A high-performance image processor for the near infrared. A modular, low-power, microprocessor based image processor was developed for use in a variety of applications. It incorporates an optimized data path architecture for real-time processing of the signal, a large amount of on-chip RAM for buffering of image data, a programmable memory controller for use with both memory and peripheral modules, as well as peripherals for image I/O, video sync generation and other applications. The use of a dual ported frame buffer enables the image processor to be used in medical imaging systems as well as in other applications where the data processing load is primarily based on digital image acquisition. The processor's speed and data throughput are adequate for real-time imaging at 10 Hz or greater for a variety of medical imaging modalities. A unique approach to image processing is presented in this paper which is based on a dual-ported frame buffer for acquiring and storing the raw image data. This approach supports frame-by-frame image processing based on the output from the image sensor. This dual-ported frame buffer based image processor architecture incorporates image compression and is therefore ideal for a wide range of imaging applications. The ability to process images in frame-by-frame, rather than line-by-line, mode provides the advantage of flexibility in the data organization, image processing, and compression. The unique structure of the frame buffer allows it to support high performance as well as a programmable data access protocol. The data access protocol offers the flexibility required to make the processor compatible with a variety of sensor modalities and to permit a wide range of customized image processing algorithms. The hardware designed for implementing the frame buffer for this application is based on a fully pipelined structure. An optimum usage of the processing resources makes this architecture ideally suited for low-power, high performance image processing applications. A high-performance image processor for medical imaging applications has been developed and implemented as a product. The architecture and implementation of this product is described in this paper. This device can be tailored to image processing requirements by a software package with the assistance of a programmable digital interface and/or an external microprocessor. Linking in vitro gene expression to the phenotypic properties of malignant melanoma cell lines: using a novel, high density cDNA microarray Authors: Mark M. Trusheim, Jing Sun, Xiang Wang, Anthony J. Ching. Institutions: University of Nebraska Medical Center. The development and ultimate therapeutic application of appropriate model systems for cancer therapy and studying melanoma progression is dependent on the availability of melanoma cell lines with defined and tractable genotypes. We describe a novel high density cDNA microarray comprised of cDNAs derived from a well defined set of melanoma tumor-associated genes that we have used to study expression of these genes in a panel of melanoma cell lines. The high density microarray is comprised of approximately 60,000 cDNA clones arranged in duplicate, each spotted in quadruplicate on epoxy-coated glass slides. Efficient methods are provided for the production of fluorescently labeled cDNA from total RNA and for the use of this high density array in expression profiling experiments. Examples of how this high density cDNA microarray can be used to study melanoma cell biology are presented. We demonstrate that this microarray platform provides a highly efficient method for assessing differential gene expression. This high density array is superior to currently available cDNA microarray platforms with respect to the number of unique clones that can be included. This microarray also allows for a more comprehensive survey of gene expression as it utilizes the entire expressed genome rather than a pre-defined set of sequences. Microalgae have gained much interest due to the possibilities of using them as sources of valuable lipids (e.g., triacylglycerides, (TAGs)) that can be converted into biofuels1, 2. In the past decade, genetic engineering of oleaginous microalgae for the production of valuable lipids has emerged as a promising approach for both basic research and commercial production of biofuels and valuable metabolites3, 4. The algal fatty acid (FA) biosynthesis pathway has been extensively engineered in various oleaginous microalgae for the production of long-chain (>20 carbons) polyunsaturated FAs (PUFAs)3-8. These engineered PUFA-overproducing strains could accumulate fatty acids in the cellular lipids up to over 50% of the total cellular lipid content8. However, the production of short-chain (<20 carbons) FAs was barely improved3, 4. This is because, unlike the long-chain FAs, the metabolic pathway(s) for the biosynthesis of short-chain FAs in oleaginous microalgae are not clearly understood. In the last few years, microalgal biosynthesis of short-chain FAs and other valuable metabolites has been investigated using multiple omics technologies, including transcriptomics, metabolomics, proteomics9-12. Metabolomics is a powerful tool for examining the changes in metabolites during specific biological processes that are critical for lipid accumulation13. It has also been used as a tool to examine metabolic variations in engineered strains with different genetic background9, 12, 14. For example, it was shown that the deletion of acyl-ACP thioesterase (FatB) gene in the green alga Chlamydomonas reinhardtii increased total cellular lipids and increased the accumulation of several short-chain FAs10, 15. However, in spite of such successful examples9, 10, 12, 14, the potential impact of different genetic backgrounds on the accumulation of metabolites is still largely unknown. In this study, we report the impact of different genetic backgrounds (such as different fatty acyl-CoA desaturases (FADs) genes) on the accumulation of short-chain FAs (e.g., C8, C10, and C12) in engineered Chlorella sp. strain c-27 (http://chlorella.umh.es/). We demonstrate the effectiveness of metabolic engineering using a combination of multiple “omics” analysis including: metabol