Search

The free energy released during the interaction of the 16S rRNA tail with the mRNA sequence during translation contains a weak sinusoidal pattern of frequency 1/3 cycles/nucleotide. We hypothesize that this signal encodes information related to the maintenance of reading frame during elongation. In the case of the well-studied +1 frameshifter, prfB in E. coli, we have observed a direct relationship between cumulative signal phase and reading frame. Based on this observation, we have developed a model that indicates how likely it is for the ribosome to stay in frame throughout the process of elongation. We validate this model by analyzing verified coding sequences in E. coli.

Two methods, power spectrum density analysis (PSD) and synchronization signal approximation, were investigated to determine if underlying periodic, free energy signals could be detected for the individual genes in this paper. These signals could be revealed assuming Watson-Crick type hybridization between the eight, 3'-terminal nucleotides of the 18S rRNA and pre- and mature-mRNA sequences in Saccharomyces cerevisiae in a manner similar to that used to analyze coding region sequences in prokaryotic genes. Using PSD, a periodic signal could only be detected in 35 of 106 genes tested; using the synchronization signal approximation, 91 of 106 genes showed linearly increasing magnitude and phase, characteristics consistent with the presence of an underlying periodic signal with an assumed frequency of one-third. The majority of introns did not show magnitude and phase behavior consistent with an underlying non-periodic signal. The periodicity property for the free energy on the protein-coding regions can contribute to finding the approximate boundaries of the exons (protein coding regions) and the introns, which provides a foundation for future studies in identifying the exact positions of the splice sites, especially for the higher eukaryotic genes.

We describe a system that allows researchers to control an implantable neural microstimulator from a PC via a USB 2.0 interface and a novel dual-carrier wireless link, which provides separate data and power transmission. Our wireless stimulator, Interestim-2B (IS-2B), is a modular device capable of generating controlled-current stimulation pulse trains across 32 sites per module with support for a variety of stimulation schemes (biphasic/monophasic, bipolar/monopolar). We have developed software to generate multi-site stimulation commands for the IS-2B based on streaming data from artificial sensory devices such as cameras and microphones. For PC interfacing, we have developed a USB 2.0 microcontroller-based interface. Data is transmitted using frequency-shift keying (FSK) at 6/12 MHz to achieve a data rate of 3 Mb/s via a pair of rectangular coils. Power is generated using a class-E power amplifier operating at 1 MHz and transmitted via a separate pair of spiral planar coils which are oriented perpendicular to the data coils to minimize cross-coupling. We have successfully demonstrated the operation of the system by applying it as a visual prosthesis. Pulse-frequency modulated stimuli are generated in real-time based on a grayscale image from a webcam. These pulses are projected onto an 11x11 LED matrix that represents a 2D microelectrode array.

This work reports on a novel signal processing method that can be applied to analysis of genetic sequences in prokaryotes to identify their translational characteristics. The methodology involves computation of a signal based on free-energy score of the interaction between the 3' tail end of the 16S rRNA and the mRNA sequence of interest. We find that in the coding region of prokaryotes this signal is has a strong harmonic corresponding to every 3/sup rd/ base position. Noncoding regions appears not have such a signal. We discuss the methodology in detail and we demonstrate its ability to clearly recognize a) the coding region of a single prokaryotic gene, and b) special characteristics of the gene, such as frameshifts. We use E. Coli K-12 genes to illustrate the findings.

The 16s ribosomal tail end has been conjectured to play an important role in the regulation of protein production and of translation efficiency. Using E. coli K-12 as our model organism, we generate sequences of 13 base pairs as hypothetical ribosome tail ends. We analyzed the distributions of these random hypothetical ribosome tail ends and found the actual E. coli ribosome tail end to be significantly different from a randomly generated ribosome tail in the magnitude of the lock and synchronization signals, and the signal to noise ratio. We then designed and ran a genetic algorithm to optimize hypothetical ribosome tail ends simultaneously for these three signal criteria. We found that the actual E. coli ribosome tail end was among the best by these measures.

Blood transfusion errors are not uncommon. In some cases the error is fatal. This is primarily due to lack of an automated system at the point of application and over-reliance on bar-coding and paperwork to catch these critical errors. In emergency situations, human errors contribute to transfusion and transplantation of incompatible blood types and organs resulting in rejection and possible fatality. We present a sensing concept that will monitor blood compatibility between the patient and the transfusion bag before allowing the valve to open for transfusion to take place. This will eliminate all transfusion errors and provide 100% safe transfusions automatically. The operating principle of the sensor is based on the light scattering characteristics of dilute blood and the effect of agglutination on scattering. The device proposed is an optical system based on spectrophotometric methods. The device configuration, and results from several tests with combinations of known blood samples will be presented.