VMD-L Mailing List

From: Laura Ernstsen (laura.ernstsen_at_uit.no)
Date: Wed Aug 21 2024 - 01:54:04 CDT

Sorry for possibly double posting since the first time I didn’t use the correct e-mail address.

Hello,

I am working with membrane simulations with 8 antimicrobial peptides (AMP) in the same simulation box. I am facing the following problem: after wrapping the system such that the membrane is centered at z=0, in some frames, some of the AMPs are situated between two periodic cells and long bond are being drawn between the parts. Previously when I have encountered this problem I have used the “join” command from pbc tools directly in the VMD TkConsole (pbc join connected -sel “AMP selection” -all

) while the whole dcd file has been uploaded and that has done the trick.

Now I am facing the same issue but VMD gives me the following error while using “join”: expected boolean value but got ""

What is different from before is that I have another membrane composition and size, another type of peptide (cyclic hexapeptide) and the number of waters and ions is different. In addition the simulations are run with NAMD a few years a part.

Separating the AMPs to their own .psf and .dcd file seems to fix the problem and the “join” tool works just fine, which is strange since the atom selection is identical in both cases. I could perhaps use separate .dcd files as a bypass, but I have yet to find a way to merge the two resulting .dcd files back together (AMP.dcd and MEMB.dcd).

I have double check the PBCtools webpage, github page, other users problems and also modifying the membrane wrapping such that the AMPs do not split between cells. I’ve also checked atom selection ambiquities by adding more precision (as in “resname TRP LYS and not segname MEMB”).

Does anyone have any clue how to fix this problem with “join”?

Best,

Laura Ernstsen