Speculation on biological computing

The Register reports on developments in DNA storage, allowing vast amounts of data to be stored in a tiny volume, albeit with very slow access times making it only useful for archival.

Meanwhile, researchers have built a compiler from Verilog to DNA allowing cells to be programmed in a high level language. It seems possible to me (knowing nothing about the details) that this could maybe be combined with DNA storage to speed up access to data. The limited size of the programs could benefit from demo scene techniques to do a lot with a little.

Combined with gene therapy techniques to transfer DNA between organisms, I wonder if it might be possible to create a virus that transmits the complete works of Shakespeare (etc) together with machinery that allows neurons to interact with it, like a biological Matrix-esque knowledge implant system.

It should be relatively simple on the other hand to inject material to be archived into organisms as "junk DNA", so you could grow plants with the entire library of human written knowledge, and distribute seeds to your friends and neighbours. Still IO is the problem - getting the data out in human-readable form.