Hi Hans, Thanks for piping up.
On Wed, Oct 15, 2008 at 9:19 PM, Hans Purkey <[email protected]> wrote: > If the intention is to follow Lipinski's definitions of Hbond acceptors, > then it should be a simple N+O count (look back at the original paper and > that is how he difined it "for simplicity"). > Agreed. The Lipinski def'n for HAcceptor is encoded as the function NOCounts. There's actually some explanation of this in the comments of $RDBASE/Python/Chem/Lipinski.py: # 2 definitions from Gobbi Paper # NOTE: if you want traditional Lipinski numbers, you # should use NOCounts (below) instead of HAcceptor # The Gobbi paper referred to is: Gobbi and Poppinger, Biotech. Bioeng. _61_ 47-54 (1998) > However, if the descriptor is intended to match a more intuitive/realistic > definition of HBA, then N-H shouldn't be a part of it. Here I mostly also agree too. There's a build-up of a more restrictive definition in the file $RDBASE/Data/BaseFeatures.fdef. Documentation for that file format (which Hans is familiar with) is in: $RDBASE/Docs/Programs/RDPharm3D/FDefFile.htm It would be simple to completely exclude atoms with Hs from the definition of H bond acceptors. I'd like to make this change once, so before I do any fixes, it would be nice to see if there are dissenting voices. Best Regards, -greg
