This smells like a missing residue(s) problem--there is a loop or some
other part of the protein that is not observed in the X-ray structure due
to disorder. These residues have been omitted. Write the following input
for mdgx:
cat > mdgx.in << EOF
&files
-p <your prmtop>
-c <your inpcrd>
&end
&configs
count 1
verbose 1
maxcyc 2
outbase 'Conf',
write 'pdb',
&end
EOF
Then run mdgx -O -i mdgx.in... this will apply the 'configuration sampling'
protocol which is really for making large numbers of conformational
variants of small molecules (I know oyu know this, I'm saying it for the
listserv). Here, we're taking advantage of the checks it puts on each
resulting configuration: it tells us where the strain was. The maxcyc 2
won't converge, and mdgx will print no results (not even the 1 we asked
for) because the sanity checks fail. (This is assuming that the problem is
a missing few residues in the protein). Sure enough, here's the result at
the bottom of the mdgx mdout file:
- Configuration 1:
Strain is 27322.3495 for bond C -N (4048, 4050)
Strain is 35.0364 for angle CA -C -N (4045, 4048, 4050)
Restraint binding atoms () strained to 0.0000
So look at atoms 4048 and 4050--a peptide bond is stretch between some
residues that look skipped. The atom numbering got a little jumbled when
tleap created the topology, but I reproduced your error with the "bond of
8.882 between..." and we can run ambpdb -p [your topology] < [your
coordinates] > [new pdb file] to see what the topology really looks like.
Here are the aforementioned atoms in that structure:
ATOM 4046 HA2 GLY 259 29.274 48.036 -30.126 1.00
0.00 H
ATOM 4047 HA3 GLY 259 30.160 47.170 -29.179 1.00
0.00 H
ATOM 4048 C GLY 259 30.907 49.072 -29.422 1.00
0.00 C
ATOM 4049 O GLY 259 30.492 49.715 -28.456 1.00
0.00 O
ATOM 4050 N ASN 260 31.807 49.055 -38.259 1.00
0.00 N
ATOM 4051 H ASN 260 32.316 49.075 -39.131 1.00
0.00 H
ATOM 4052 CA ASN 260 30.347 49.097 -38.308 1.00
0.00 C
So, missing residues between GLY 259 and ASN 260.
Dave
On Tue, Jul 9, 2019 at 12:16 PM Carlos Simmerling <
carlos.simmerling.gmail.com> wrote:
> sorry I meant to visualize your prmtop and crd in something like VMD.
>
> On Tue, Jul 9, 2019 at 11:49 AM Anthony Bogetti <anthony.bogetti.gmail.com
> >
> wrote:
>
> > The pdb file was originally a dimer, but we removed one of the chains.
> We
> > tried adding in TER line at the end and it didn't fix the issue.
> >
> > How can I load prmtop and inpcrd files directly into leap?
> >
> > Thanks,
> > Anthony
> >
> > On Tue, Jul 9, 2019 at 11:44 AM Carlos Simmerling <
> > carlos.simmerling.gmail.com> wrote:
> >
> > > you might try to load in the prmtop and inpcrd generated by leap, since
> > > that will show the bonds leap determined.
> > > could you have a dimer without a TER line?
> > >
> > > On Tue, Jul 9, 2019 at 11:41 AM Anthony Bogetti <
> > anthony.bogetti.gmail.com
> > > >
> > > wrote:
> > >
> > > > Hello,
> > > >
> > > > I'm trying to create a topology and coordinate file for a protein in
> > > leap.
> > > > I obtained the pdb file and added hydrogens and changed nothing else.
> > > When
> > > > I load the pdb file in tleap, everything seems fine but I'm getting
> an
> > > > error message that says:
> > > >
> > > > There is a bond of 8.882728 angstroms between: [Blank]
> > > >
> > > > It doesn't tell me what atoms this "bond" is between. What could be
> > > > causing this?
> > > > I've visualized my structure and everything looks fine.
> > > > I'm using ff14SB and I've checked all of my atom names/types and they
> > are
> > > > correct.
> > > >
> > > > Thanks,
> > > >
> > > > Anthony Bogetti
> > > > Graduate Student at University of Pittsburgh
> > > > Advisor: Dr. Lillian Chong
> > > > _______________________________________________
> > > > 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
> > >
> > _______________________________________________
> > 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
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 09 2019 - 10:00:02 PDT
