[gmx-users] Re: QMMM cpmd

[stefhoor]

Hi Christian, that's a pity, isn't it? I don't think I will use this
ancient version of gromacs. It is actually quite hard to use QM/MM
with gromacs, I've tried to compile with mopac, and failed. I still
don't know what the problem is. Gerrit answered my email saying u
should link everything with g2c and I still don't have a clue what he
meant with that. Gamess-UK is paid, and I am not willing to pay for
something if I don't even know if I will be able to compile with
gromacs. Same thing for Gaussian. ORCA has no support for lam-mpi, and
I don't know if I can hold two versions of mpi on my computers.
I would really appreciate some suggestion know. I need to model an
enzymatic reaction and was planning to do it using QM/MM. Actually, I
am still planning, but I need just a bit of, let's say, guidance,
right now.
Thank you in advance


>
> Hi Stefan,
>
> unfortunately, cpmd-qmmm does not seem to be part of the standard
> gromacs version. If you want to use it, you have to use some ancient
> versions of gromacs and the patch from this page:
> http://www.tougaloo.edu/research/qmmm/index.htm
>
> If you have a big QM box, the superiority of cpmd in scalability is
> worth the afford of using an old GMX version. AFAIK there is no other
> qmmm possibility if you want to use the planar wave technique (and
> therefor the fantastic scalability). [Please correct me, if I am
> wrong...]
>
> Greets,
> Christian
>
> On Mon, 2010-05-31 at 20:49 -0300, Stefan Hoorman wrote:
>> I've downloaded gromacs from the git repository. I type ./bootstrap
>> and get loads of warnings like this
>>
>> configure.ac:414: warning: AC_CACHE_VAL(ac_m_cpu_7450, ...):
>> suspicious cache-id, must contain _cv_ to be cached
>> configure.ac:414: warning: AC_CACHE_VAL(ac_m_tune_970, ...):
>> suspicious cache-id, must contain _cv_ to be cached
>>
>> Plus, when I type ./configure CFLAGS=-DGMX_QMMM_CPMD --with-qmmm-cpmd,
>> configure runs fine except for this warning:
>> configure: WARNING: unrecognized options: --with-qmmm-cpmd
>>
>> I figure that if ./configure does not recognize the cpmd option, there
>> is no point in trying to compile gromacs with cpmd, I mean, the "make"
>> command is not going to include cpmd, or is it?
>> Some help on the matter would be appreciated.
>> Thank you
> --
> M.Sc. Christian Seifert
> Department of Biophysics
> University of Bochum
> ND 04/67
> 44780 Bochum
> Germany
> Tel: +49 (0)234 32 28363
> Fax: +49 (0)234 32 14626
> E-Mail: cseifert at bph.rub.de
> Web: http://www.bph.rub.de
>
>