- modeller_usage - salilab.org

building models strategy
by Gulshat Ibragimova 19 Nov '98

19 Nov '98

Dear Modellers, what could be the best strategy for building models (especially when one has to build lot of them -:)? The "search_sequence" procedure gives a list of proteins with similar sequences. However, their structures are generally misaligned. So this very diverse structural information might lead to bad models. Perhaps, it would make sense to find out structural relatives of the most identical protein (using the FSSP, at example) and to use them for building models despite their (relatives') low sequence similarity? Thanking for any comments, Gulshat --------------------------------------------------------------------- Gulshat Ibragimova, Ph.D. EMBL, Postfach 102209 69012 Heidelberg, FRG Tel: +49 (0)6221 387466, Fax: +49 (0)6221 387517

2 1

Executable file for Silicon Graphics O2
by chunhua yuan 16 Nov '98

16 Nov '98

Hi, Dear MODELLER users: Our Silicon Graphics is O2, which has operating system IRIX Release 6.3 and CPU MIPS R5000, Rev 2.1. Can anybody tell me which executable file should I download besides the modeller4-data.tar.Z? and what is the "HOSTTYPE" for O2 computer? Thank you very much. Best Regards, Chunhua Yuan

1 0

Loop modelling in a large system
by Graham tyrrell 05 Nov '98

05 Nov '98

Dear Modellers, I am attempting to use Modeller to model a number of loops into a very large protein system, and am having a few problems. (I am relatively new to Modeller so please excuse any dumb questions) I have 4 suitable structures which I wish to use as templates for the missing loop regions in my main structure. The problem I have is that I want to constrain the coordinates of my model to a greater extent towards that of the template which has the loops missing, I then want to use the remaining templates to build in the loops. This seems a similar problem to one described in the FAQ: ('5. I don't want to use one region of a template for construction of my model') however, due to the large number of loops missing, and the length of my sequence (~1000 residues) this would be extremely laborious, with plenty of room for error. I was wondering if anyone had any similar experience, and could advise me as to an appropriate strategy. Regards, Graham Tyrrell -- ---------------------------------------------------------------------- Graham Tyrrell Email (Work) : tyrrell(a)yorvic.york.ac.uk (Personal) : gpt103(a)york.ac.uk Telephone : (01904) 432550 -----------------------------------------------------------------------

3 2

Re: gt5templatesonPC
by sali＠rockvax 03 Nov '98

03 Nov '98

Hi, I will compile a Linux version that allows more templates. Please look for it on our ftp site a little later today. You will probably need to increase your swap space size and possibly also do 'limit stacksize 256000' as root and run it as root ... I agree with you that Linux on PCs is the way to go! Version 5, due in a couple of weeks, has dynamic memory allocation and does not have such size problems. Best, Andrej > Dear Sali, > > I m very interested to run modeler with 6 or seven templates on a PC with > LINUX. > Is there a possibility to do this. (I have seen only the number 5 for PCs > and and a higher number for SGIs > in your handbook, but I think to use a PC for the modelerjob is very > usefull, bacause I tested, that a PII 450MHz has nearly double the speed of > a SGI R10000 O2 in calculating with modeler). > > Your help is welcome. > > Yours Volker > -- -- Andrej Sali, Assistant Professor The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399 tel +1 212 327 7550; lab +1 212 327 7206 ; fax +1 212 327 7540 e-mail sali(a)rockvax.rockefeller.edu; http://salilab.org

1 0

What is MPRMCNS?
by Susan Crennell 03 Nov '98

03 Nov '98

Hello, I am a new Modeller user and I've managed to read my sequence alignment in and CHECK it using the model-default.top script, Modeller even writes out a .ini initial model, but then stops saying: mkspline_E> too many parameters, increase MPRMCNS Please could you tell me how to do this, or how I can get round this problem? Thank you, Susan Crennell

2 1

Re: Modeller on linux
by Lutfullah Kakakhel 18 Oct '98

18 Oct '98

Hello, I have discovered that 'i386-linux' is set in my linux kernel and it would require re-compilation of kernel to change it to 'i486-linux' or whatever and that it also depends on the system and the configuration file for re-compilation. Under the circumstances what is the best possible option? Should the relevant scripts read the system data base and then alter the name of the file and directories accordingly or is it a must to retain the 'i486' and one should go through the hard task of re-compiling the kernel. I am not getting any errors on Run_tests and 'mod4 model-default' but no file 1fdx.B9999001 is produced. Instead I get a core dump. Is this due to the low memory and swap space I have (32 MB each) or something else. Since I am going in single steps, I am receiving newer problems each time i go ahead with making the program run. Looking forward to hear from you (Andreji Sali) or any other linux user of MODELLER. Lutfullah

1 0

Re: Modeller on linux
by Lutfullah Kakakhel 16 Oct '98

16 Oct '98

Hello again, Is it possible to increase the limit of 1000 on residues to about 1150 without switching to the larger moddler? I have the linux SMALL executable. Is the larger version available for LINUX? Lutfullah

1 0

Re: Modeller on linux
by Lutfullah Kakakhel 15 Oct '98

15 Oct '98

>I am not sure what is going on without looking at your system ... > >Try this: > >cd /usr/local/bin/modeller4/bin >rm mod4 >mv mod4_linux-i486 mod4 ^ ^ 4? is this a typo? The one I found on the ftp site was mod_i486-linux-SMALL.Z >chmod a+x mod4 > >Also make sure /usr/local/bin/modeller4/bin is in your execution path >(echo $path). > Its now working. Strange as it may sound but my linux turns hosttype into i386-linux on its own no matter what you put in the .cshrc or even the system-wide /etc/profile. Each time I boot, the hosttype is turned into i386-linux. I used to force it into i486-linux in the step ./Install but then the script /bin/hosttype turns it back into i386-linux which gives an error on creation of the proper directory and "mod4" in its proper place thereby causing the subsequent errors i reported earlier. I changed your executable to mod_i386-linux (after all it is an executable) where it existed and apparently it has worked. Run_tests did not cause any error and has successfully gone through. I will try to find out where linux stores its machine type but may be there is no need. Thanks once again, Lutfullah >Best, Andrej > > >Lutfullah Kakakhel wrote: >> >> Thanks for your prompt reply. >> I am running csh and i have the suggested line in the .cshrc file. >> When I type: >> # hosttype i get the reply >> i486-linux >> when i type simply: >> #mod4 (with nothing else) i get >> /usr/local/bin/modeller4/bin/mod4: hosttype : command not found >> /usr/local/bin/modeller4/bin/mode: /usr/local/bin/modeller4/bin/mod_: no >> such file or directory. >> Have I installed the program in the right directories? >> >> I just copied the zipped archives to /usr/local and did whatever is written >> in the installation procedure. >> No error is generated till the end. But >> ./Run_tests >> produces the same errors: >> >> ./run_tops1:mod4: command not found >> >> Lutfullah >> >> -----Original Message----- >> From: Andrej Sali <sali(a)rockvax.rockefeller.edu> >> To: Lutfullah Kakakhel <seen(a)paknet2.ptc.pk> >> Cc: modeller_usage(a)salilab.org >> <modeller_usage(a)salilab.org> >> Date: Thursday, October 15, 1998 11:32 AM >> Subject: Re: Modeller on linux >> >> >Please define >> > >> >setenv HOSTTYPE i486-linux >> > >> >in your .cshrc file. >> > >> >Best, Andrej >> > >> > >> >> Lutfullah Kakakhel wrote: >> >> >> >> Hello, >> >> >> >> I just subscribed to the list and have the following to resolve: >> >> >> >> 1. I am trying to install MODELLER on a REDHAT 5.0 Linux system >> >> which is a Pentium 133MH, 32 MB RAM and 1.2 GB hard disk. Would it >> >> give me troubles >> >> later and if yes what kind? >> >> >> >> 2. The installation passes through all the stages successfully but >> >> i get a screenful of error messages: >> >> ./run_tops1: mod4: command not found >> >> when i try to test run as suggested in the documentation. >> >> >> >> 3. The system prompts a 'hosttype' i386-linux during the ./Install >> >> while the executable i have downloaded is mod_i486-linux. Which one >> >> would require change, if does? Does the above error has anything to >> >> do with this? >> >> >> >> Sorry if these sound dumb questions but i am totaly new in this field. >> >> >> >> >> >> Lutfullah >> >> >> >> >> > >> >-- >> >Andrej Sali, Assistant Professor >> >The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399 >> >tel +1 212 327 7550; lab +1 212 327 7206 ; fax +1 212 327 7540 >> >e-mail sali(a)rockvax.rockefeller.edu; http://salilab.org >> > > >-- >Andrej Sali, Assistant Professor >The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399 >tel +1 212 327 7550; lab +1 212 327 7206 ; fax +1 212 327 7540 >e-mail sali(a)rockvax.rockefeller.edu; http://salilab.org >

1 0

Re: Modeller on linux
by Lutfullah Kakakhel 15 Oct '98

15 Oct '98

Thanks for your prompt reply. I am running csh and i have the suggested line in the .cshrc file. When I type: # hosttype i get the reply i486-linux when i type simply: #mod4 (with nothing else) i get /usr/local/bin/modeller4/bin/mod4: hosttype : command not found /usr/local/bin/modeller4/bin/mode: /usr/local/bin/modeller4/bin/mod_: no such file or directory. Have I installed the program in the right directories? I just copied the zipped archives to /usr/local and did whatever is written in the installation procedure. No error is generated till the end. But ./Run_tests produces the same errors: ./run_tops1:mod4: command not found Lutfullah -----Original Message----- From: Andrej Sali <sali(a)rockvax.rockefeller.edu> To: Lutfullah Kakakhel <seen(a)paknet2.ptc.pk> Cc: modeller_usage(a)salilab.org <modeller_usage(a)salilab.org> Date: Thursday, October 15, 1998 11:32 AM Subject: Re: Modeller on linux >Please define > >setenv HOSTTYPE i486-linux > >in your .cshrc file. > >Best, Andrej > > >> Lutfullah Kakakhel wrote: >> >> Hello, >> >> I just subscribed to the list and have the following to resolve: >> >> 1. I am trying to install MODELLER on a REDHAT 5.0 Linux system >> which is a Pentium 133MH, 32 MB RAM and 1.2 GB hard disk. Would it >> give me troubles >> later and if yes what kind? >> >> 2. The installation passes through all the stages successfully but >> i get a screenful of error messages: >> ./run_tops1: mod4: command not found >> when i try to test run as suggested in the documentation. >> >> 3. The system prompts a 'hosttype' i386-linux during the ./Install >> while the executable i have downloaded is mod_i486-linux. Which one >> would require change, if does? Does the above error has anything to >> do with this? >> >> Sorry if these sound dumb questions but i am totaly new in this field. >> >> >> Lutfullah >> >> > >-- >Andrej Sali, Assistant Professor >The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399 >tel +1 212 327 7550; lab +1 212 327 7206 ; fax +1 212 327 7540 >e-mail sali(a)rockvax.rockefeller.edu; http://salilab.org >

2 1

Modeller on linux
by Lutfullah Kakakhel 14 Oct '98

14 Oct '98

Hello, I just subscribed to the list and have the following to resolve: 1. I am trying to install MODELLER on a REDHAT 5.0 Linux system which is a Pentium 133MH, 32 MB RAM and 1.2 GB hard disk. Would it give me troubles later and if yes what kind? 2. The installation passes through all the stages successfully but i get a screenful of error messages: ./run_tops1: mod4: command not found when i try to test run as suggested in the documentation. 3. The system prompts a 'hosttype' i386-linux during the ./Install while the executable i have downloaded is mod_i486-linux. Which one would require change, if does? Does the above error has anything to do with this? Sorry if these sound dumb questions but i am totaly new in this field. Lutfullah

2 1