On Thu, Mar 14, 2019 at 2:12 AM Marco Podda <marcopodda1...@gmail.com> wrote:
> awesome! Just what I was trying to achieve! > Great. > > Just one additional question, although it's more curiosity than anything. > I am supposing that if you "lose" the parent molecule, it is not > possible to go back and retrieve it deterministically from the fragments > alone. One should just apply BRICSBuild up to some depth and then "assume" > (or rather, "hope") the original molecule is among the ones that the > algorithm has recombined. Am I correct? > The process is definitely deterministic, so it's not "assume". If you know the maximum dept that you fragment to, then you know how to ensure a maximum depth for the rebuilding process. The group of molecules that comes out of this will contain the input molecule. > I was thinking about whether there is a particular fragment order that > could foster the recovery of the original molecule, but that doesn't seem > to be the case. > Are you aware of any fragmentation algorithm that is able to go both ways > (i.e. from the molecule to the fragments, and from the fragments to the > same molecule)? > There is almost certainly a way to implement the BRICS decomposition/build process such that this is possible, but I'm wondering why you would want to. It's much easier to just keep the original molecule around with the fragments. -greg
_______________________________________________ Rdkit-discuss mailing list Rdkit-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
