On Apr 11, 2006, at 5:07 PM, Donald Ephraim Curtis wrote: > information, along with requiring the code to find molecules that > are the same. For example, selecting one molecule and finding all > similar molecules (finding all water molecules). I was just > wondering what facilities are already included in the ghemical code > that would help do this, or any algorithms that would help. One > idea was to convert the molecules using openbabel to the SMILE > format and then doing a string comparison. Any suggestions would > be helpful. You don't need to even do that. Besides, it's generally a bad idea to do straight string comparisons with SMILES. There's no guarantee that equivalent molecules necessarily have equivalent SMILES strings. * Inside Open Babel, you can easily do SMARTS matching (e.g., if this set of atoms matches SMARTS of [OD2]). * There isn't support for "equivalence" (e.g. mol1 == mol2) but that's certainly a reasonable feature request. * Open Babel 2 also supports similarity searching using molecular fingerprints. So to answer your question, using code like project::HtoMethyl() you can sync to an OBMol and write code to search for water molecules. Cheers, -Geoff