[gmx-users] Re: Gromacs_trouble

[Justin A. Lemkul]

Please keep this discussion on the gmx-users list.  I have only limited 
experience with MARTINI, but there are others on the list more experienced than 
I.  See comments below.

Du Jiangfeng (BIOCH) wrote:
> Dear Justin,
> 
> Thank you very much again of your help.
> 
> So far, I appended some sentences to make carboxylated residues in Martini force field and I got my protein model. However, no matter how to perform EM, the MD simulation was always blown up because of too many links warnings.
> 
> I presume the problem is still in the values of the modified residues in the itp file. 
> 
> What I did is :
> 1   the usage of "atom2cg_v2.1.awk" converting my GLA domain into coarse grained model. (In "atom2cg_v2.1.awk", I added new lines for GLA).
> 2  the usage of "seq2itp.pl" making itp file for my GLA domain. (In "seq2itp,pl", I appended some lines for GLA residue).
> 
> That's it.
> 
> But it seems not to work. 
> 
> Does my way is correct? Or should I shift to another method to figure it out? 
> 

There is nothing wrong with the way you've constructed the topology, from what I 
can see.  That's the standard MARTINI protocol for proteins, anyway.  What's 
more likely the problem is the values you've assigned either for bonded or 
nonbonded interactions of your new species.  You said earlier you assigned some 
arbitrary values, but you haven't provided any of that information.

> I also attach some files that I changed and the GLA domain. (In the files of "atom2cg"  and "seq2itp", the things what I added were marked with "#append", so you can track them easily).
> 

The structure file emcgmR.gro shows that residue CGU-26 is especially strange. 
It appears as if it is shearing apart.  I suspect that may be part of your problem.

-Justin

> Thank you in advance,
> Best Wishes,
> Jiang. 
> 
>  
> 
> 
> Du Jiangfeng (BIOCH) wrote:
>> Hi Guys,
>>
>> I want to run a protein which contains some carboxylated residues into a membrane. I have had to add a special residue into itp file since there is no any description in normal itp for GLA residue. I admit some values I added for GLA residue were ambiguous because I don't know the exact values.
>>
>> After that, the system was blowing up when MD simulation with the following warning:
>>
>> Step 0, time 0 (ps)  LINCS WARNING
>> relative constraint deviation after LINCS:
>> rms 0.029057, max 0.152326 (between atoms 81 and 82)
>> bonds that rotated more than 30 degrees:
>>  atom 1 atom 2  angle  previous, current, constraint length
>>      43     44   37.0    0.0145   0.0148      0.0145
>>      81     82   67.5    0.0078   0.0090      0.0078
>>      89     90   34.7    0.0059   0.0060      0.0059
>>
>> Step 0, time 0 (ps)  LINCS WARNING
>> relative constraint deviation after LINCS:
>> rms 0.056864, max 0.359219 (between atoms 81 and 82)
>> bonds that rotated more than 30 degrees:
>>  atom 1 atom 2  angle  previous, current, constraint length
>>      43     44   35.6    0.0145   0.0147      0.0145
>>      81     82   78.6    0.0078   0.0106      0.0078
>>      89     90   34.3    0.0059   0.0060      0.0059
>>
>>
>> Apparently, some values about GLA residue are wrong in itp file. But I really don't know how to set a correct value for it.
>>
> 
> Without seeing what you actually did and knowing what force field you're using, 
> it's impossible to help.  Missing parameters need to be derived in a manner 
> compatible with the original force field (no small task).
> 
> http://www.gromacs.org/Documentation/How-tos/Parameterization
> 
> -Justin
> 

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

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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