if ParmEd reports that a bond exists, then the prmtop defines
that bond... VMD *should* report all bonds when loading a prmtop
In my hands, ParmEd reports that bonds commanded on leap between the active
center (provided through a mol2 file and related frcmod) and the protein
backbone (CA atom of glycine-like residue) do exist. However, VMD does not
show such bonds, while it shows correctly those bonds indicated in the mol2
file.
Not because I do not rely on ParmED but due to the unusual case, I checked
manually that such bonds are written on the parm7 file. Let me take an
example (I am using the PDB numbering of the PDB file saved by leap)
-C2 (ID 7893) is the carbon atom name of a CH2 of the active center that
was indicated to leap to be bound to the CA atom (ID 1628) of the protein
backbone. Both carbons are free-valence, i.e., they are ready for bonding
without having to delete hydrogens.
7893 (PDB) -> 23676 (parm7)
1628 (PDB) -> 4881 (parm7)
The bond between 4881 and 23676 is written on the parm7 file. Likewise for
the other bonds between the active center and the protein backbone.
So, your 'should' above seems to be appropriate. I am curious why those
bonds are not shown by VMD (and I'll have problems when I need a rendering
of the active area of the enzyme if the work goes on well). For this reason
I described above the nature of what is bonded.
francesco
On Thu, Jan 15, 2015 at 2:10 PM, Jason Swails <jason.swails.gmail.com>
wrote:
> On Thu, 2015-01-15 at 09:18 +0100, Francesco Pietra wrote:
> > I am trying to verify chemical bonds commanded from leap with
> >
> > ---VMD
> > ---parmed.py
> > ---parm7 file
> >
> > I assume that in parm7 file:
> > BONDS INC HYDROGEN
> > This section contains a list of every bond in the system in which at
> least
> > one atom is Hydrogen. Each bond is identified by 3 integers—the two
> > atoms involved in the bond and the index into the BOND FORCE CONSTANT
> > and BOND EQUIL VALUE. For run-time efficiency, the atom indexes are actu-
> > ally indexes into a coordinate array, so the actual atom index A is
> > calculated
> > from the coordinate array index N by A = N/3 + 1. (N is the value in the
> > topology file)
> > %FORMAT(10I8)
> > There are 3 × NBONH integers in this section.
> > BONDS WITHOUT HYDROGEN
> > This section contains a list of every bond in the system in which neither
> > atom
> > is Hydrogen. It has the same structure as BONDS INC HYDROGEN described
> > above.
> > %FORMAT(10I8)
> > There are 3 × NBONA integers in this section.
> >
> >
> > Following is the initial portion of my parm7 file (ambertools14):
> > FLAG
> > BONDS_WITHOUT_HYDROGEN
> > %FORMAT(10I8)
> >
> > 30 33 1 30 36 2
>
> Let's start with these. 30 33 1 means that atoms 30/3+1 and 33/3+1 are
> defined by the first bond. These are atoms 11 and 12 -- the carbonyl
> bond. These are parametrized by the first entry in BOND_EQUIL_VALUE and
> BOND_FORCE_CONSTANT.
>
> The next bond is 30 36 2, meaning that atoms 30/3+1 and 36/3+1 are
> bonded to each other. These are atoms 11 and 13 -- the amide bond
> between the first two residues. They are parametrized by the second
> entry in BOND_EQUIL_VALUE and BOND_FORCE_CONSTANT.
>
> > Could you please instruct me how to verify bonds in the above list for
> the
> > two initial residues in the corresponding PDB file?
>
> Note that tleap typically re-orders atoms in the original PDB file so
> that you can't use the original PDB file to compare atom numbers like
> this. If you save a PDB from tleap or use ambpdb (or cpptraj) to create
> a PDB file (which it appears you have done), then that PDB file works
> for this.
>
> > ATOM 1 N ALA 1 39.886 -67.858 8.967 1.00 0.00
> > ATOM 2 H1 ALA 1 40.139 -66.882 8.910 1.00 0.00
> > ATOM 3 H2 ALA 1 39.435 -68.128 8.104 1.00 0.00
> > ATOM 4 H3 ALA 1 40.681 -68.381 9.306 1.00 0.00
> > ATOM 5 CA ALA 1 38.841 -68.004 10.020 1.00 0.00
> > ATOM 6 HA ALA 1 37.934 -68.417 9.578 1.00 0.00
> > ATOM 7 CB ALA 1 39.276 -69.022 11.092 1.00 0.00
> > ATOM 8 HB1 ALA 1 39.319 -68.531 12.064 1.00 0.00
> > ATOM 9 HB2 ALA 1 38.557 -69.840 11.130 1.00 0.00
> > ATOM 10 HB3 ALA 1 40.261 -69.416 10.841 1.00 0.00
> > ATOM 11 C ALA 1 38.486 -66.675 10.668 1.00 0.00
> > ATOM 12 O ALA 1 39.265 -65.721 10.620 1.00 0.00
> > ATOM 13 N MET 2 37.288 -66.634 11.251 1.00 0.00
> > ATOM 14 H MET 2 36.670 -67.430 11.184 1.00 0.00
> > ATOM 15 CA MET 2 36.788 -65.488 12.003 1.00 0.00
> > ATOM 16 HA MET 2 37.498 -64.664 11.926 1.00 0.00
> > ATOM 17 CB MET 2 35.425 -65.022 11.462 1.00 0.00
> > ATOM 18 HB2 MET 2 34.698 -65.828 11.565 1.00 0.00
> > ATOM 19 HB3 MET 2 35.087 -64.154 12.027 1.00 0.00
> > ATOM 20 CG MET 2 35.418 -64.613 9.987 1.00 0.00
> > ATOM 21 HG2 MET 2 35.721 -65.461 9.373 1.00 0.00
> > ATOM 22 HG3 MET 2 34.414 -64.298 9.701 1.00 0.00
> > ATOM 23 SD MET 2 36.551 -63.245 9.652 1.00 0.00
> > ATOM 24 CE MET 2 35.781 -61.890 10.516 1.00 0.00
> > ATOM 25 HE1 MET 2 36.464 -61.506 11.274 1.00 0.00
> > ATOM 26 HE2 MET 2 35.541 -61.096 9.808 1.00 0.00
> > ATOM 27 HE3 MET 2 34.866 -62.237 10.995 1.00 0.00
> > ATOM 28 C MET 2 36.619 -65.876 13.464 1.00 0.00
> > ATOM 29 O MET 2 36.237 -67.031 13.793 1.00 0.00
> >
> >
> > Embarrassed enough to pose such a naive question, however I am in trouble
> > in verifying the active center with transition metals (where not all
> bonds
> > indicated by both leap log and parmed.py are displayed on VMD; I know
> that
> > in such a case VMD only relies on bonds registered in the parm7 file,
> > abandoning any guess on interatomic distances)
>
> Going through each bond entry like this will be much too tedious. That
> was part of the motivation behind writing ParmEd (and, originally,
> rdparm). ParmEd does exactly what I described above when listing the
> bonds. So if ParmEd reports that a bond exists, then the prmtop defines
> that bond.
>
> VMD *should* report all bonds when loading a prmtop (but if you load a
> PDB file -- even generated from a prmtop file -- then VMD will do a
> simple bond-by-distance criteria when displaying them). There are some
> representations (e.g., "dynamic bonds") that ignores this "bond"
> information and simply uses a distance-based cutoff. But most
> representations should respect the bond info in the topology file.
>
> HTH,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
>
> _______________________________________________
> 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
Received on Fri Jan 16 2015 - 09:30:03 PST
