Re: CCL:protein cross the boundary of the box in PBC

Hi, Rick,
 Would you mind I ask you a question about the relationship between 1-4
 interaction and the NPHI, the number of dihedrals in the psf file? I have
 a very simple peptide system composed of 5 residues without using periodic
 boundary condition. From the psf file, I know that 210 !NPHI. I check
 that all the 210 dihidrals in the psf file are unique, i.e, their is no
 case like A-B-C-D, and A-e-f-D, or D-g-h-A in the dihedral part. Here, I
 use A, B, C, and D to denote the number of atoms, like 12 14 17 19 in the
 dihedral part of the psf file, etc. Hence, I think the
 number of 1-4 interactions should be equal to 210 when the nonbonded
 interaction is calculated. However, from the output of CHARMm,
 it says "with mode 5 found 230 exclusions and 205 interactions(1-4)".
 Obviously, 205 is less than 210. What is wrong with my thought?
 By the way, if I use the periodic boundary condition, what does that mean
 for 1-4 interaction? For eample, if I have A-B-C-D in the primary box and
 A'-B'-C'-D' in one of the image cells, how the 1-4 interaction is
 calculated if distance of AD' is less than AD? I think, in this case,
 CHARMm will neglect AD interaction, but consider AD' interaction. If AD'
 is calculated, CHARMm use 1-4 VDW parameters or other VDW parameters?
 I am sorry to ask you this question. I have posted this question in the
 CHARMm bbs but without reply.
 Thanks very much for your help,
 Wei Shi