[gmx-users] g_rama/xrama problem in version 3.1.4.

[istvan at kolossvary.hu]

Hi, 

I ran a 1ns md on ACE-ALA-NAC (sometimes called alanine dipeptide) and 
wanted to look at the Ramachandran plot. Both g_rama and xrama abort after 
finding the correct number of a single phi-psi pair. A fatal error message 
comes from nrama.c that reads something like "Dihedral about 6,8 cannot be 
found in topology". "6" is a C=O oxygen of ACE and "8" is a NH hydrogen of 
ALA, there shouldn't be any dihedral between them, should it. I guess the 
topology should be fine since I can run a whole simulation with it. I wonder 
if this has something to do with the same problem that I posted yesterday 
(atomtype mismatch in the .rtp and .hdb files for NAC). Can you give me a 
hint. I am kind of discouraged since I am doing very simple things. 

Many thanks, 

Istvan 

P.S. Erik had this idea: 

Hi, 

Ah, I'm almost positive it's just a problem in the analysis code. 

I didn't write the core of g_rama myself, so you might want to post it to 
the mailing list where Berk & David read things too. 

ACE-ALA-NAC isn't a dipeptive, but a capped single amino acid. g_rama 
detects phi/psi dihedrals automatically, and the original gromacs/gromos87 
forcefields didn't use caps a lot, so my guess would be that the automatic 
code to find the phi/psi dihedrals based on atom names fails. 

Try it with a normal terminated (NH3+ and COO-) ALA-ALA and see if that 
works. If it does, it is just the detection code in g_rama that needs to be 
updated; you might even want to have a look, it's a fairly small program. 

Cheers, 

Erik 

> 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.