[gmx-users] regarding gen-pairs

[Mark Abraham]

On 30/03/2010 1:57 AM, Gavin Melaugh wrote:
> Hi guys
>> Sorry about the wrong subject in the previous e-mail. (I just hit reply).

It's preferable to start a new email entirely. Archival software 
sometimes uses the information in email headers, which might not match 
up properly if people start emails from the middle of other discussions' 
threads.

>> I have been running gromacs for a few months now with no real problems
>> however some of my numbers differ from my supervisor's who uses DL-POLY.
>> Anyway discussion of the discrpnacies has led me to questioning aspects
>> of my topology file below (exerpts). We created the topology ourselves.
>> I always thought that the set up below would calculate 1-4 interactions,
>> but having read through the archives I am now confused. Do I have to
>> give a pair list in topology file, or is gen-pairs = yes sufficient. I
>> never created a pairlist as I thought that gen-pairs would create it
>> from the parameters in atomtypes.

The example in manual section 5.7.1 seems to describe the use of 
gen_pairs well enough. 5.3.4 may also help. You need to be clear about 
which force field you are using and how its parameters are expected to 
be combined to produce 1-4 interactions. You haven't given us enough 
information to help with that...

>> ;Parameter level
>> [defaults]
>> ; nbfunc    comb-rule     gen-pairs        fudgeLJ     fudgeQQ
>>       1             3                    yes
>> 0.5           0.5
>>
>> [atomtypes]
>> ;type     mass           charge      ptype     sigma(nm)
>> epsilon(kjmol-1)
>>     CB     12.011000      0.000000       A      0.355000      0.292880
>>     CA     12.011000     -0.115000       A      0.355000      0.292880
>>     HC      1.008000      0.115000       A      0.242000      0.125520
>>     CU     13.019000      0.265000       A      0.350000      0.334720
>>     NU     14.007000     -0.597000       A      0.325000      0.711280
>>     CH     13.019000      0.332000       A      0.385000      0.334720
>>     C3     15.035000      0.000000       A      0.390500      0.732200
>>     C2     14.027000      0.000000       A      0.390500      0.493712
>>
>> ;Molecular level
>> [moleculetype]
>> ;       name         nrexcl
>>         subhexylcage      3
>>
>> [atoms]
>> ; atomnr  type   resnr  residue   name    cgnr     charge     mass
>>      1      CA       1    CGE       CA       1     -0.1150     12.0110
>>      2      CB       1    CGE       CB       1      0.0000     12.0110

If this is your topology, you have much bigger problems than pairlists. 
You have only two atoms defined such that your molecule has a net 
partial charge. If you're exerpting stuff, please indicate that explicitly.

Mark

>> [bonds]
>> ;  ai    aj   func   b0(nm)       kb(kjmol-1 nm-2)
>>      1     2     1     0.1400      392459.2
>>      1     6     1     0.1400      392459.2
>>
>> [angles]
>> ;  ai    aj    ak   func  theta(deg)    ktheta(kjmol-1 rad-2)
>>      2     1     6     1     120.00      711.2800
>>      2     1     7     1     120.00      292.8800
>>
>> [dihedrals]
>> ;  ai    aj    ak    al    func       V1           V2
>> V3            V4 (all kj mol-1)
>>      1     2     3     4       5       0.00000      30.33400
>> 0.00000       0.00000
>>      1     2     3    10       5       0.00000      30.33400
>> 0.00000       0.00000
>>
>> ;System level
>> [system]
>> ;the name of this system
>> config test of 32 subhexyl cage molecules
>>
>> [molecules]
>> ; moleculename     number
>>   subhexylcage        32
>>
>>
>