ You've got daak! Get free email for life: http://daak.net a service of http://www.epakistan.com --------------------------------------------------------------------- Express yourself with a super cool email address from BigMailBox.com. Hundreds of choices. It's free! http://www.bigmailbox.com ---------------------------------------------------------------------

1 0

Modelling multi subunit proteins
by Norbert Straeter 05 Jun '02

05 Jun '02

Dear Modeller users, if I want to model a protein consisting of two subunits, how can I include the already modelled subunit into the calculation when modelling the second subunit such that new loops "see" the other subunit? I am currently using the simple scripts from the tutorial of the manual. Best regards Norbert -------------------------------------------------------------- Dr. Norbert Straeter Tel. +49 (0)30 838-53456 Free University of Berlin Fax -56702 Institute of Chemistry -Crystallography- Takustrasse 6 14195 Berlin, Germany strater(a)chemie.fu-berlin.de http://userpage.chemie.fu-berlin.de/~strater/ --------------------------------------------------------------

2 2

Which loop energy?
by J Simms 30 May '02

30 May '02

Hi all, I have recently started to model GPCRs, but as with everyone else I am having trouble with the loops. I have followed the tutorials , kindly given, and have resulted with a number of files with different loops starting and ending in different locations. I have investingated the log files for energy terms, as I understand it the loop model with the lowest energy term is a good starting point. My problem which energy term in the log file do I use, as there seems to be a number of energy terms even for the same loop model number. > Many thanks for any help > John >

2 1