[gmx-users] how to interpretate a tpr file

Yang Ye leafyoung81-group at yahoo.com
Mon Jul 10 19:53:05 CEST 2006


Longshoremen Zhang wrote:
> On Mon, 2006-07-10 at 13:58 +0800, Yang Ye wrote:
>   
>> Dongsheng Zhang wrote:
>>     
>>> Yang Ye,
>>>
>>> Thank you very much for your help. I still have some questions:
>>>
>>> On Mon, 2006-07-10 at 11:35 +0800, Yang Ye wrote:
>>>   
>>>       
>>>> Dongsheng Zhang wrote:
>>>>     
>>>>         
>>>>> Dear GMX users:
>>>>>
>>>>>
>>>>> Could someone help me to understand these information from a position
>>>>> restrained tpr file?
>>>>>
>>>>> 1.   excl[0][0..17]={0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 13, 14, 15,     
>>>>>                      16, 18, 26, 28}
>>>>>   
>>>>>       
>>>>>           
>>>> Exclusion group 0, there are total 17 exclusions in this group. between 
>>>> 0 (the first one) and remaining 17 atoms.
>>>>     
>>>>         
>>>  Is this exclusion for nonbonded interactions, including 1-2, 1-3
>>> interactions? If so, I don't understand why there are 17 atoms to be
>>> excluded for atom 0. It is a capped carbon nanotube. Each C has three
>>> single covalent bond, so the total exluded atoms for each atom should be
>>> 6.
>>>
>>> This is next term for exlusion:
>>>
>>> excl[1][18..35]={0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 
>>>                13, 14, 15, 20, 22}
>>> Why are there two different exclusions for atom 0?
>>>   
>>>       
>> Please check your top file, including those files included into it. 
>> Remove/Change some and compare the dump.
>>     
>>>>> 2.   blocks[CGS][0][0..19]={0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 
>>>>>                12, 13, 14, 15, 16, 17, 18, 19}
>>>>>
>>>>> 3.   blocks[MOLS][2][1802..1805]={1802, 1803, 1804, 1805}
>>>>>   
>>>>>       
>>>>>           
>>>> Both are useful for dividing the atoms for parallel processing (before 
>>>> domain decomposition arrives in gmx 4.0 :-).
>>>> CGS means charge group. MOLS means molecules. Division shall not break 
>>>> one charge group or a molecule (hope that I am not wrong here).
>>>>     
>>>>         
>>>>> 4.   blocks[SBLOCKS][0][0..1039]={0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 
>>>>>                10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 
>>>>>                24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 
>>>>>                38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 
>>>>>                52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,  
>>>>>                65, ........... 1039}
>>>>>   
>>>>>       
>>>>>           
>>>> Not sure...
>>>>     
>>>>         
>>> Anyone can help me for this one? Thanks!
>>>
>>>   
>>>       
>>>>> 5.  functype[0]=LJ_SR, c6= 5.72640402e-03, c12= 1.86486031e-05
>>>>>  I know this is for the parameters of LJ potential. My question is where
>>>>> is the information that tells gromacs to use these parameters for the
>>>>> interaction between two specific atoms.
>>>>>   
>>>>>       
>>>>>           
>>>> see atoms: section; type there refers to the functype here. Any functype 
>>>> number for LJ_SR beyond the atom number there will not be used.
>>>>     
>>>>         
>>> I am sorry I am not clear here. I know C6 and C12 is for the LJ
>>> interaction between C and C in nanotube by looking at **nb.itp file. How
>>> can gromacs know it in tpr file (I believe all information should be in
>>> tpr file)? Let me give you an example. In tpr file, type=81 (ANGLES)
>>> 1041 1040 1042 shows that the angle bending parameters for atoms  1041
>>> 1040 1042 are from functype[81], but I can't find any information which
>>> says  the LJ parameters for C and C are from functype[0].
>>>   
>>>       
>> mdrun will associate functype[i] with atom type. check gmx source code 
>> if you are still not sure.
>>     
>
> my questions is how gromacs makes the association, especially for the LJ
> potential between two different atoms. Do you mean atom type in:
>
>  atom[     0]={type=  0, typeB=  0, ptype=    Atom, m= 1.20110e+01, q=
> 0.00000e+00, mB= 1.20110e+01, qB= 0.00000e+00, resnr=    0} grpnrs=[ 1 1
> 0 0 0 0 0 0 0 0 ]}
>
> The number 0 in "type=  0" is an index number for different atoms in the
> top file. It doesn't match with i in " functype[i] ". 
>   
I take one small system as an example. I got 126 atoms. So I have 
following lines in my dump file for tpr.

   topology:
      name="GGCC"
      atoms:
         atom (126):
...

 From the atom list below, the largest number in type is 12. So I have 
13 types of atoms.

In the later sections of the dump file, I have

      idef:
         catnap=13
         nodeid=0
         ntypes=296

Notice that here atnr=13. atnr means atom type number.

functype[0] to functype[168] are for LJ_SR (169 lines, so 169 sets 
parameters c6, c12). So they are the all possible combination of the LJ 
interaction parameters between any two types of atom. 13*13 = 169. Exact 
order of them can be interpreted by recalculation.

Yang Ye
> If you want me to check the source code, could please tell which portion
> of the source code can answer my question? I have checked sim_util.c,
> but haven't found the answer.
>
> Thank you for your help!
>
>
>   
>>>   
>>>       
>>>>> As someone has suggested, if gromacs website can offer the
>>>>> interpretation for each file, I believe many people will highly
>>>>> appreciate it.
>>>>>
>>>>> All the best!
>>>>>
>>>>> Dongsheng
>>>>> _______________________________________________
>>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>>>> 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
>>>>>
>>>>>   
>>>>>       
>>>>>           
>>>> _______________________________________________
>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>>> 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
>>>>     
>>>>         
>>> _______________________________________________
>>> gmx-users mailing list    gmx-users at gromacs.org
>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>> 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
>>>
>>>   
>>>       
>>
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> 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
>>     
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> 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
>
>   

-- 
/Regards,/
Yang Ye
/Computational Biology Lab
School of Biological Sciences
Nanyang Technological University
Singapore
Tel: 6316-2884
/




More information about the gromacs.org_gmx-users mailing list