Compiling a virtual structural library of metabolic intermediates and natural ligands

Greetings,
 I am in the process of compiling a database of metabolic intermediates
 (small molecules) and natural ligands. I am aware that this is
 not a trivial undertaking. - I am planning to make this available on the
 internet as an interactive database, but I will need a fairly
 complete startingpoint of annotated 2D/3D structures. I do have a library I
 have compiled myself already and would like to have user input.
 Some general questions that someone might like to ponder about are:
 	1) How large would such a library need to be if we consider both
 procaryotic and eucaryotic metabolism?
 	2) How should one deal with tautomers in such a database?
 Any other suggestions and reflections on the idea are also welcome !
 I do NOT need pointers to other databases as you cannot compile databases
 > from other databases without infringing copyright and ownership -
 especially
 if you want to make the data publicly available.
 Please if you have any sets of ligands/metabolites/catabolites which you
 feel should be in there I would be happy to add them in.
 A redundancy check will be performed when adding them to the database. The
 annotation could contain any of the following fields:
 	Trivial name
 	IUPAC name
 	Structure (sdf)
 	Pathway/Reaction
 	etc ...
 Many regards
 		-- Jan
 Dr. Jan Torleif Pedersen
 Section head, Bioinformatics
 Department of Molecular Genetics
 H. Lundbeck A/S
 Ottiliavej 9, 2500 DK-Valby
 Denmark
 phone: +45 36 43 28 87
 FAX: 36438271
 email: jatp$at$lundbeck.com