We prepared the atom type table file
($AMBERHOME/dat/antechamber/ATOMTYPE_SYBYL.DEF) based on the sybyl
document. As we know, no aromatic atom types (C.ar and N.ar) are assigned
to five-memberred rings. You can verify this by retrieving sample molecules
from the sybyl's fragment library, or you may simply scratch your molecules
and check what kinds of atom types are assigned automatically.
Babel may assign atom types based on their rules. In antechamber, however,
we try to reproduce the atom types generated by the sybyl software
packages. You could define your atom types according the rules described in
the atom type table files and the Antechamber paper.
Indeed, there is a problem with bond types assignment for your molecule:
some bonds of your molecule are assigned as aromatic bonds. The problem has
been fixed and the new mol2.c file has been checked into amber branch.
Please also find the mol2.c attached in this email.
Best
Junmei
On Thu, Jul 18, 2013 at 12:09 PM, <hannes.loeffler.stfc.ac.uk> wrote:
> Hi,
>
> thanks for your reply. I want to have Sybyl atom types in the output from
> a gaff mol2 file. I got those but the wrong ones. I don't think that
> cc/cd, ss, na should be translated to C.2, S.3, N.3. I would have expected
> to see C.ar, S.2 and N.ar as they were original. You see, a ful cycle of
> backward and forward conversions should not result in different types. But
> it does.
>
> Cheers,
> Hannes.
>
> ________________________________________
> From: Junmei Wang [junmwang.gmail.com]
> Sent: 18 July 2013 17:55
> To: AMBER Mailing List
> Subject: Re: [AMBER] antechamber issues
>
> Hello,
>
> For the first issue, I guess you misunderstand the '-at ' flag as the
> atom types of the input file. As a matter of fact, you use '-at' to
> specify the atom type of the output file. The default atom type is
> 'gaff'. So you could get gaff atom types simply running command:
>
> antechamber -i ring.mol2 -fi mol2 -o ring-sybyl.mol2 -fo mol2
>
> For the second issue, please add more stuff immediately after the
> following three lines in antechamber.c
> if (strcmp("gesp", cinfo.outtype) == 0
> || strcmp("25", cinfo.outtype) == 0)
> wgesp();
>
>
> stuff you need to add:
> if (strcmp("sqmcrt", cinfo.outtype) == 0
> || strcmp("23", cinfo.outtype) == 0)
> wsqmcrt(ofilename, atomnum, atom, minfo);
> if (strcmp("sqmout", cinfo.outtype) == 0
> || strcmp("24", cinfo.outtype) == 0)
> wsqmout(ofilename, atomnum, atom, cinfo, minfo);
>
> We will also file a bug report soon
>
> Best
>
> Junmei
>
>
>
> On Wed, Jul 17, 2013 at 4:02 AM, Hannes Loeffler
> <Hannes.Loeffler.stfc.ac.uk> wrote:
> >
> > Hi,
> >
> > I have found two issues with antechamber:
> >
> > 1) When I convert the attached pdb file to mol2 with e.g.
> >
> > antechamber -i ring.pdb -fi pdb -o ring.mol2 -fo mol2 -j 4 -s 2
> >
> > I get a mol2 file with reasonable GAFF atom types.
> >
> > But when I try to get a "standard" mol2 with Sybyl atom types via
> >
> > antechamber -i ring.mol2 -fi mol2 -o ring-sybyl.mol2 -fo mol2 -at sybyl
> > -s 2 -j 4 (or -j 1)
> >
> > I get wrong atom types (see attachment). The same happens when
> > directly converting from the PDB (identical mol2).
> >
> > Converting the original pdb to mol2 with openbabel yield Sybyl atom
> > types that I would think are a reasonable match of the GAFF types (see
> > attachment).
> >
> >
> > 2) I would expect a command like
> >
> > antechamber -i ring.pdb -fi pdb -o sqm.in -fo sqmcrt -j 0
> >
> > to write the SQM input file but it doesn't (not output generated at
> > all).
> >
> >
> > Thanks,
> > Hannes.
> > --
> > Scanned by iCritical.
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> --
> Scanned by iCritical.
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
- text/x-csrc attachment: mol2.c
Received on Tue Jul 23 2013 - 12:00:03 PDT
