Enzymes synthesizing diamond and similar macromolecules or molecular networks can easily be developed by mechanomeric selection (MeSe).

For the critical argument with regard to mechanomeric selection, see the blog-post here

http://www.nanopaprika.eu/profiles/blogs/enzymes-and-molecular-mach...

A very concise report giving the background to that argument, the argument itself, a technological description of mechanomeric selection, and a small number of brief descriptions of medical and industrial applications of that technology can be seen at

http://mechanomers.blogspot.com/2012/03/enzymes-and-molecular-machi...

And a more expansive and easier-going introduction to mechanomeric selection in the form of a pseudoblog/FAQ can be seen at

http://mechanomers.blogspot.com

(start at the bottom, in blog fashion).

The mechanomeric selection of an enzyme for such synthesis will involve matricial mechanomeric selection (see the report and pseudoblog/FAQ linked to above): matriciating (usually meaning chromatographically two-dimensionally arraying) replicated random mechanomers (functional copolymers) of some artificial class (to avoid unintended medical and/or ecological consequences), say polyesters or polysulfonamides; overlaying the replicated random mechanomer so produced with the solvent and substrate or reactant to be used in such synthesis; and examining that matrix for the emergence of diamond, preferably using mechanomeric indication by overlay of that matrix with a previously-developed enzyme, an indicase, and its substrate or indicator together reacting to and therefore indicating the presence of diamond by catalyzing an indicating reaction causing a color-change (see the report and pseudoblog/FAQ), but possibly by scanning x-ray looking for bursts of the typical diffraction pattern of diamond; followed by extraction of the mechanomers from locations where diamond is being produced, or from corresponding locations on a parallel matrix (see the report and pseudoblog/FAQ); replication of those mechanomers; their further testing for the desired catalysis, for example by further selections using different matriciations; and final production of the enzyme selected to be used in such synthesis by replication by the replicase used to create the original replicated random mechanomer stocks used in that enzyme's selection (see the report and pseudoblog/FAQ).