Hi Greg,
Should that be :
dm = Chem.GetDistanceMatrix(mol)
I can find no Get3DDistanceMatrix defined?
For a list of molecules, do we recalculate the dm for each one, or do we use
one molecule’s dm as a ‘reference’?
Without trawling through the source code, I’m not clear what’s actually being
done here as the documentation is a bit Spartan.
Is there any reference to a journal article?
Cheers,
Steve.
From: Greg Landrum [mailto:greg.land...@gmail.com]
Sent: 01 May 2014 14:57
To: Stephen O'hagan
Cc: rdkit-discuss@lists.sourceforge.net
Subject: Re: [Rdkit-discuss] 3D-Pharmacophore fingerprints ?
Steve,
On Thu, May 1, 2014 at 12:23 PM, Stephen O'hagan
<soha...@manchester.ac.uk<mailto:soha...@manchester.ac.uk>> wrote:
Would it be possible to generate 3D-pharmacophore fingerprints similar to the
existing 2D ones?
Yes. The function Generate.Gen2DFingerprint() takes an optional argument "dMat"
which can be used to provide the distance matrix. If you pass this a 3D
distance matrix, you get a 3D pharmacophore fingerprint.
Here's a crude example:
In [34]: m = Chem.MolFromSmiles('OCCCCCN')
In [35]: AllChem.EmbedMolecule(m)
Out[35]: 0
In [36]: dm = Chem.Get3DDistanceMatrix(m)
In [37]: from rdkit.Chem.Pharm2D import Gobbi_Pharm2D,Generate
In [38]: factory = Gobbi_Pharm2D.factory
In [39]: sig1 = Generate.Gen2DFingerprint(m,factory)
In [40]: sig2 = Generate.Gen2DFingerprint(m,factory,dMat=dm)
In [41]: sig1==sig2
Out[41]: False
In [42]: sig1.GetOnBits()[0]
Out[42]: 116
In [43]: sig2.GetOnBits()[0]
Out[43]: 115
In [44]: factory.GetBitDescription(115)
Out[44]: 'BG HA |0 3|3 0|'
In [45]: factory.GetBitDescription(116)
Out[45]: 'BG HA |0 4|4 0|'
I hope this helps,
-greg
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
