[gmx-users]query regarding unfolding of protein at higher temperature

[Tsjerk Wassenaar]

Hi Sangeeta,

>
> 1)      Will the result be erroneous if I keep box size unaltered at higher
> temperature?

If your protein starts unfolding, you are likely to get direct
interactions between your periodic images. That will severely distort
your results (to the point that you can not draw conclusions from
them).

>
> 2)      What should be minimum distance between protein atoms and box edges?
> I have found that some authors say minimum distance will be 0.85 nm, but I
> also have found that several authors carried on simulations with dist 0.5nm
> between protein atoms and box edges, so I am getting confused.
>

In principle, you should allow the solvent in between periodic images
to behave rather bulk-like. This means that you'd have to have about 2
nm between the images (i.e. 1 nm between the solute and the edge of
the periodic unit cell). Some argue that if you're using a cut-off,
you at least have to have a distance between periodic images larger
than the cut-off distance. But this would usually come down to 1.4 nm,
ergo 6 water layers, which would not be sufficient for bulk-like
behaviour.

> 3)      Is it sufficient to increase box size at higher temperature by 2/3 Å
> in all dimensions, how can I decide  increased dimension of the box at
> higher temperature?
>

This depends on the geometry of your system.

Cheers,

Tsjerk

-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623