AMBER Archive (2005)

Subject: Re: AMBER: antechamber -- missing improper torsions

From: David Mobley (dmobley_at_gmail.com)
Date: Fri Oct 21 2005 - 12:21:01 CDT

Ross,

I'm attaching relevant files (actually, only one is necessary:
fmn.pdb). Here are the steps I follow:

antechamber -i fmn.pdb -fi pdb -o fmn.mol2 -fo mol2 -c bcc -s 02
parmchk -i fmn.mol2 -f mol2 -o frcmod

Then I open tleap and do the following:
source leaprc.gaff
mods = loadAmberParams frcmod
ligand=loadMol2 fmn.mol2
saveAmberParm ligand ligand_nowater.prmtop ligand_nowater.crd
quit

Then I "cat" leap.log and see the following errors (warnings, really):
 ** Warning: No sp2 improper torsion term for c-ne-cd-cd
        atoms are: C4 N5 C4A C10
 ** Warning: No sp2 improper torsion term for ne-ca-ca-ca
        atoms are: N5 C6 C5A C9A
 ** Warning: No sp2 improper torsion term for ca-ca-ca-ha
        atoms are: C5A C7 C6 H6
 ** Warning: No sp2 improper torsion term for ca-ca-ca-ha
        atoms are: C8 C9A C9 H9
 ** Warning: No sp2 improper torsion term for ca-ca-ca-nh
        atoms are: C5A C9 C9A N10
 ** Warning: No sp2 improper torsion term for ca-cd-nh-c3
        atoms are: C9A C10 N10 C1*
 ** Warning: No sp2 improper torsion term for nc-cd-cd-nh
        atoms are: N1 C4A C10 N10

Note that these torsions do seem to be missing from frcmod and
gaff.dat, but for some reason adding them to the frcmod file doesn't
help. And from looking at it, it looks like these torsions SHOULD be
present, so I can see why leap is complaining.

Note also that this procedure normally works fine for most of the
small molecules I've tried, but for some reason I'm getting these
warnings in this case. Further, if I ignore the warnings and go ahead
and put the molecule in water and minimize, minimization usually exits
with a segmentation fault after several hundred steps.

Thanks,
David

On 10/18/05, Ross Walker <ross_at_rosswalker.co.uk> wrote:
>
> Hi David,
>
> Oh this is more complicated then. Is it possible for you to send me the
> files you are using and the steps you are doing so I can reproduce it and
> take a look.
>
> Thanks
> Ross
>
> /\
> \/
> |\oss Walker
>
> | Department of Molecular Biology TPC15 |
> | The Scripps Research Institute |
> | Tel: +1 858 784 8889 | 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.
>
>
>
> ________________________________
> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of
> David Mobley
> Sent: Tuesday, October 18, 2005 12:22
> To: amber_at_scripps.edu
> Subject: Re: AMBER: antechamber -- missing improper torsions
>
>
> Ross,
>
> No, it isn't putting them in the frcmod file. The errors I show are in leap
> when it tries to generate the prmtop file; it can't find some of the
> torsions it's looking for. And I had just applied all of the bugfixes two
> days before sending my original e-mail, so if it's a bug that has been
> fixed, the patch should be included in my compilation of amber.
>
> Thanks,
> David
>
>
>
> On 10/17/05, Ross Walker <ross_at_rosswalker.co.uk> wrote:
> >
> > This may be a leap bug rather than an antechamber bug. Is antechamber
> putting the impropers in the frcmod file but leap just not putting them in
> the prmtop file?
> >
> > If so then try applying bugfix.53 from
> http://amber.scripps.edu/bugfixes80.html
> >
> > All the best
> > Ross
> >
> >
> > /\
> > \/
> > |\oss Walker
> >
> > | Department of Molecular Biology TPC15 |
> > | The Scripps Research Institute |
> > | Tel: +1 858 784 8889 | 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.
> >
> >
> >
> > ________________________________
> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of
> David Mobley
> > Sent: Monday, October 17, 2005 11:34
> > To: amber_at_scripps.edu
> >
> > Subject: AMBER: antechamber -- missing improper torsions
> >
> >
> >
> > Dear all,
> >
> > I wrote some time ago to the list because antechamber is failing to put
> some of the necessary improper torsions for a small molecule I'm trying to
> set up in the frcmod file it generates. Can anyone shed some light on why
> antechamber is missing these, and what the fix is?
> >
> > See my original e-mail below, and attached are some files you should be
> able to use to reproduce the problem.
> >
> > Thanks,
> > David
> >
> >
> > ---------- Forwarded message ----------
> > From: David Mobley <dmobley_at_gmail.com>
> > Date: Sep 28, 2005 3:35 PM
> > Subject: AMBER: antechamber -- missing improper torsions
> > To: amber_at_scripps.edu
> >
> > Dear all,
> >
> > I'm generating AMBER parameters for a small molecule, flavin
> mononucleotide (FMN) with AMBER using antechamber on the pdb file (with all
> hydrogens), and antechamber is failing to put some of the required improper
> torsions for this molecule in the frcmod file that parmchk generates. Any
> idea why? Here are the particulars:
> >
> >
> > I use "antechamber -i (pdb file) -fi pdb -o (mol2 file) -fo mol2 -c bcc",
> then run "parmchk -i (mol2 file) -f mol2 -o frcmod". I'm attaching a copy of
> the pdb file, which comes from here:
> http://www.ebi.ac.uk/msd-srv/msdchem/cgi-bin/cgi.pl?FUNCTION=record&ENTITY=CHEM_COMP&PRIMARYKEY=1569&PARENTINDEX=-1&APPLICATION=1
> >
> > I am using the latest version of AMBER and just applied all of the bug
> fixes two days ago. The errors I get only show up in my leap.log (leap
> writes out the prmtop and crd files without complaining) and are:
> >
> > ** Warning: No sp2 improper torsion term for c-ne-cd-cd
> > atoms are: C4 N5 C4A C10
> > ** Warning: No sp2 improper torsion term for ne-ca-ca-ca
> > atoms are: N5 C6 C5A C9A
> > ** Warning: No sp2 improper torsion term for ca-ca-ca-ha
> > atoms are: C5A C7 C6 H6
> > ** Warning: No sp2 improper torsion term for ca-ca-ca-ha
> > atoms are: C8 C9A C9 H9
> > ** Warning: No sp2 improper torsion term for ca-ca-ca-nh
> > atoms are: C5A C9 C9A N10
> > ** Warning: No sp2 improper torsion term for ca-cd-nh-c3
> > atoms are: C9A C10 N10 C1*
> > ** Warning: No sp2 improper torsion term for nc-cd-cd-nh
> > atoms are: N1 C4A C10 N10
> >
> > If I proceed while ignoring this, the molecule in question will begin
> minimizing in water OK, then exit with an "address error" after several
> hundred steps.
> >
> > If I look through gaff.dat, I can't find improper torsions for the bonds
> it's complaining about, so I add the X -X -X -X entry from gaff.dat to my
> frcmod file under improper torsions. Strangely, however, it still seems not
> to recognize the improper torsions, as it still gives the same warnings in
> the leap.log file.
> >
> > Note: It isn't just that I don't know how to use antechamber or something.
> It's worked fine for a number of other small molecules, but seems to be
> having problems only with this particular one, for some reason.
> >
> > Any ideas?
> >
> > Thanks!
> > David
> >
> >
> >
> >
> >
>
>


  • chemical/x-pdb attachment: fmn.pdb
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu