Duane L. Marcy - US grants

EIA-0129731
Robert R. Birge
Syracuse University

Title: Optimized Bacteriorhodopsin Proteins for Photonic and HolographicMemory Storage

The protein bacteriorhodopsin is being optimized for bioelectronic and photonic applications through site directed mutagenesis, random mutagenesis and directed evolution. The native protein has many characteristics that make it a nearly optimal photonic material, but further optimization is necessary to achieve competitive performance in computer applications. The goal is to generate protein variants that are optimized for holographic associative memories and large scale volumetric memories. The former memory system is based on Fourier-transform holography, and the protein will be optimized for thermal stability, quantum efficiency and holographic efficiency.

The large scale volumetric memory is based on a branching reaction out of the main photocycle. The protein is being optimized for this application by increasing thermal stability, branching efficiency and long-term stability of the branched photoproduct. Genetic engineering is the best method for achieving all three goals simultaneously, and directed evolution is being studied as one possible method of rapidly achieving these goals simultaneously. This is a collaborative project involving researchers at the W. M. Keck Center for Molecular Electronics at Syracuse University and molecular biologists at the University of Connecticut.

The increasing demands of modern computation require innovative solutions to data storage and processing. Conventional architectures are often limited not by processor speed, but by the ability to efficiently handle large amounts of data. Optical architectures for these applications promise potential gains in speed, data throughput, & storage density. In addition, an entirely new class of materials has emerged that is expected to provide practical solutions to these demanding applications; these materials include not only conventional optical & holographic materials, but also of those of biological origin. Bacteriorhodopsin (BR) was among the first proteins to attract attention as a viable component in molecular electronics & nanotechnology. In the native organism, Halobacterium salinarum, the protein acts as a photosynthetic sunlight to chemical energy transducer. Through billions of years of evolution, nature has produced a protein that is remarkably rugged and responds quickly and efficiently to light. The protein's unique properties make it an appealing candidate for a number of devices that use light to record or process information, including holography and optical computer memory storage.

Despite bacteriorhodopsin's unique properties, it still lacks the efficiency necessary for commercialization. Fortunately, advanced techniques in molecular biology can be used to optimize the protein for specific device architectures. Through techniques such as random mutagenesis, semi-random mutagenesis, and directed evolution, researchers can custom-tailor protein properties in ways never before possible. The use of directed evolution techniques to produce BR variants for optical recording materials will result in reusable holographic media that require no processing or fixing, and are capable of real-time operation. The goal of this research effort is to develop a new class of fully write-read-erasable dynamic holographic recording media based on genetically-engineered bacteriorhodopsin variants. Each new material will be evaluated to determine which is the most efficient, with respect to a number of standard holographic benchmarks.

The targeted application we hope to develop is a bacteriorhodopsin-based holographic associative memory, which simulates the way the human brain works. Associative memory architectures are not new, but their utility has been limited by the absence of materials that facilitate highly efficient implementation (i.e., the lack of truly reusable media). Associative memories allow computers to identify objects and concepts faster and more efficiently- applications include any technology that requires autonomous general-pattern recognition and/or fast & efficient large-scale database search capabilities. Information processing techniques such as data mining, data reduction, and large-scale complex database searches will be enhanced through the successful development of these architectures. Furthermore, this technology has the potential to play a critical role in the development of artificial intelligence (AI)-successful implementation of AI architectures will require a fast & efficient large-scale database search capability. Incorporation of dynamic write-read-erase materials into pre-existing associative architectures will introduce a level of flexibility not previously possible.

To summarize, the proposed effort uses advanced molecular biological techniques to produce genetically-engineered bacteriorhodopsin (BR) proteins for holographic associative memories. The technical merit & broader impacts of this technology must therefore be considered at multiple levels, including (1) development of novel holographic materials & (2) the ramifications of viable associative memories. The former will impact any holographic technology that will benefit from a real-time reusable media (e.g., optical memory & non-destructive testing architectures), while the latter will facilitate any technology that will benefit from the ability to utilize fast & efficient large scale database capabilities (e.g., artificial intelligence, proteomics, and the human genome project). Perhaps the most basic impact of this technology will be the demonstration of random mutagenesis and directed evolution as viable techniques for the production of custom-tailored proteins to be used in a variety of applications.