VMD-L Mailing List

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Fri Nov 04 2016 - 10:35:30 CDT

On Fri, Nov 4, 2016 at 10:43 AM, Vermaas, Joshua
<Joshua.Vermaas_at_nrel.gov> wrote:
> Ajasja is exactly right. Depending on how many frames I have loaded, the smoothing can be anywhere from 2-20 frames. What VMD is doing is literally averaging the positions of each atom over a range of frames, which averages out high frequency motions that obscure the slower motions we are interested in. This does have a side effect though, particularly for molecular rings, which will “wink” as they rotate due to the averaging.

FWIW, this "pumping/shrinking" effect can be significantly reduced by
using a higher order polynomial fitting to the sliding window instead
of using (0th order) averaging.

this movie uses this and a typical setting here is to average over
~100 frames with a 8th or 10th order polynomial.
https://www.youtube.com/watch?v=lMBRnxbcRH4

for systems with periodic boundaries and particles moving in and out,
there is another effect that needs to be taken care of. here is an
example, that has the polynomial filtering, but not the proper
treatment of PBC:
https://www.youtube.com/watch?v=6b2WK5rmzBc
you can see that it is still very busy due to the coarse grained water
particles (the spheres) jumping in and out of the frame.

this can be addressed by processing the trajectory with a segment wise
unwrapping, as it is done in this movie:
https://www.youtube.com/watch?v=lm-dAvbl330

all of these visualizations/movies were created with a custom VMD
version, where i had hacked the trajectory averaging code to use the
savitzky-golay filter mentioned above.

axel.

>
> Josh Vermaas
>
> Director’s Postdoctoral Fellow
> National Renewable Energy Laboratory
> joshua.vermaas_at_nrel.gov<mailto:joshua.vermaas_at_nrel.gov>
>
>
>
>
> On Nov 4, 2016, at 4:54 AM, Ajasja Ljubetič <ajasja.ljubetic_at_gmail.com<mailto:ajasja.ljubetic_at_gmail.com>> wrote:
>
> Hi!
>
> Probably the Trajectory Smoothing Window size (under Graphic representations->Trajectory) was set to something higher than 0.
>
> <image.png>
>
> Best,
> Ajasja
>
> On 4 November 2016 at 11:26, Kevin C Chan <cchan2242-c_at_my.cityu.edu.hk<mailto:cchan2242-c_at_my.cityu.edu.hk>> wrote:
> Dear Users,
>
> I am currently making some MD movies. I looked at some examples on the web (e.g. https://www.youtube.com/watch?v=GSi5-y6NHjY from TCBG UIUC). However, when I made mine, the movement of particles seem very jumpy due to the frame intervals (e.g. https://www.youtube.com/watch?v=uO9-k_HXy0c), especially for the lipid tails. I thought reducing frame intervals could improve but still not working for 20ps per frame, which already made my movie >50000 frames. It looks like the movements have been interpolated or smoothened in some ways.
>
> I know that some of the great movies on webs were made by users here, so I am asking if any post-modifications have been applied to the trajectory. Or you are just using very very small frame intervals.
>
> Thanks in advance,
> Kevin
> City University of Hong Kong
>
>
> 

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com  http://goo.gl/1wk0
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.