July 2004 - modeller_usage

Does Modeller work on x86_64?
by benne278 02 Aug '04

02 Aug '04

Has anyone used modeller successfully on an Intel x86_64 system? I am trying to use 6v2 on a 3.4 GHz machine running RedHat Enterprise 3 and Modeller seg-faults even on the supplied example jobs. My actual jobs crash too, of course, and they run fine on a PIII Linux system I have access to. So either my system is not configured correctly or there is a problem running on x86_64. Are there any libraries I should check, or any other issues I should be aware of for running on x86_64? The examples do this: [bennette@localhost examples]$ ./Run_tests DIR: ./align-model-steps /HOME/software/modeller/bin/mod6v2: line 38: 13124 Segmentation fault (core dumped) $NICE ${EXECUTABLE} $STEERF >$LOGF DIR: ./all-steps /HOME/software/modeller/bin/mod6v2: line 38: 13132 Segmentation fault (core dumped) $NICE ${EXECUTABLE} $STEERF >$LOGF STOP ... etc. ... The log file for my job shows that the program doesn't report any errors, it just dies after saying it is calculating distance restraints: http://abacavir.pharmacy.umn.edu/~bennette/modeller/model-default.log.txt Thanks for any suggestions, Dr. Eric Bennett Center for Drug Design University of Minnesota

2 1

RE: [modeller_usage] More Doubts
by Cesar Avila 30 Jul '04

30 Jul '04

>I'm trying to model another protein (HSP) and I'm having trouble again. >Please look the second line of my alignment file >structureX:1uy7:2:A:236:A:::-1.00:-1.00 Dear Elton, I have seen your log messages and downloaded the pdb file to test it on my PC. It seems that you have the same problem I had some time ago. The problem appears when there are missing atoms in the structure. If you see the REMARKS section at the pdb file you will get a complete list of missing atoms in the model. You have to delete these from the alignment files, so modeller can match the residues in the alignment with the ones in the pdb file. Replace the missing residues with gaps in the alignment and you will be able to continue with your work. You have to note also that you will have to modify the second line of the alignment since the first residue in the structure is 16:A, not 2A. I hope you solve your problem. Best regards, Cesar Avila Universidad Nacional de Tucuman Argentina ___________________________________________________________ 100mb gratis, Antivirus y Antispam Correo Yahoo!, el mejor correo web del mundo http://correo.yahoo.com.ar

1 0

Loop refinement and symmetry
by David Piper 29 Jul '04

29 Jul '04

Modellers, When refining loops, can symmetry restraints be included also? For instance, if I have a tetrameric model protein and want to model a ~15 residue loop, constrain a portion of that loop to an alpha helix, and maintain tetrameric symmetry, can I simply add the special_restraints one would use for secondary structure and symmetry after the loop definition? I tried that approach and the resulting PDB file has absurdly bizarre dihdrals etc. Attached is an example of the .top file I tried. Any insight would be appreciated, thanks, David Piper

1 0

RE: [modeller_usage] More Doubts - missing aminoacids in template
by Schweikardt, Thorsten 29 Jul '04

29 Jul '04

Hi Elton, please take a look at your template-pdb file, it does not begin with aminoacid 2, but with aminoacid 16; same problem for the ending aminoacid. You have to make sure your alignment and pdb-file contain the same aminoacids. If the xray structure analysis does not reveal all aminoacids, you have to correct the alignment before modelling. If you want to know which aa are missing, extract the sequence of the existing aminoacids in the pdb-file with Modeller or DeepView (File-save-Sequence) and compare it to the sequence you are using. Happy modeling, Thorsten Schweikardt Molecular Biophysics University of Mainz > Dear Mr. Schweikardt and others, > > I'm trying to model another protein (HSP) and I'm having > trouble again. > Please look the second line of my alignment file > structureX:1uy7:2:A:236:A:::-1.00:-1.00 > > I have followed your previous instructions for the Prion > protein and it doesn't > seem to work out. Can you help me again? > The files (.ali .log .top and) are compressed by tar -cvzf command. > > Best regards, > > Vasconcelos, EJR > >

1 0

More Doubts.
by Elton Jose Rosas de Vasconcelos 28 Jul '04

28 Jul '04

Dear Mr. Schweikardt and others, I'm trying to model another protein (HSP) and I'm having trouble again. Please look the second line of my alignment file structureX:1uy7:2:A:236:A:::-1.00:-1.00 I have followed your previous instructions for the Prion protein and it doesn't seem to work out. Can you help me again? The files (.ali .log .top and) are compressed by tar -cvzf command. Best regards, Vasconcelos, EJR

1 0

RE: [modeller_usage] Some Doubts
by Schweikardt, Thorsten 28 Jul '04

28 Jul '04

Hello Elton, your alignment file states "structure" where it should be "structureX", change it and Modeller should read in the pdb file, as long as you have stored it in the path you gave Modeller in the "SET ATOM_FILES_DIRECTORY"-command. You should also give Modeller the starting/ending aminoacid number and chain number in the alignment file. The second line should be structureX:1uw3:125:A:230:A:::-1.00:-1.00 Another thing: as long as you do not want to model water or ions, just protein, you should delete all but the aminoacid atoms from the input pdb-file, as modeller will try to match the residues in the alignment file with the pdb-file and will stop if it encounters atoms that are not present in both. For alignment file parameters, have a look at http://salilab.org/modeller/manual/node72.html I hope that helps, best wishes, Thorsten Schweikardt Molecular Biophysics University of Mainz > -----Original Message----- > From: modeller_usage-bounces(a)salilab.org > [mailto:modeller_usage-bounces@salilab.org] On Behalf Of > Elton Jose Rosas de Vasconcelos > Sent: Tuesday, July 27, 2004 11:43 PM > To: modeller_usage(a)salilab.org > Subject: [modeller_usage] Some Doubts > > Dear Doctors and Modeller users, > > We are graduate students in Fortaleza, Brazil and we are > trying to model > an aminoacid sequence deduced from genotyping data using the > MODELLER for > Linux (RedHat 9.0)! Thus, we would like to know if you could > help us on the > script for modelling our target sequence, cause we are having > some troubles > with that. Our log output file stops after the CHECK_ALIGNMENT, in the > BEGINNING OF COMMAND, telling us the template's structure had > not been read. > We put our atomic coordenates files (1uw3.atm and pdb1uw3.ent) into > the /examples/atom_files/ subdirectory. > > Another error message appears when we put, in alignment file > (pri.ali), > "structure" and "sequence" in front of the template and > target sequence > respectively. This error message is "rdpir___648E> Alignment > sequence not found > in PDB file: 1 ../atom_files/pdb1uw3.ent" > > Please check our *.top *.ali and *.log files attached. > > We would appreciate very much if you could give us some light > on solving this > problem. Beforehand I thank you for your attention... > > Vasconcelos, EJR and Nugen (Núcleo de Genômica e > Bioinformática) Group - > Fortaleza, Brazil >

1 0

Dimer modelling with metals
by V Nadeem Ahmad 27 Jul '04

27 Jul '04

Dear Modellers, I would like to know how to do Dimer modelling in the vicinity of a metal ion present. else even modelling a protein chain i need to place a Metal ion in the protein how to go about it tell which script i can use and my alignment. with regards, Nadeem Ahmad V *************************************************************** V.Nadeem Ahmad B.Tech G-04,Techpark Mall Institute of Bioinformatics & Applied Biotechnology (IBAB) International Tech Park (ITPL) Whitefield, Banglore Phone: 080 -28410029 *************************************************************** "The harder you work, the luckier you get"

1 0

Some Doubts
by Elton Jose Rosas de Vasconcelos 27 Jul '04

27 Jul '04

Dear Doctors and Modeller users, We are graduate students in Fortaleza, Brazil and we are trying to model an aminoacid sequence deduced from genotyping data using the MODELLER for Linux (RedHat 9.0)! Thus, we would like to know if you could help us on the script for modelling our target sequence, cause we are having some troubles with that. Our log output file stops after the CHECK_ALIGNMENT, in the BEGINNING OF COMMAND, telling us the template's structure had not been read. We put our atomic coordenates files (1uw3.atm and pdb1uw3.ent) into the /examples/atom_files/ subdirectory. Another error message appears when we put, in alignment file (pri.ali), "structure" and "sequence" in front of the template and target sequence respectively. This error message is "rdpir___648E> Alignment sequence not found in PDB file: 1 ../atom_files/pdb1uw3.ent" Please check our *.top *.ali and *.log files attached. We would appreciate very much if you could give us some light on solving this problem. Beforehand I thank you for your attention... Vasconcelos, EJR and Nugen (Núcleo de Genômica e Bioinformática) Group - Fortaleza, Brazil

1 0

modeling mutations
by Miguel Relloso-Cereceda 26 Jul '04

26 Jul '04

Dear Modeler mailing list, I am a user of MODELLER 6v2 for Linux, and I am using this program to model the mutations that I am making on CD1b. I am looking for new interactions between amino acids that could be generated by these mutations and for possible changes generated in loops. This is the modeller that I am using: INCLUDE SET ATOM_FILES_DIRECTORY = '/home/moodyb/Miguel/37t/' SET OUTPUT_CONTROL = 1 1 1 1 1 #SET PDB_EXT = '.pdb' #SET MD_LEVEL = 'refine1' SET STARTING_MODEL = 1 SET ENDING_MODEL = 5 #SET DEVIATION = 4.0 SET KNOWNS = 'CD1b' #SET HETATM_IO = off #SET WATER_IO = off #SET HYDROGEN_IO = off #SET ALIGNMENT_FORMAT = 'PIR' SET SEQUENCE = 'CD1b-37t' SET ALNFILE = '37t.pir' CALL ROUTINE = 'model' Are these parameters the best way to get my objectives? I do not have much experience in computer science, and I went through your manual and I did not understand it at all. Please contact me with any advice and if it is not convenient, could you tell me who might be able to help me? Thank you very much. Miguel Relloso. ----------------------oOO-(_)-OOo---------------- Miguel Relloso Ph.D. Brigham & Women's Hospital Harvard Medical School Smith 514 1 Jimmy Fund Way Boston, MA 02115 phone: 617.525.1027 fax: 617.525.1010

2 1

Re: [modeller_usage] Restraints Problem
by akshay17 23 Jul '04

23 Jul '04

Dear Sir I tried rearranging parameters and Atom_ids which were written in wrong order in my .rsr file. This time, it says "segmentation fault (core dumped)". I recognize my problem somewhat similar to Mr. Daniele Bemporad's but as advised by you, i think in his .rsr file also, the parameters and Atom_ids are other way around. Anyway, i guess there might be some other problem apart from that. I am attaching my .rsr, .top & .log files. Is it possible that the User restraints i am giving might be unreasonable to be met during the model generation due to which it halts? I am not sure. Thanks for your help. Akshay Patny > > > On Tue, Jul 20, 2004 at 12:05:36AM -0500, akshay17 > > > wrote: Dear Sir, I tried removing Extra lines from my > > > .rsr file and still got the following error. > > > > > > > > > __________________________________________________________ > > > __ _ Executing Modeler Additional output in file > > > angim_06.log and in angim_06.log.moltabfile, > > > angim_06.log.restabfile, angim_06.log.viotabfile* > > > > > > lib-4190 : UNRECOVERABLE library error > > > A numeric input field contains an invalid character. > > > > > > Encountered during a list-directed READ from an > > > internal file (character variable) > > > IOT Trap > > > Abort (core dumped) > > > > > > > > > __________________________________________________________ > > __ _ > > > > Please advise. Shall, i write MODELLER 5 or MODELLER > > > 6v2 in my .rsr file. I am attaching my .rsr file > again. > > > The format of your restraints file is incorrect; see > > http://salilab.org/modeller/manual/node98.html. The > > first line must read either 'MODELLER5 VERSION: MODELLER > > FORMAT' or 'MODELLER5 VERSION: USER FORMAT'. There is no > > 'MODELLER6v2' format. Additionally, your user format > > lines are incorrect - the format should be: > > > > Id Form Modality Feature Group Numb_atoms > > Numb_parameters 0 Parameters Atom_ids > > > > - i.e., you have your parameters and Atom_ids the wrong > > way round. Your restraints file should look something > > like the following: > > > > MODELLER5 VERSION: USER FORMAT > > R 2 1 1 22 2 2 0 5.0 1.0 NZ:102 > > OD1:278 R 2 1 1 22 2 2 0 5.0 1.0 NZ:102 > > CZ:92 R 2 1 1 22 2 2 0 5.0 1.0 CZ:167 > > CG:263 R 2 1 1 22 2 2 0 5.0 1.0 NZ:199 > > CZ:259 R 2 1 1 22 2 2 0 5.0 1.0 OH:292 > > OD1:295 R 2 1 1 22 2 2 0 5.0 1.0 OH:292 > > OD1:111 R 2 1 1 22 2 2 0 5.0 1.0 NZ:240 > OD1:237 > > > Ben Webb, Modeller Caretaker > > -- > > modeller-care(a)salilab.org > > http://www.salilab.org/modeller/ Modeller mailing list: > > http://salilab.org/mailman/listinfo/modeller_usage > > > > Akshay Patny > (Graduate Student @ PhD) > Department of Medicinal Chemistry > School of Pharmacy, University of Mississippi > 417, Faser Hall, University, MS - 38677 > Tel : 662-915-1286(Off.), 662-236-3059(Res.) > E-mail : akshaypatny(a)indiatimes.com, akshay17(a)olemiss.edu > Web : www.olemiss.edu > > Akshay Patny (Graduate Student @ PhD) Department of Medicinal Chemistry School of Pharmacy, University of Mississippi 417, Faser Hall, University, MS - 38677 Tel : 662-915-1286(Off.), 662-236-3059(Res.) E-mail : akshaypatny(a)indiatimes.com, akshay17(a)olemiss.edu Web : www.olemiss.edu

2 2