Learning from enzymes
For piezochemical mechanosynthesis we do not want to recreate the complex folded background framework of proteins.
We mostly want too look at and learn from the local geometry right around the reaction site.
A question is in how far the optimal trajectories differ with the big advantage of piezochemical mechanosynthesis to be capable of applying huge forces and torques onto the chemical bonds.
Regarding the fat finger problem
Most of the floppy "fingers" (side chains) of enzymes are actually just there to recognize and capture the molecule(s) to be transformed.
Correct position is encoded in shape. This is not necessary for machine phase systems.
Generally:
- Getting eight "fingers" together is quite easy in case of resource molecule preparation
- Getting four "fingers" together is quite easy above a flat surface
In a streaming assembly line like setup (which subtracts one degree of freedom):
- Getting three "fingers" together is quite easy in case of resource molecule preparation
- Getting two "fingers" together is quite easy above a flat surface
Ideal targets for piezochemical mechanosynthesis are
- small point like resource molecules and
- volume filling gemstone like crystolecules.
Both stiff.
But even chain like molecules can be handled when constrained sufficiently by stretching sections of them out.
Mechanosynthesis of diamond actually requires synthesizing short chain appendages to form the loops of the diamond crystal structure.
This is outlined in meticulous detail in the tooltip cycle paper.
