[gmx-users] interaction lists in idef.h

[David van der Spoel]

wnoid at hec.utah.edu wrote:
> hello again.
> 
> 
> thank you very much to drs. Groenhof, Abraham, and van der Spoel.  i 
> greatly appreciate your time and help.  dr. Groenhof's email and the pdf 
> file that he linked were particularly helpful.  i would suggest that the 
> pdf file would be a great addition to the gromacs wiki.
> 
> 
> with your help, i now can access the lists of interacting pairs, 
> triples, etc. for each type of bonded interaction.  however, i am still 
> having difficulty accessing the lists for the nonbonded interactions.  
> as the comments in types/idef.h suggests, idef.il[F_LJ].nr=0, so the 
> lists of pairs of nonbonded interactions are not stored there.
> 
> 
> dr. van der Spoel mentioned that the nonbonded interactions are 
> organized as a matrix at the beginning of the idef.  i can see that the 
> first interaction types (and their corresponding sets of iparams and 
> functypes) correspond to these nonbonded interactions.  however, i do 
> not understand how to access the information that must be stored in this 
> interaction matrix.  (i don't quite understand where this interaction 
> matrix is in the idef structure either.  sometimes these c structures 
> are a little bewildering to me - fortran is my native programming 
> language.)  if someone could indicate how to access these lists of 
> nonbonded pairs that would be greatly appreciated.
> 
> 

the list of nonbonded pairs is determined dynamically to accomodate for 
diffusion etc. what you find in idef is the interaction parameters. each 
atom has an atom type, and the types of two atoms interacting are then 
used as index in the matrix.

in other words there is no list of nonbonded pairs in the tpr file.

-- 
David van der Spoel, Ph.D.
Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205. Fax: +4618511755.
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se