Dear Sefano, Thanks for the reply, I was suspecting the same. However I need the PDB 3D coordinates mapped on the original SMILES, that's why I was trying reading the PDB. I am wondering if there are any methods to map the PDB coordinates of a molecule to its original SMILES?
Thanks, Sam On Wed, Aug 31, 2016 at 5:14 PM, Stefano Forli <fo...@scripps.edu> wrote: > Sam, > I wouldn't trust the PDB for either bond order or connectivity. > In OB the order is inferred from the distances, which are dependent on the > quality of the structure and the software used for the model refinement. > > For example, I found that in structures generated a version of Refmac, the > oxygen double bond lenght in the DMSO is wrong, so OB assigns an sp3 > hybridization to it. > > If you have to do that in a high-throughput fashion, I would use the > ligand ID to download the actual SMILES from the ligand repository at the > PDB and use that for your analysis. > > Best, > > S > > > > On 08/31/2016 01:51 PM, Sam Tonddast-Navaei wrote: > >> I am trying to read a small molecule from PDB file and match its atom >> numbers to the same >> molecule in a SDF file. I have tried both matching SMART patterns and >> using OBIsomorphismMapper, both work for 70% of cases. However there are >> cases for which >> OpenBabel can not simply get the right SMILES from just the PDB file >> (e.g. adds extra >> double bonds or miss ones). For example in PDB ID 1KF6 and ligand name >> IPE, if you >> generate SMILES from PDB you will get OCCO/C=C/OCCO/C=C/O/C=C/O while the >> correct one >> should be OCCOCCOCCOCCOCCO. I would appreciate if someone can give me >> some advise on this. >> >> Thanks, >> Sam >> ------------------------------------------------------------ >> ------------------------------ >> NOTE: This message was trained as non-spam. If this is wrong, please >> correct the training >> as soon as possible. >> >> >> ------------------------------------------------------------ >> ------------------ >> >> >> >> _______________________________________________ >> OpenBabel-discuss mailing list >> OpenBabel-discuss@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss >> >> > -- > > Stefano Forli, PhD > > Assistant Professor of ISCB > Molecular Graphics Laboratory > > Dept. of Integrative Structural > and Computational Biology, MB-112A > The Scripps Research Institute > 10550 North Torrey Pines Road > La Jolla, CA 92037-1000, USA. > > tel: +1 (858)784-2055 > fax: +1 (858)784-2860 > email: fo...@scripps.edu > http://www.scripps.edu/~forli/ >
------------------------------------------------------------------------------
_______________________________________________ OpenBabel-discuss mailing list OpenBabel-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
