[gmx-users] diffusion constant level off

[Ahmet yıldırım]

Dear users,

I used the following commands to get diffusion constants (every 10 ns) of a
simulation of 100 ns . The number of frame is 5001 (write .xtc trajectory
every 20 ps).  I looked at RMSD vs average structure, RMSD vs starting
structure, Radius of gyration, RMSD matrix. This simulation has reached to
converge at last 50 ns.
g_msd -f traj.xtc -s topol.tpr -o msd_1.xvg -b 0 -e 10000
g_msd -f traj.xtc -s topol.tpr -o msd_2.xvg -b 10000 -e 20000
g_msd -f traj.xtc -s topol.tpr -o msd_3.xvg -b 20000 -e 30000
...
g_msd -f traj.xtc -s topol.tpr -o msd_10.xvg -b 90000 -e 100000

1.) I used the above commands without the following flags ( -type, -lateral
and -ten). Which diffusion will the above comands give? is it bulk
diffusion?
Gromacs manual:
-type:Compute diffusion coefficient in one direction:no, x, y or z
-lateral:Calculate the lateral diffusion in a plane perpendicular to: no,
x, y or z
-ten:Calculate the full tensor
2.) I plotted diffusions (10 values) as function of time. Diffusions dont
converge. Did I do any steps by mistake?
3.) From manual:
The diffusion constant is calculated by least squares fitting a straight
line (D*t + c)...
What is (D*t + c)? What are the meaning of D and c?
4.) What should be "Time between restarting points in trajectory"?

Thanks in advance
-- 
Ahmet Yıldırım