CHARMM Development Project Forums User Discussion & Questions Molecular Dynamics How to get the average structure of protein in a trajectory

I have run a trajectory for 1 ns, and I want to obtain the average structure of protein in the trajectory.

Have a look at the examples in the Script Archive forum, they can help you get started with basic tasks.
The command you need is "COOR DYNA" (corman.info):
read rtf card name ...
read para card name ..
read psf card name ..
read coor card name starting_coordinate_file
coor copy comp
open file unit 101 name name_of_your_trajectory_file
coor dyna orient firstunit 101 nunit 1 select segi segment_id_of_your_protein end
write coor card name average.crd select segi segment_id_of_your_protein end
* average coordinates over 1 ns
*

Keep in mind that an average structure may be highly non-physical, esp. internal coordinates such as bond angles and dihedrals. I often choose to obtain a real coordinate set that is closest to the average in terms of RMS difference for Calphas of conserved structural elements such as helices, sheets, and turns.

Totally agree with Rick, who as usual focuses on science. This is something you should discuss at depth with you supervisor. One may also add that:
1/ 1 nanosecond may be too short
2/ It may be more relevant to describe the trajectory either in terms of a real structure close to the average as Rick suggests, or in terms of representative members of the main clusters obtained from a clustering analysis.