[gmx-users] simulated annealing for Anton

[Weihua Li]

> Hi all,
>   I want to perform a MD with simulated annealing.But i don't know how to use the parameters in .mdp file(e.g. cooling down from 500k to 100k).It seems to only have a parameter that cooled down to 0K,Zero_temp_time. 
> How to control the system to cool down to 100K?
>  Could anyone give me some suggestions? Thank you very much!
>Simple: stop your simulation as soon as the temperature reaches
>100K, then continue with regular T-coupling w/o simulated anealing.
>i.e. you would do something like:
>grompp -f cool.mdp -o cool.tpr
>mdrun -f cool.tpr -o cool.trr
>grompp -f 100K.mdp -t cool.trr
>mdrun
>(I might have some of the option names wrong,
>check for a trajectory input for grompp.)
>In cool.mdp you should tune nsteps such that at the final
>step your simulated annealing will have dropped to 100K.

 Thank Anton for your answer. 
In my case (from 500k to 100k), the zero_temp_time parameter is necessary? If i don't use it , the grompp seems not to work.
So, what is the meaning of zero_temp_time parameter in  my case since the system does not cooled down to zero ? 
And is the ref_t set to 100K? 
Looking forward to your reply. Thank you very much.
Cheers,
Weihua Li



Cheers,
Weihua Li
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20030320/4919aa70/attachment.html>