Re: some recommendations for parametrizing new molecules required

From: Christopher Gillespie (gillescche_at_gmail.com)
Date: Fri Jan 23 2009 - 07:36:30 CST

Hi Vlad,

I have recently started a parameterization of my own and here is what
I have seen thus far.

  GAMESS supposedly has decent parallelization and potentially can be
faster than Gaussian. I say supposedly because I have only tried it
once and have used Gaussian for everything else. Gaussian has
parallelization as well but I have been informed GAMESS does a better
job of it. Also, I have used the US version and have not tried another.

As for time, I have a 43 atom optimization (MP2/cc-PVDZ) with
counterpoise correction running for the last 13 days with Gaussian,
which is higher than the recommended HF/6-31g* level for CHARMM FFs
but still your system is quite a bit larger, QM speaking.

Hopefully someone with more GAMESS experience can supply more
information on its speed and use with Paratool.

Best
Chris

On Jan 23, 2009, at 4:31 AM, CHINDEA Vlad wrote:

> Dear NAMD users
>
> At this moment I am pondering on the strategy I will follow in order
> to parametrize for NAMD a drug molecule with 124 atoms. It is quite
> a complex tricyclic molecule, not baring much resemblance with
> molecules already parametrized. I have read the Paratool guideline
> and the additional information from Mackerell and now I would need
> some clarifications before starting:
>
> I seems quite possible that I will not get Gaussian through our
> university which means that I will have to use other programs. In
> Paratool tutorial it is stated that GAMESS could be used but there
> is not much information given on how to proceed with it. Before
> embarking on using this program I would need reassurance that it is
> indeed possible to use GAMESS for this purpose (is there maybe some
> progress on developing Paratool compatibility with GAMESS ?) and
> maybe some pointers on the differences between using Gaussian and
> GAMESS (I am guessing at least the manual input of the QM results
> from GAMESS in Paratool).
>
> Which of the 3 GAMESS (US, UK, PC/Firefly) to use ? Important
> criteria here would be speed, availability of support and previous
> work done with any of these programs for preparing NAMD input. I
> have checked a little bit all programs and I only found a limited
> number of benchmarks which are favorable to PC/Firefly. So what is
> your opinion here ?
>
> I would appreciate a rough idea on the computation resources
> required. As my molecules splits naturally by cycles and the largest
> of them has about 80 atoms it seems to me that this will be the
> largest fragment on which I will have to do single-point QM. Is this
> reasonable to calculate on a single Intel P4 2,66 Ghz processor at
> the required level of theory ? As I understand the threshold for
> such calculations is usually 120 atoms, but with what computational
> power ? I could live with calculations taking days, but not weeks !
>
> Is there a difference in using CHARMM or AMBER charges ? I
> understand that they are calculated differently but once calculate
> does NAMD cares about this distinction (as it does for example for
> parameters file format: either CHARMM or X-PLOR) ? Or is there any
> limitation in using one or the other ?
>
> Sorry for the long mail and thank you for your support
>
> All the best !
>
> Vlad Chindea
>
> What can you do with the new Windows Live? Find out

----------------------------------------------------------------------------------------
| Christopher Gillespie
| Ph.D. Candidate
| University of Delaware
| Department of Chemical Engineering
| 150 Academy Street
| Newark, DE 19716
| gillesc (at) udel.edu
| http://udel.edu/~gillesc
----------------------------------------------------------------------------------------

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:17 CST