Hi Paolo,
Thanks so much for your quick response.
Hmm… would you mind sharing how you are calculating those RMSDs? I’m seeing
RMSDs with constraints of over 1.0 Angstrom, so clearly some detail is
still escaping me.
My goal here is to reproduce the strain energy calculations that are part
of Chemalot’s sdfMMConfAnalysis (
https://github.com/chemalot/chemalot/blob/master/bin/sdfMMConfAnalysis.pl),
which currently rely on software licenses to generate the intermediate
local minima for the final strain energy curve. The idea is to have a few
local minima that haven’t moved very much from the input coordinates, but
which help fill in that bottom left “constrained minima” in the
conformational profile (Figure 2 from paper:
https://jcheminf.springeropen.com/articles/10.1186/s13321-017-0228-9)
Thanks again,
Katrina
On September 27, 2017 at 2:09:10 PM, Paolo Tosco (paolo.to...@unito.it)
wrote:
Dear Katrina,
I have just tried running your script computing the RMSD after minimization
with and without the per-atom harmonic constraints. Without constraints,
the heavy atom RMSD ranges from 0.07 to 0.68 angstrom for the various
conformers, whereas with the constraints ranges from 0.046 to 0.087
angstrom, which seems OK to me.
This said, doing a minimization with such a strong restraint placed on
every atom won't basically do anything: you are asking that each atom does
not move more than 0.1 angstrom from its initial position, which means that
basically the coordinates will remain unchanged after the minimization.
Cheers,
p.
On 09/27/2017 08:43 PM, Katrina Lexa wrote:
Hi Paolo,
Thanks so much for this suggestion (ages ago now). Based on Francois’
suggestion and your suggestion I have cobbled together a script that seems
like it ought to work to apply position constraints on each atom, minimize,
and then return the local minima conformations. Perhaps I’m still missing
something, as in my simple test case, it doesn’t seem like the position
constraints are doing anything - I’m asking for a constraint of 0.1
Angstroms with a fairly heavy force constant, but in some cases (especially
the oxygen) atoms are moving further from the starting conformation than
that. Any input would be much appreciated.
Thanks,
Katrina
On July 26, 2017 at 10:53:14 AM, Paolo Tosco (paolo.to...@unito.it) wrote:
Hi Katrina,
you may add multiple constraints, of the same (e.g., positional) or
different kinds. Each of them will result in an additional force field
contribution being added to your system. So you may, for example, identify
a group that you wish to constrain by a SMARTS pattern, identify the
relevant atom indices with GetSubstructMatch(), and add a position
constraint for each of them calling ff.MMFFAddPositionConstraint() in a for
loop.
Hope that helps,
p.
On 07/26/17 17:32, Katrina Lexa wrote:
Thanks Greg!
Thanks Francois for your response.
Where I’m struggling is with generating locally restrained conformations.
Essentially, my question is about this line of code:
ff.MMFFAddPositionConstraint(1, 0.3, 1.0e5)
My reading of the code page is that the first variable, 1, describes which
atom the position constraint is being added to. So in this case, it’s being
added to atom 1. What if I want to add the constraint to all (or some) of
the atoms in my input compound?
On July 25, 2017 at 8:47:49 PM, Greg Landrum (greg.land...@gmail.com) wrote:
Hi Katrina,
welcome to the RDKit community!
On Mon, Jul 24, 2017 at 10:45 PM, Katrina Lexa < <kl...@umich.edu>
kl...@umich.edu> wrote:
>
> I'm relatively new to RDKit, so I apologize for what may be a silly
> question. I'd like to generate a set of local minimum conformations around
> my input conformation, using a set of defined flat bottom potentials (0.2,
> 0.6, 1.0, and 1.4), in order to better compare my bound conformation with
> its unrestrained minimum.
>
> Ultimately, I'd like to write a script that reads in one input 3D
> conformation and prints out the energy and rmsd (and structure) of those
> local minimums as well as the lowest energy unrestrained conformation.
>
> There are plenty of examples of conformer generation scripts for RDKit,
> but I'm struggling with the constraints piece. I see the MMFFFixedAtoms &
> MMFFPositionConstraints code, but based on the code it's not obvious to me
> whether it's possible to implement either as a constraint on all atoms in
> my input molecule. Is there a way or is there another route I should be
> taking?
>
Apologies for asking what may be an obvious clarification question, but
what are you planning on constraining? It sounds like you want to attach
"springs" (well, flat bottom potentials) between the atoms in your
generated conformations and the corresponding atoms in a reference
conformation and calculate the energy of that system. Is that right?
If so, you can do this by generating a force field for the molecule with
your generated conformations, adding fixed points corresponding to your
reference conformation, and then adding distance contraints with the
appropriate min (I guess 0) and max values and reasonable force constants.
Subtracting the energy of the generated conformation from the energy of
this full system gives you the energy contributed by the constraints.
I can provide a bit of sample code showing how to do this, but I want to be
sure that this is actually what you're asking for.
-greg
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Rdkit-discuss mailing
listRdkit-discuss@lists.sourceforge.nethttps://lists.sourceforge.net/lists/listinfo/rdkit-discuss
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org!
<http://sdm.link/slashdot_______________________________________________>
http://sdm.link/slashdot_______________________________________________
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
