Build 3D-coordinates from 2D drawings or 1D string: Difference between revisions

no edit summary
No edit summary
 
Line 1: Line 1:
There are several options to create 3D coordinates from it's 2D drawing:
There are several options to create 3D coordinates from it's 2D drawing or a SMILES string:


You can build any 3D molecule easily in Discovery Studio Visualizer - It is free but not for Mac OS. Also see http://www.p212121.com/2009/11/05/discovery-studio-visualizer-2-5/ and http://www.p212121.com/2009/12/08/3-selection-methods-in-discovery-studio-visualizer/ .
You can build any 3D molecule easily in Discovery Studio Visualizer - It is free but not for Mac OS. Also see http://www.p212121.com/2009/11/05/discovery-studio-visualizer-2-5/ and http://www.p212121.com/2009/12/08/3-selection-methods-in-discovery-studio-visualizer/ .
Line 10: Line 10:


Or with the following three easy steps:
Or with the following three easy steps:
# Define a SMILES string for your compound.  The java molecular editor makes this easy:  http://www.molinspiration.com/jme/
# Define a SMILES string for your compound.  The java molecular editor makes this easy:  http://www.molinspiration.com/jme/
# Paste the smiles string into COOT, or feed it into phenix.elbow
# Paste the smiles string into COOT, or feed it into phenix.elbow
# Output the pdb file.
# Output the pdb file.
COOT is faster but phenix.elbow allows you to minimize the structure, even using third-party QM programs like GAMESS.


COOT is faster but phenix.elbow allows you to minimize the structure, even using third-party QM programs like GAMESS.
Next possibility: http://zinc.docking.org/ - draw the molecule using the widget, save the smiles file and let their server make the 3D version for you, properly protonated etc. And you get things like logP, and other descriptors as well. True you get it as .mol2.gz or .sdf.gz but babel is your friend
1,328

edits