[gmx-users] cannot print velocity information to trr/xtc files (Carbon Naotube inside water)

[Mark Abraham]

Nadir Kaplan wrote:
>>> What is going on here? How can I get velocity info? Any help will be
>>> greatly appreciated. Note that, there is an external acceleration on
>>> the water molecules, where is the position of CNT is fixed. So
>>> obviously water molecules should have finite velocities.
>>
>> Turn off the weird stuff (freeze groups and accelerate groups) until 
>> you've got your basic issues sorted out. MD simulations in general and 
>> CNT simulations in particular are not easy things to do right. Learn 
>> to walk before you try to run! :-)
> 
> 
> 
> Freezing is for simplicity. One needs velocity to run.

Yeah, but freezing and accelerating do things to velocities. If you're 
having problems, a good strategy is to simplify the situation until you 
can identify where the problem arises... thus, no freezing and accelerating.

> Below I paste another mdp file which I used for the MD part of the same
> simulation. That MD simulation was succesfully over and I was able to
> reproduce some results in the literature by using position info. Again,
> in the .trr file of that simulation, there are no velocities.

Well, those reproduced results probably mean that there *were* 
velocities :-) Velocities must be there in the .trr for the last 
timestep, so long as mdrun was allowed to terminate normally.

If you really had "nstvout = 5000" and you are looking at the correct 
non-post-processed trajectory with gmxdump on a suitable timestep, then 
your observations are inexplicable, unless there's some weird "gotcha" 
with no velocities being written for atoms in either a freeze or 
accelerate group... not that that makes much sense... In short, it seems 
much more likely you've made an error than that there's a weird bug.

My earlier advice is still good - get a clean working directory and do 
the same kind of simulation but without the freeze and accelerate groups 
and see if you can reproduce your problem while recording everything you 
did on the way. If it now works, do the same but add in the weird stuff 
until you isolate the problem - if any.

Good luck!

Mark