[gmx-users] compare simulations between wt and mutate protein

[Nguyen Hoang Phuong]

Hi,

> Hi Andrea,
>
> I think it's a good thing to start another simulation by mutating the
> wt at 20ns. Besides, it would be good to mutate the mutant at 20ns back, to
> see if it goes the other way. In any case, you'll have to perform
> multiple simulations. For a strategy on comparing simulations, to see
> which properties are affected, you can have a look at chapters 2 and 4
> of my thesis, which I know you have. Comparing with respect to the
> "internal dynamics" by means of PCA is also a good possibility and
> would generally be approached by "Combined Essential Dynamics", but I
> would note there are better methods.
In addition to that I would like to mention that you should be careful 
when comparing simulations by using PCA. The conformational states 
obtained by PCA may be artificial ones. I have discussed this problem in 
the recent paper: Proteins 65, 898, 2006. I don't want to advertise the 
paper but just try to provide more information which may help you more.

Best wishes,

Phuong