VMD-L Mailing List

From: Nuno Sousa Cerqueira (nscerque_at_fc.up.pt)
Date: Thu Dec 10 2009 - 01:05:51 CST

Begin forwarded message:

> From: Nuno Sousa Cerqueira <nscerque_at_fc.up.pt>
> Date: December 10, 2009 7:05:29 AM GMT+00:00
> To: Yu Fang <y.fang2_at_ugrad.unimelb.edu.au>
> Subject: Re: vmd-l: paratool - parameter generation from gaussian
> log files
>
> Hi Yu,
>
>
> I think the error in gaussian is a mistake in one of the keywords of
> the input file.
> Instead of the FREQ you should put OPT.
> Note that the modredundant option requires some input after the
> coordinates of the molecule. It is generally used to constrain some
> bonds or freeze some atoms.
>
> Regards, Nuno
>
>
>
>> %chk=dopamine_transf.chk
>> %nproc=1
>> %mem=1GB
>> # RHF/6-31G* Geom=(AllCheck,ModRedundant) OPT=(Modredundant,ReadFC)
>> IOp(7/33=2)
>> (numbers follow)
>
>
> On Dec 10, 2009, at 4:17 AM, Yu Fang wrote:
>
>> Hi All
>>
>> I have a question regarding the vmd paratool plugin
>> As I understand it this plugin is capable of reading the gaussian QM
>> simluation log files and computing the CHARMM force field
>> parameters for
>> bonds, angles and dihedrals based on this information.
>>
>> I understand how the CHARMM topology & parameter files work for
>> NAMD, but
>> I am new to gaussian. I am trying to parametize a few small ligand
>> molecules, for which comparison to atom types in the default
>> parameter
>> file that comes with namd is not good enough.
>>
>> What I have done is
>> -selected atom types and assigned VDW parameters (using default
>> topology/parameter file that comes with namd)
>> -autogenerated internal co-ordinates
>> -performed the optimized geometry and single point calcs in
>> gaussian as
>> required by paratool and imported in the respective log files
>> successfully.
>>
>> However, I cannot figure out what to do next to get paratools to
>> generate
>> the CHARMM parameters for me.
>>
>> I tried to perform a co-ordinate transformation QM job in gaussian,
>> thinking this might be required. As I said, I am new to gaussian,
>> so I do
>> not know what this job does. Anyhow the gaussian input file paratool
>> generated for me did not work, I suspect because I am using g98
>> instead of
>> g03.
>>
>> This gaussian input file options are shown below:
>>
>> %chk=dopamine_transf.chk
>> %nproc=1
>> %mem=1GB
>> # RHF/6-31G* Geom=(AllCheck,ModRedundant) Freq=(Modredundant,ReadFC)
>> IOp(7/33=2)
>> (numbers follow)
>>
>> running this file, g98 seems to complain that
>> "Freq=(Modredundant,ReadFC)"
>> contained a syntax error.
>>
>> The gaussian log file looks like the following:
>>
>> *********************************************
>> Gaussian 98: x86-Linux-G98RevA.7 11-Apr-1999
>> 10-Dec-2009
>> *********************************************
>> %chk=dopamine_transf.chk
>> %nproc=1
>> Will use up to 1 processors via shared memory.
>> %mem=1GB
>> Default route: MaxDisk=2000MW
>> ----------------------------------------------------------------------
>> # RHF/6-31G* Geom=(AllCheck,ModRedundant)
>> Freq=(Modredundant,ReadFC) I
>> Op(7/33=2)
>> ----------------------------------------------------------------------
>> QPERR --- A SYNTAX ERROR WAS DETECTED IN THE INPUT LINE.
>> MODREDUNDANT) FREQ=(MODREDUNDANT,READFC)
>>
>> So, in short, I would like to know what I need to do next to
>> generate the
>> CHARMM parameters using paratools. If I need to perform a co-ordinate
>> transformation gaussian job, what are the options I should use in the
>> input file for g98.
>>
>> Any help much appreciated.
>> Yu
>>
>>
>>
>>
>>
>>
>>
>>
>>
>