In <3fpdbb$d8u@planchet.rutgers.edu>, Grant Callaghan <grantc@cts.com> writes:
>I wonder why I don't see anyone talking about using the data
>encoded in DNA to create self-replicating machines and tools
>that will only have to be engineered, not manufactured. It
>seems the obvious route to go, since we are already deciphering
....
>engineering methods. Trying to assemble nanotools through current
>manufacturing techniques seems to be barking up the wrong tree.
>
>grantc@crash.cts.com
Ah, but there is one real problem with this: DNA encodes sequences of amino acids which form a long chain which then fold into proteins. There is currently no effective way to predict protein folding, save by experience. One can, to some extent, design the structure of the protein first and then sequence the necessary amino acids from that, but that still implies full knowledge of the functionality of the proteins and enzymes in the system. We are nowhere near that level. We are, in my opinion, closer to the mechanical route. (One added problem is the distinct lack of rigidity in almost all proteins, which makes them so versatile and yet so unpredictable.) Not to shoot down the idea, as I think someday we will use DNA to engineer structures, but those will be protein based life structures highly different from rigid mechanical structures.
Arioch