Re: [Rdkit-discuss] GetSubstructMatch vs MMFFOptimize

[Paolo T. <pao...@un...>]

Dear Robert,

the reason of the failure is that MMFF uses its own aromaticity model 
(see 
http://www.rdkit.org/docs/GettingStartedInPython.html#working-with-3d-molecules). 
Therefore, after calling

AllChem.MMFFOptimizeMolecule(mols[0])

you will need to add the following call:

AllChem.SanitizeMol(mols[0],
   sanitizeOps = AllChem.SanitizeFlags.SANITIZE_KEKULIZE \
   | AllChem.SanitizeFlags.SANITIZE_SETAROMATICITY)

This will fix your problem.

Kind regards,
Paolo

On 05/14/2016 07:07 AM, Robert DeLisle wrote:
> RDKitters,
>
> I'm working on a project in which I want to align a collection of 
> structures with their most similar structures and display the results 
> in PyMOL.  To accomplish this, I've built a Python script similar to 
> the one attached here in which I start with pairs of structures, find 
> the MCS of those structures, create a template based on the MCS and a 
> 3D conformation of the structure of interest, and then generate a 
> constrained conformation of a query structure.  I tried to comment the 
> attached code enough to lead you through the process.
>
> What I find is that quite often, the ConstrainedEmbed() function fails 
> with the error "molecule doesn't match the core" which seems very odd 
> since the pairs for which it fails are very similar. The attached .png 
> shows one such pair and their MCS.
>
> What I've found is that when I generate a 3D conformation for the 
> first structure and optimize it with MMFF (MMFFOptimize), this often 
> causes GetSubstructMatch to fail finding the MCS within the 
> structure.  If instead I used UFFOptimize, everything seems to work OK 
> most of the time.
>
> In my code, I've noted where the error occurs and flanked it with some 
> print statements to show what happens. Specficially, at like 36 I have 
> the MMFFOptimize line, and at 37 the UFFOptimize line.   I've also 
> attached a set of structures for which MMFF fails.
>
> While using UFFOptimize produces great results, I'm curious regarding 
> why MMFFOptimize creates a problem.  And, whether this is a bug which 
> should be fixed, or just a glitch related to atom typing and other 
> parameterizations that occur with MMFF.
>
> Thanks for any explanation or ideas.
>
> -Kirk
>
>
>
>
> ------------------------------------------------------------------------------
> Mobile security can be enabling, not merely restricting. Employees who
> bring their own devices (BYOD) to work are irked by the imposition of MDM
> restrictions. Mobile Device Manager Plus allows you to control only the
> apps on BYO-devices by containerizing them, leaving personal data untouched!
> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
>
>
> _______________________________________________
> Rdkit-discuss mailing list
> Rdk...@li...
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss