[gmx-users] R: problems running REMD on grids

[Mark Abraham]

Anna Marabotti wrote:
> I'm re-sending the message below because it seems to me that it has not
> arrived to the list.
> Please forgive me in case of duplicate sending.
> Best regards
> Anna
> 
>   _____  
> 
> Da: Anna Marabotti [mailto:anna.marabotti at isa.cnr.it] 
> Inviato: venerdì 11 settembre 2009 14.05
> A: 'gmx-users at gromacs.org'
> Oggetto: problems running REMD on grids
> 
> 
> Dear gmx-users,
> 
> I’m trying to perform some REMD calculations with GROMACS 4.0.5, and I found
> on the Wiki site these instructions:
> 
>  
> 
> http://oldwiki.gromacs.org/index.php/REMD
> 
>  
> 
> that I’m trying to follow, but I encountered some problems.
> 
> If I understand well the instructions, if I want to perform replicas at N
> different temperatures, I have to create N different .mdp files and then N
> different .tpr files, that I should name “filename”_0, “filename”_1...
> "filename”_N-1. That’s what I did: I created 16 different .mdp files and 16
> .tpr files that I called REMD_0.tpr. REMD_1.tpr REMD_2.tpr ... REMD_15.tpr.
> At this point, I have to launch the mdrun command with the –multi, -replex
> (optionally, -reseed) options. Following the example in "Execution step", my
> command line should be:
> 
>  
> 
> mdrun –s REMD_.tpr –multi 16 –replex 1000 –deffnm REMD_  -np 16
> 
>  
> 
> That's what I did, but I experienced some difficulties on the parallel
> machines that I’m using to perform simulations. In particular, I inserted
> the above command into a script that is used on the machine to launch the
> program in an interactive way, but GROMACS returns an error saying that
> “cannot open file REMD_.tpr” (and in fact, the file REMD_.tpr does not
> exist, only files REMD_0.tpr, REMD_1.tpr etc do exist!) I also retried
> launching the same command from the shell, but the error was the same.
> 
> So my question is: how does GROMACS “understand” that it has to interpret
> “REMD_.tpr” not as a single file, but as a group of files indexed _0, _1, _2
> and so on? Does it depend on the order of the options? Could you please give
> me some hints about?

I'd guess you're confusing the mechanism that internally appends a 
suitable integer to the filename by providing both -s and -deffnm. 
Please try without the redundant -s argument and let us know if that 
worked. If so, perhaps this bit of code needs a small fix.

Mark

> BTW, another problem: I tried to find solution searching in the
> oldwww.gromacs.org site, but I am still unable to consult the gmx-users
> archives using a keyword as it was possible in the past. How can I find
> information in the GROMACS archives?
> 
>  
> 
> Many thanks and regards
> 
> Anna
> 
> ____________________________________________________
> Anna Marabotti, Ph.D.
> Laboratory of Bioinformatics and Computational Biology
> Institute of Food Science, CNR
> Via Roma, 64
> 83100 Avellino (Italy)
> Phone: +39 0825 299651
> Fax: +39 0825 781585
> Email: anna.marabotti at isa.cnr.it
> Skype account: annam1972
> Web page: http://bioinformatica.isa.cnr.it/anna/anna.htm
>  
> "If you think you are too small to make a difference, try sleeping with a
> mosquito"
>  
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php