Dear all, Well, now I'm incredibly behind in all this. I will try to slowly catch up.
Andrew: thanks for putting in the effort to do this analysis. I have some notes from an earlier email of yours that I will post this morning. It's probably going to take me a while to mull through the rest of the new stuff and to do an RDKit reimplementation of Andrew's scripts (you don't mind, do you?). I do have one quick question: I didn't see anything about inclusion of branched paths. I think (and not just because the RDKit fps include branching!) that this might help add some nicely discriminating bits at relatively low cost. More on that later. I've generated a new, larger, testing dataset using the pubchem HTS compounds. I will also post the details on those (hopefully this morning). -greg
