[gmx-users] g_rmsf and pbc

[Mark Abraham]

tangxuan wrote:
> Dear all,
>    The protein I am working on is rubisco, consisting of 8 identical
> large subunits and 8 identical small subunits. I try to calculate RMSF
> for each large subunits, but the rmsf values seems  much large to some
> large subunits. So I check the first frame of the protein structutre,
> and I found that these large subunits having large rmsf values seems
> separate in the box. Then I tried to use trjconv -pbc nojump to remove
> the jump for each of them in the xtc file. When I use new xtc file and
> original tpr file to calculate the rmsf, it shows " Segmentation fault"
> error. This may be because  the coordinators of atoms in the tpr file
> are very diffrent from that in the fist frame. Do you know how to solve
> this problem?

GROMACS does an inexplicable segfault almost never. It's highly likely 
that either there's more of an error message than you've said, or that 
the problem is the result some catastrophic filesystem or OS issue.

Mark