[gmx-users] supervised MD in Gromacs?

[Albert]

Dear Joao:

thanks again for such kind comments.

I know Plumed can perform similar work probably if we define the 
distance between the ligand and binding pocket as the CV. However, I am 
not so sure whether such kind of metadynamic simulation will lead to a 
nice ligand binding mode in the end, since we can place the ligand with 
countless pose..... In the mentioned "supervised MD", it seems that at 
least they can reproduce the ligand binding mode observed crystal 
structure in such a short time......

Do you know any other similar methods can do this?

Thanks a lot

Albert

On 12/08/2015 02:46 PM, João M. Damas wrote:
> Dear Albert,
>
> Unfortunately, the methods section of some papers does not reflect every
> detail of what was done to achieve the results. In the particular case of
> computational papers, this means the use of scripting (or even hacking) to
> assist a more wide-spread code (like GROMACS). That being said, I do not
> know what was done in those papers.
>
> I have not read the papers, but these "supervised" techniques seem fairly
> similar to the milestoning ones. Maybe find an alternative there?
>
> Best,
> João