We have developed a model system in Caenorhabditis elegans to perform genetic and molecular analysis of peptidergic neurotransmission using green fluorescent protein (GFP)-tagged IDA-1. IDA-1 represents the nematode ortholog of the transmembrane proteins ICA512 and phogrin that are localized to dense core secretory vesicles (DCVs) of mammalian neuroendocrine tissues. IDA-1::GFP was expressed in a small subset of neurons and present in both axonal and dendritic extensions, where it was localized to small mobile vesicular elements that at the ultrastructural level corresponded to 50 nm electron-dense objects in the neuronal processes. The post-translational processing of IDA-1::GFP in transgenic worms was dependent on the neuropeptide proprotein convertase EGL-3, indicating that the protein was efficiently targeted to the peptidergic secretory pathway. Time-lapse epifluorescence microscopy of IDA-1::GFP revealed that DCVs moved in a saltatory and bidirectional manner. DCV velocity profiles exhibited multiple distinct peaks, suggesting the participation of multiple molecular motors with distinct properties. Differences between velocity profiles for axonal and dendritic processes furthermore suggested a polarized distribution of the molecular transport machinery. Study of a number of candidate mutants identified the kinesin UNC-104 (KIF1A) as the microtubule motor that is specifically responsible for anterograde axonal transport of DCVs at velocities of 1.6 mm/sÀ2.7 mm/s.

Parallel optical memories have been proposed to meet high speed, high capacity storage requirements for input/output intensive computer applications. This technology offers the capability for storage and retrieval of optical data in twodimensional pages resulting in high capacity and high throughput. Current raw bit error rates ofexperimental systems fall significantly short ofthe minimum industry requirement of 1012 for binary data. Thus, error control techniques for two-dimensional data are necessary in order for such memories to be commercially feasible. In this paper, we discuss severalerror control schemes suitable forpage-oriented data and specifically applicable to page-oriented optical memories. To better interface between the memory and the electronic host computer, we propose "smart" photodetector array devices in which the input is optical but the output is electronic. These arrays receive the optical signal from the memory and convert it to electronic data. Utilizing the speed of VLSI technology, the arrays perform fast paralleldecoding and data correction, thereby providing an efficient optoelectronic interface between the memory and the electronic computer.

Page-oriented optical memories combine high capacity with massive data-transfer rates and could provide the next generation of secondary storage systems. Several technological barriers need to be overcome before these memories become commercial products. One is the need for efficient interfaces to electronic computers. To assist in the analysis of all issues pertaining to such an interface, we have developed the optical storage interactive simulator (oasis), an interactive software tool. oasis can simulate data-encoding schemes, noise sources present in a particular memory system, and data-recovery mechanisms. Bit-error rates and other useful statistics can also be measured. The user has the option of studying the effects of individual error sources to the system output or of applying several of them in any order dictated by the system architecture. This multifaceted analysis will assist the user in evaluating the effectiveness of a particular error-correcting code and choosing the right one for the system.

Page-oriented optical memories (POMs) have the potential to become very high capacity and throughput data storage devices. During readout, data stored in a POM system implemented, e.g. as a volume holographic memory [1,2], may be distorted by a combination of noise sources present in the system, such as the gaussian profile of the beam, misalignment of the various optical components, blurring due to low pass filtering, scattering, and others. Such effects may substantially degrade the quality of the retrieved data and cause a number of bits to be incorrectly identified as 0 or 1 (assuming binary data). Thus, an error correcting scheme is needed to improve the reliability of the system.