Termination control

Examples of termination control:
– rotationally symmetric assembly with specifiable number of segments left out
– in self-assemlbly terminating A grid of voxels (carterian, hexagonal, or whatever) that when self-assembled deterministically terminates this self assembly at some point.

Full control:
The shape after final termination of self assembly shall be arbitrary.
Arbitrary shape means both arbitrary shape of external surface and internal surfaces.
Thus full termination control to the highest possible degree implies individual addressability of all voxels.
Voxels can be set or unset (addressed) individually.
Kinda like "random access memory" (but only once in the potentially irreversible assembly process).

Using the terminology of "addressing voxels" may be confusing:
There are no changes after assembly. Static structural structures are assumed here.

Some control:
Algorithmic selfassembly is allowing at least some but not full control over termination of self-assembly
Only (not necessarily contiguous) subsets of those voxels can be set or unset (addressed) indirectly.

No control:
⚠️ Zero termination control does not imply no termination!
When a selfassembly goes around 360° and meets of with it's start then it terminates despite no termination control imposed.
One could count adjustment in curvature as limited amount of termination control but this will not be done here since
the aim here is to find a definition that helps in pushing forward technological capability generally rather than
for each new problem (each new curvature) individually.

Delineation from superficially impressive firatures like size and symmetry:
Infinite symmetries like a non-terminating rod, plane, or volume are considered to possess zero termination control.
Note that this also holds for full 360° rotational symmetry despite there actually being termination.
Full circle termination is considered to be not a controlled one.
Full circle termination is rather considered to be accidental by the fact that ends happen to meet up.

Limiting factors in termination control and countermeasures

De-novo proteins

• Artificial de-novo proteins can only provide a small set of orthogonal interfaces.
  And it is hard to get high specificity and activity at the same time
• Their otherwise desired stiffness gets in the way.
  There is no DNA like step by step unzipping process possible.
  All bond breaking energy needs to be supplied in one fell swoop.
  That is: Kinetic traps can become a bigger problem.
• Internal cohesion, external intefaces and eventual other external functionality are
  three mutually competing factors on the choice of side-chain sequences.
  Such a situation is not present in structural DNA nanotechnology.

Countermeasures:

• Squigglesembly
• Circumsembly
• Integrate them into more termination-control-scalable background framework
  like e.g. structural DNA nanotechnology

Structural DNA nanotechnology

• Diffusuion time.
  Especially for larger 3D structures exclusively made from short strands (oglionucleotide staple strands) the achievable yields drop down far.

Countermeasures:

• Inclusion of longer strands (backbone scaffold strands) can increase effective concentration and
  thereby speed up selfassembly considerably. This comes with tradeoffs though. (wiki-TODO: refresh on that one)
• Hierarchical self-assembly aka convergent self-assembly
• Maybe tether-based / hinged-based assembly approaches?

Spiroligomers

• Their otherwise desires stiffness gets in the way of self assembly and termination control.
  Due to them being less conforming they are less and tolerant to variations in complementary surface shapes. Also they are small-ish limiting the amount of encodable shape.
• They are still quite small molecules which limits their reachion-site-choosing-specificity. Similar to the site specificity limitation problem in conventional chemistry.

Countermeasures:

• Integrate into more termination-control-scalable background framework like e.g. as artificial side-chains in de-novo proteins.

Related

• Incremental path
• Self assembly
• Algorithmic selfassembly
• Fully addressable lattice selfassembly
• Hierarchical selfassembly

(TODO: look into formally quantifying the quantity of termination control)