Re: wrap is On to Diffusion phenomena ?

From: Jérôme Hénin (
Date: Tue Jul 27 2021 - 07:13:34 CDT

Dear Geo,

----- On 16 Juil 21, at 0:45, Ropón-Palacios <> wrote:

> Dear NAMD Users,

> I really have a problem, which does not let me sleep. During the diffusion
> process (where I want to observe the input of an ion to a channel), I have been
> using the wrapAll and wrapWater on options in the config file, at the end of
> the simulation I have been able to see that there is really the input of ions,
> but Many authors say that you have to "unwrap" the trajectory, when doing
> diffusion analysis,

You need to unwrap if you are interested in the diffusive dynamics of the ion, because you don't want big artificial jumps.

> then, I have done it, using pbc both only for the ions and for the whole system
> and in both cases it is not seen that the ion penetrates,

The ion is still inside the channel even it is far away. You can imagine that it is inside a periodic image of the channel (you can display them in VMD). Here it is simpler to look at wrapped coordinates, with the channel centered.

> so I thought about extending the dynamics, but what happens is that when I
> extend the dynamics, the structure with which I start is the one that is inside
> the channel, so all this causes me enormous confusion.

> Please, could you tell me in which cases wrapAll or wrapWater can be used in a
> dynamic.

For the simulation it does not matter. To analyze binding, look at wrapped coordinates. For diffusion, look at unwrapped coordinates.


This archive was generated by hypermail 2.1.6 : Fri Dec 31 2021 - 23:17:11 CST