On Thu, Dec 19, 2013 at 7:47 AM, <agarai.physics.iisc.ernet.in> wrote:
>
> Hi all,
>
> I want to calculate the end-to-end distance of a DNA molecule using ptraj.
>
> I used the following input file:
>
> trajin md_prod_38_ns.trj
> trajin md_prod_39_ns.trj
> trajin md_prod_40_ns.trj
> trajin md_prod_41_ns.trj
> trajin md_prod_42_ns.trj
> trajin md_prod_43_ns.trj
> trajin md_prod_44_ns.trj
> trajin md_prod_45_ns.trj
> trajin md_prod_46_ns.trj
> trajin md_prod_47_ns.trj
> trajin md_prod_48_ns.trj
> trajin md_prod_49_ns.trj
> trajin md_prod_50_ns.trj
> distance dist_tail_nuc :1,76 :38,39 out distance_dna_end
>
Try adding the "noimage" keyword here. If your DNA chain is large compared
to the size of the box, then the distance cpptraj is calculating might be
the distance between the head of the nucleotide sequence to the tail of the
nucleotide sequence in the adjacent box. By default, cpptraj always takes
the minimum distance between two masks among all of the periodic images.
The mask :1,76 corresponds to resid of the residues at one end of the DNA
> and the mask :38,39 correspond to the residues on the other end. Using
> This input file, ptraj must calculate the distance between the center of
> mass of residues 1,76 and 38,39. But the values that I get as output are
> very much smaller than the expected values. I have checked the trajectory
> files and the .prmtop file and they seem to be fine. I am in doubt,
> whether my input file is wrong or is there some other source for the
> error. Kindly help.
>
Distances are very easy to compute, and as such are not very error-prone.
If the image issue is not causing the lower-than-expected distances, then
you should visualize your trajectory in VMD (or some related program) to
see if you can find anything wrong. This is _always_ a good idea (and with
tools like VMD you can always check that the distances you measured in
cpptraj are 'correct').
Good luck,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Dec 19 2013 - 08:00:03 PST
