AMBER Archive (2008)

Subject: RE: AMBER: Clarification re qmmask

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Thu Oct 16 2008 - 13:14:04 CDT

Hi Ben,

> Is this deliberate, and if so, is the rationale behind it known? It's
> certainly not something I could have guessed from the manual, which says,
> "see mask documentation for more info" (p. 93), implying, "just observe
> one
> of the mask formats and all will be well".
>
> I could understand this if periodic attempts to rebuild the QM region from
> the mask take place, thus tampering with energy conservation. On the other
> hand, I would have expected (not having seen for myself, admittedly) that
> the QM region gets built only once, at the start of the simulation. Is
> this
> rebuild going on, or is there some other cause?

This is certainly not deliberate - the QM code simply calls the same mask
routine as anything else in amber and this returns the atom numbers that are
then used to build the QM region. Thus in theory there should be no reason
why it shouldn't work although I can think of lots of reasons why you
wouldn't want to do it this way.

With regards to the sander_bomb that error is occurring because there are
various checks to make sure the QM region is legal and a key one is whether
it fits inside the periodic box within the minimum image convention. Hence
seeing these errors if you have a decent sized water box suggests that it is
not picking the nearest atoms correctly - note I have absolutely no idea how
this deals with imaging so if you have an unimaged inpcrd file this may
cause problems. Why you would see this problem with a value of 0.01 though
I don't know. Can you send me your input and output so I can take a look at
it.

To debug further I would also start by building a non-periodic system - say
in a solvent cap and playing with that first since this won't trigger the
minimum image sander bomb even if the QM region is too large - although you
may get other errors about spin / charge etc. The key is to set write_pdb=1
and take a look at it to see what your QM region actually looks like once it
selects it.

AS for why you might not actually want to use the nearest option for mask
(in fact I have never tried it myself and didn't even know it existed) is
that this will indiscriminately pick residues near the QM region, this might
make sense if they are only waters but if they are not you are always going
to be cutting at peptide bonds which is not really the best place to be
cutting the QM/MM region. Also note that the selection won't be dynamically
updated during the run (this is actually quite hard to do - especially if
you want to do it 'properly' and not actually change your QM/MM Hamiltonian
during the simulation) however Gustavo and I are currently working on a
method by which one can keep a specific number of solvent molecules close to
the QM region and have it change dynamically during the simulation as
solvent molecules move.

All the best
Ross

/\
\/
|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu