Re: [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)

Visvaldas
Perhaps this should have gone a to a new thread...

I have a couple of questions about RMSD. Maybe they could lead to "requested features":)

1) from the documentation, it seems that rmsd calc _always_ does superposition of the molecules before calculating RMSD.  From some older (2007) openbabel thread it seams there is no way to compute RMSD without superposition. This would be very useful, say, in ligand-protein docking calculations.

2) The computed RMSDs are "symmetry corrected". There is a global symmetry and a local symmetry. Is the local symmetry also calculated for the purposes of symmetry correction when doing RMSD? Actually, it's not a trivial problem; one way to tackle it is to permute rotations around all rotatable dihedrals in a reference molecule, and to try to obtain a nearly identical molecule ("nearly" because of imperfections). Perhaps other algorithms exist. An example would be a local symmetry of the phenyl ring in phenylalanine: phenylalanine is non-symmetric, but phenyl inside it has two pairs of equivalent carbons. 

Best regards,

Vis

------------------------------------------------------------------------------
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
_______________________________________________
OpenBabel-scripting mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
Reply | Threaded
Open this post in threaded view
|

Re: [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)

Noel O'Boyle
Administrator
Regarding (2), it does handle what you call "local symmetry" (automorphisms).

Regarding (1), it wouldn't take long to write a C++ program (not sure
about Python) that used Open Babel to calculate the RMSD in the way
you describe. But I guess you are asking for an option to obabel,
right?

- Noel

On 16 January 2012 20:25, Visvaldas K. <[hidden email]> wrote:

> Perhaps this should have gone a to a new thread...
>
> I have a couple of questions about RMSD. Maybe they could lead to "requested
> features":)
>
> 1) from the documentation, it seems that rmsd calc _always_ does
> superposition of the molecules before calculating RMSD.  From some older
> (2007) openbabel thread it seams there is no way to compute RMSD without
> superposition. This would be very useful, say, in ligand-protein docking
> calculations.
>
> 2) The computed RMSDs are "symmetry corrected". There is a global symmetry
> and a local symmetry. Is the local symmetry also calculated for the purposes
> of symmetry correction when doing RMSD? Actually, it's not a trivial
> problem; one way to tackle it is to permute rotations around all rotatable
> dihedrals in a reference molecule, and to try to obtain a nearly identical
> molecule ("nearly" because of imperfections). Perhaps other algorithms
> exist. An example would be a local symmetry of the phenyl ring in
> phenylalanine: phenylalanine is non-symmetric, but phenyl inside it has two
> pairs of equivalent carbons.
>
> Best regards,
>
> Vis
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2
> _______________________________________________
> OpenBabel-scripting mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
>

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
OpenBabel-scripting mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
Reply | Threaded
Open this post in threaded view
|

Re: [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)

Visvaldas
In reply to this post by Visvaldas
Dear Noel,


Thank you for your answers!

For my 1st question, I had in mind something along the lines of openbabel python bindings. However, a "standalone" version of the rmsd calculation could be probably useful for quite a few people.

Vis




________________________________
From: Noel O'Boyle <[hidden email]>
To: Visvaldas K. <[hidden email]>
Sent: Tuesday, January 17, 2012 11:40 AM
Subject: Re: [OpenBabel-scripting] [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)

Regarding (2), it does handle what you call "local symmetry" (automorphisms).

Regarding (1), it wouldn't take long to write a C++ program (not sure
about Python) that used Open Babel to calculate the RMSD in the way
you describe. But I guess you are asking for an option to obabel,
right?

- Noel

On 16 January 2012 20:25, Visvaldas K. <[hidden email]> wrote:

> Perhaps this should have gone a to a new thread...
>
> I have a couple of questions about RMSD. Maybe they could lead to "requested
> features":)
>
> 1) from the documentation, it seems that rmsd calc _always_ does
> superposition of the molecules before calculating RMSD.  From some older
> (2007) openbabel thread it seams there is no way to compute RMSD without
> superposition. This would be very useful, say, in ligand-protein docking
> calculations.
>
> 2) The computed RMSDs are "symmetry corrected". There is a global symmetry
> and a local symmetry. Is the local symmetry also calculated for the purposes
> of symmetry correction when doing RMSD? Actually, it's not a trivial
> problem; one way to tackle it is to permute rotations around all rotatable
> dihedrals in a reference molecule, and to try to obtain a nearly identical
> molecule ("nearly" because of imperfections). Perhaps other algorithms
> exist. An example would be a local symmetry of the phenyl ring in
> phenylalanine: phenylalanine is non-symmetric, but phenyl inside it has two
> pairs of equivalent carbons.
>
> Best regards,
>
> Vis
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2
> _______________________________________________
> OpenBabel-scripting mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
>

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
OpenBabel-scripting mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
Reply | Threaded
Open this post in threaded view
|

Re: [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)

Noel O'Boyle
Administrator
Oh yes, I forgot. I've already done this some time ago...I was just
waiting for 2.3.1 to be released...

https://gist.github.com/974477

- Noel

On 17 January 2012 09:48, Visvaldas K. <[hidden email]> wrote:

> Dear Noel,
>
>
> Thank you for your answers!
>
> For my 1st question, I had in mind something along the lines of openbabel python bindings. However, a "standalone" version of the rmsd calculation could be probably useful for quite a few people.
>
> Vis
>
>
>
>
> ________________________________
> From: Noel O'Boyle <[hidden email]>
> To: Visvaldas K. <[hidden email]>
> Sent: Tuesday, January 17, 2012 11:40 AM
> Subject: Re: [OpenBabel-scripting] [Open Babel] pybel - (howto) calculation of RMSD for two molecules (but different conformers)
>
> Regarding (2), it does handle what you call "local symmetry" (automorphisms).
>
> Regarding (1), it wouldn't take long to write a C++ program (not sure
> about Python) that used Open Babel to calculate the RMSD in the way
> you describe. But I guess you are asking for an option to obabel,
> right?
>
> - Noel
>
> On 16 January 2012 20:25, Visvaldas K. <[hidden email]> wrote:
>> Perhaps this should have gone a to a new thread...
>>
>> I have a couple of questions about RMSD. Maybe they could lead to "requested
>> features":)
>>
>> 1) from the documentation, it seems that rmsd calc _always_ does
>> superposition of the molecules before calculating RMSD.  From some older
>> (2007) openbabel thread it seams there is no way to compute RMSD without
>> superposition. This would be very useful, say, in ligand-protein docking
>> calculations.
>>
>> 2) The computed RMSDs are "symmetry corrected". There is a global symmetry
>> and a local symmetry. Is the local symmetry also calculated for the purposes
>> of symmetry correction when doing RMSD? Actually, it's not a trivial
>> problem; one way to tackle it is to permute rotations around all rotatable
>> dihedrals in a reference molecule, and to try to obtain a nearly identical
>> molecule ("nearly" because of imperfections). Perhaps other algorithms
>> exist. An example would be a local symmetry of the phenyl ring in
>> phenylalanine: phenylalanine is non-symmetric, but phenyl inside it has two
>> pairs of equivalent carbons.
>>
>> Best regards,
>>
>> Vis
>>
>> ------------------------------------------------------------------------------
>> RSA(R) Conference 2012
>> Mar 27 - Feb 2
>> Save $400 by Jan. 27
>> Register now!
>> http://p.sf.net/sfu/rsa-sfdev2dev2
>> _______________________________________________
>> OpenBabel-scripting mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/openbabel-scripting
>>

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
OpenBabel-scripting mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-scripting