Protein-Based Memories and Associative Processors
aDept. of Chemistry, University of Connecticut,
Storrs, CT 06269 USA
bW. M. Keck Center for Molecular Electronics
This is an abstract
for a presentation given at the
Foresight Conference on Molecular Nanotechnology.
There will be a link from here to the full article when it is
available on the web.
This presentation will explore the use of the protein, bacteriorhodopsin, in optical three-dimensional memories and parallel associative processors. Three-dimensional memories store information in a memory volume element, and provide as much as a thousand-fold improvement in memory storage capacity over current technology. The comparative advantages and disadvantages of holographic, two-photon and sequential one-photon volumetric architectures will be discussed. The associative memory operates in a fashion somewhat analogous to the human brain and responds to input data by finding (in a few nanoseconds) the closest match within the data base and feeding this information, and any associated information, to the output. Such a memory is critical to the development of artificial intelligence. The use of site directed mutagenesis and directed evolution to improve the properties of the protein for specific applications will also be discussed. Although working prototypes have been developed, a number of cost/performance and architectural issues must be resolved prior to commercialization.
Dept. of Chemistry, University of Connecticut
55 North Eagleville Rd, Storrs, CT 06269 USA