Dear Igor, On Tue, Aug 5, 2008 at 10:17 PM, Igor Filippov [Contr] <[email protected]> wrote: > > Great, I got it to work - there is one small problem though. > When I get SMILES string there is always an extra property with the > value of "0", doesn't exist in the SD output and I'm not sure where it's > coming from: > [H]Oc1c2c3c(cc1)CC1([H])N(C)CCC33C1([H])C=CC([H])(O[H])C3([H])O2 0 72 > 0.94463 > ^^^ > this zero after SMILES and before "72".
That's the slot for the molecule name (it's "_Name" property). If the molecule doesn't have a name, the Smiles writer dumps the sequence number of the molecule. So in your example above, that was the first molecule (sequence number zero) in the file. So that's what it is... now as to how to make it stop... that's trickier. The "easiest" (by which I mean easiest for me) thing is to assign _Name properties to your molecules, then you won't see the sequence numbers. At the moment there is no way to suppress the _Name output to SMILES files. That's an oversight which I'll fix. I have to do some investigating to see how much I will break if I modify the SmilesWriter to not generate names automatically; if it doesn't cause too many problems I will do that. If there are problems associated with switching off the automatically generated names, I will instead add an option to suppress names on-demand. -greg
