[gmx-users] energy groups

[Justin A. Lemkul]

Liu Shiyong wrote:
> Hi, Mark,
> 
> I tried your suggestion.
> 
> + pdb2gmx -ignh -ff G53a6 -f b.pdb -p a6.top -i a6.posre.itp -o a6.pdb
> + editconf -f a6.pdb -o a6.gro -d 20.0
> + make_ndx -f a6.pdb -o a6.ndx
> + grompp -maxwarn 10 -f em.mdp -c a6.gro -n a6.ndx -p a6.top -o a6.input.tpr
> 
> title               =  b
> cpp                 =  /usr/bin/cpp
> integrator          =  cg
> dt                  =  0.002    ; ps !
> nsteps              =  1
> rlist               =  0.55
> nstlist             =  0
> vdwtype             =  Cut-off
> rvdw                =  0.6
> coulombtype         =  Cut-off
> epsilon_r           =  2
> rcoulomb            =  0.6
> energy_grps         =  chAANDB_&_!H* chC_&_!H*
> ;
> ;       Energy minimizing stuff
> ;
> emtol               =  100.0
> emstep              =  0.01
> 
> I got the same error.
> 
> -------------------------------------------------------
> Program grompp, VERSION 4.0.2
> Source code file: grompp.c, line: 150
> 
> Fatal error:
> atoms 1 and 2 in charge group 1 of molecule type 'Protein_A' are in
> different energy groups
> -------------------------------------------------------
> 

Now I see it.  This error is coming up because the atoms are bonded together. 
Atoms 1 and 2 are N and H of the N-terminus, right?  Actually, Mark explained 
this one some time ago:

http://www.gromacs.org/pipermail/gmx-users/2008-January/031725.html

Also, the cut-off's you've chosen (and using cut-off for electrostatics) are 
likely to lead to artifacts.  Interestingly, this is the third time in as many 
days as I have seen bizarre values of rlist, rcoulomb, and rvdw in conjunction 
with Gromos96...

-Justin

> "It's Against the Rules" (Pulp Fiction)
> 
> 
> 
> 
> On Mon, Feb 2, 2009 at 7:05 PM, Mark Abraham <Mark.Abraham at anu.edu.au> wrote:
>> Liu Shiyong wrote:
>>> command:
>>>
>>> + pdb2gmx -ignh -ff G53a6 -f b.pdb -p a6.top -i a6.posre.itp -o a6.gro
>>> + pdb2gmx -ignh -ff G53a6 -f b.pdb -o a.pdb
>>> + editconf -f a6.gro -o a6.gro -d 20.0
>>> + make_ndx -f a.pdb -o a6.ndx
>>> + grompp -maxwarn 10 -f em.mdp -c a6.gro -n a6.ndx -p a6.top -o a6.input.tpr
>>
>> I think you still don't need two invocations of pdb2gmx. Your only constraint is that you want chain IDs preserved for your make_ndx command. The PDB format carries the system size just like the GRO format. Thus
>>
>> + pdb2gmx -ignh -ff G53a6 -f b.pdb -p a6.top -i a6.posre.itp -o a6.pdb
>> + editconf -f a6.pdb -o a6.gro -d 20.0
>> + make_ndx -f a6.pdb -o a6.ndx
>>
>> + grompp -maxwarn 10 -f em.mdp -c a6.gro -n a6.ndx -p a6.top -o a6.input.tpr
>>
>> Simpler still is to define your index groups using information other than the chain ID - as I think someone suggested days ago. Ranges of atom numbers are also effective.
>>
>>
>>>        -------------------------------------------------------
>>>                      Program grompp, VERSION 4.0.2
>>>                      Source code file: grompp.c, line: 150
>>>                      Fatal error:
>>>                      atoms 1 and 2 in charge group 1 of molecule type
>>>               'Protein_A' are
>>>                      in different energy groups
>>>                            -------------------------------------------------------
>>>
>>>
>>>    There is still some sort of disconnect between topology numbering
>>>    (which is where charge groups are defined), and the structure you
>>>    are using.  Can you post the first *few* lines of your .pdb file, [
>>>    atoms ] section of the .top, and the relevant index groups?
>>
>> Not necessarily - the assignment of energy groups in the .mdp file is still a potential source of problems.
>>
>> Mark
>>
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.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
> 
> 
> 
> 

-- 
========================================

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================