Protein Engineering

One of the fundamental goals of nanotechnology in the long run is to engineering proteins for use in nano-machinery.  There exist many forms of molecular machinery in the body, from ion pumps to power plants.  They operate on chemical principles built on protein backbones.  Ideally, we should be able to copy the machinery and create our own molecular machines for a variety of interesting nanotechnology applications.

In order to do so, however, we'd need to accurately engineer a protein to fold in the proper sequence.  Currently, most research is going into predicting how existing proteins fold.  It is a computationally intense problem that is leaps and bounds more difficult than figuring out how to make a protein to fold the right way.

Nanotechnology pioneer, K. Eric Drexler, was the first to publish an article suggesting that we could co-opt proteins to work for us exactly as designed.

In his paper titled Molecular Engineering: An Approach to the Development of General Capabilities for Molecular Manipulation, Drexler proposed that we could conceivably pursue an entire pathway for fabrication of nano-machinery through exact control over protein fabrication.

Once these machines were made, they could be used to make even more complex machines.  Drexler saw how the possibilities were endless, and avidly pushed for more research in the field.

As it stands, we haven't gotten very far towards that goal.  Protein folding is a very complex subject to treat.  They require self-consistent calculations that could lead nowhere despite days of computation time.  Some progress has been made in a variety of bottom-up techniques that are a necessary prerequisite for protein engineering.

In order to build proteins, we need to actually use machinery already present in the body.  The first step would require gene synthesis and recombinant DNA technology.  This could trick a ribosome into manufacturing a specific sequence of proteins tailored for our use. 

Unfortunately, even though protein engineering is theoretically an easier task than protein structure prediction, it's pretty much impossible to design the right sequence without first understanding how natural proteins fold.  Our understanding of proteins is still quite limited.  If we ever prefect our knowledge of prediction, then it will open up the path to the molecular machines that researchers are so excited about.