[gmx-users] Re: g_order in GROMACS 3.3.2 and 3.3.1

[Stephen M. Dutz]

Hi,

I've just begun attempting to use g_order on a protein/lipid POPC system.  There are a couple things different that I do than you are currently doing.

1) Make sure that the .pdb file you are using with g_order is a trajectory file and not merely a structure file.  See this page to double check your input files: http://www.gromacs.org/documentation/reference/online/g_order.html
2) You are grouping all the Carbons from one tail of the DPPC into one group.  What you need to do is group all like carbons in individual groups (from that tail).
       i.e. Have one group that is all DPPC C17 atoms in the system, one group that is all DPPC C18 atoms, and so on, up to C31. Thus you would end up having 15 groups.  Also, a name for the group will be generated automatically when you create it, so you can skip that part. 
3) When you launch g_order it has default inputs and outputs, only if you deviate from these files do you need to specify.  
    i.e. You use,  g_order -f YYY.pdb -n index.ndx -s .tpr -od, you are telling the program that your topology input file is ".tpr", I'm assuming this isn't the name of the file, and thus this could potentially be one of your problems.  I believe you need to specify the full file name (correct me if I am wrong).  Also it will output a deuterium file by default called deuter.xvg.  You only need to include the -od tag, followed by a name, if you want to alter this name.  That being said, assuming you have a "generic run input file" of topol.tpr and an index file called index.ndx, you would launch g_order through the following command:
g_order -f YYY.pdb

It will then output order.xvg and deuter.xvg that can be visualized in xmgrace.  

There is my 2 cents.  I've found that it is quite difficult to get a reply from anyone on this board with regards to g_order.  I'm not sure if it is because no one knows how to use it or what...  There are also a few past posts by people, however most remained unanswered.  Good luck!

Steve

----------
"Mediocrity breeds insecurity and paranoia."
- S. Sharma
----------



Dear all GROMACS user :
    After generating the .ndx file which contained only one group that one tail of DPPC,
    I use g_order to calculate order parameters. But there are nothing in deuter.xvg and order.xvg file.
    The process is listed below:
    step1: make_ndx -f XXX.pdb -o index.ndx
             then type : a C17 C18 ... C31     (C31 is methyl and C17...C30 is methylene in one tail of DPPC lipid)
             then type : name 3 sn_1        (0 is for system, 1 is for DPPC, 2 is for SOL )
             then type : q

    step2 : delete all groups in the index.ndx except sn_1 group.

    step3 : g_order -f YYY.pdb -n index.ndx -s .tpr -od

     Can someone tell me what wrong is in my process ?

         Thank you very much.