Trajectory over time

[drknnetz]

Hello all, I am brand new to using dssr and x3DNA. I started yesterday and it seems like a pretty awesome download so thanks for creating this! I was wondering if there is anything in the works for analyzing an entire trajectory? I know this current dssr analyzes pdb files and not an mdcrd file so I'm sure there is a lot more that goes into it. I was just wondering if there was anything in the works.
Thanks,
Dennis

[xiangjun]

Hi Dennis,

Thanks for using DSSR and 3DNA, and for posting your feature request on the Forum!

As to the analysis of trajectories of molecular dynamics (MD) simulations, please note the following:

• 3DNA -- you may want to try the x3dna_ensemble Ruby script, or "do_x3dna: A tool to analyze structural fluctuations of dsDNA or dsRNA from molecular dynamics simulations" by Kumar and Grubmuller, recently published in Bioinformatics.
• DSSR -- there is no similar functionality yet, but I'd very much to make DSSR more directly relevant to the increasingly important MD field. Not a practitioner of simulations myself, I need some MD expert to work with. Anyone interested in such a joint project?

Best regards,

Xiang-Jun

 

[drknnetz]

Thanks for the response, I used the ruby script and it was great. Trying to figure out a way to edit it to work for the dssr but no luck yet. I also found it useful to compile the trajectory into an output pdb file with "time stamps" of the simulation for each step. So basically my one pdb file has 500 time stamps in it and I am using scripts to analyze those. That may work for others in similar fashion. I'm still a pretty weak programmer though, so it is taking some doing.

[xiangjun]

Hi Dennis,

Thanks for posting back with your progress. For the x3dna_ensemble Ruby script distributed with 3DNA v2, the MODEL/ENDMDL ensemble PDB format was deliberately selected as the starting point. For one thing, that's what an NMR ensemble is represented in the PDB. More importantly, different MD packages (as I am aware of) have different binary formats that are beyond what I'd like to handle in the script. Presumably, each MD package should have an option to output snapshots of a trajectory into the standard MODLE/ENDMDL ensemble PDB format.

So basically my one pdb file has 500 time stamps in it and I am using scripts to analyze those. That may work for others in similar fashion. I'm still a pretty weak programmer though, so it is taking some doing.

Your 500 time stamps are well within the comfort zone that a x3dna_ensemble-like script can handle. If you have any specific questions, please post them on the Forum. I'd be more than happy to help you through the technical details. It's a learning experience for all of us. It'd be great you can share your script in the end for the benefit of other users.

Xiang-Jun