Demonstration and system analysis of the HORNET architecture

The HORNET architecture is a packet-over-wavelength-division-multiplexing ring network that utilizes fast-tunable packet transmitters and wavelength routing to enable it to scale cost-effectively to ultrahigh capacities. In this paper, we present the HORNET architecture and a novel control-channel-based media access control protocol. The survivability of the architecture is demonstrated with an experimental laboratory testbed. Mathematical analysis of the architecture shows that the wavelength routed network can scale to relatively large sizes ranging between 30 and 50 nodes, depending on the component performance. This is true even for arrangements that do not contain high-power optical amplifiers in every node.     

The alpha risk of the Taguchi method for STB type quality characteristic

One of the basic assumptions of experimental design is that the error variances are equal for all treatment combinations. On the contrary, one of the basic assumptions of Taguchi’s parameter design is that the error variances are not equal for the treatment combinations. Thus, the significant parameter levels are found by maximising the signal-to-noise ratio of the quality characteristic. In the analysis of variance (ANOVA) of the signal-to-noise ratio, the combination of column effects to better estimate error variance is referred to as pooling. Taguchi has suggested the strategy of “pooling-up”. When using the pooling-up strategy, there will be the tendency to make the alpha mistake more often. If the assumption of the former is true, then there is an alpha risk that judges some factor being significant when in fact it is not. The purpose of this paper is to investigate the alpha risk of the Taguchi method for the smaller-the-better (STB) type problem by simulation. The results show that the alpha risk is very high for several orthogonal arrays.     

Disulfide-linked head-to-head multimerization in the mechanism of ion channel clustering by PSD-95

The PSD-95/SAP90 family of PDZ-containing proteins is directly involved in the clustering of specific ion channels at synapses. We report that channel clustering depends on a conserved N-terminal domain of PSD-95 that mediates multimerization and disulfide linkage of PSD-95 protomers. This N-terminal multimerization domain confers channel clustering activity on a single PDZ domain. Thus, channel clustering depends on aggregation of PDZ domains achieved by head-to-head multimerization of PSD-95, rather than by concatenation of PDZ domains in PSD-95 monomers. This mechanism predicts that PSD-95 can organize heterogeneous membrane protein clusters via differential binding specificities of its three PDZ domains. PSD-95 and its relative chapsyn-110 exist as disulfide-linked complexes in rat brain, consistent with head-to-head multimerization of these proteins in vivo.     

Indwelling device-related bacteremia caused by serum-susceptible Campylobacter coli

Two isolates of serum-susceptible Campylobacter coli were recovered in a 7-day interval from blood from a patient with hepatocellular carcinoma and liver cirrhosis whose peritoneal-caval (Denver's) shunt malfunctioned. Identical random amplified polymorphic DNA fingerprints, cellular fatty acid chromatograms, and antibiograms of the two isolates indicate that C. coli has the ability to cause catheter-related bacteremia following its colonization of the catheter.     

Inverse-narrow-width effects and small-geometry MOSFET thresholdvoltage model

An analytical threshold voltage model is developed based on the results from a three-dimensional MOSFET simulator, called MICROMOS. The model is derived by solving Poisson's equation analytically and is used to predict the threshold voltage of MOSFETs with fully recessed oxide isolation (the trench structure). Coupling was observed between the short-channel effect and the inverse-narrow-width effect. The coupling results from the mutual modulation of the depletion depth and is used to extend the analytical inverse narrow-width model to small-geometry devices. The model is compared with experimental data obtained from the literature as well as with the three-dimensional simulator. Satisfactory agreement for channel length down to 1.5 μm and channel widths down to 1 μm has been obtained     

Clinical and microbiological characteristics of Flavobacterium indologenes infections associated with indwelling devices

Clinical infections caused by Flavobacterium indologenes have never been documented. Thirteen isolates derived from seven patients with indwelling device-associated F. indologenes infections were identified from 1 April through 30 November 1995. The antimicrobial susceptibilities to 20 antimicrobial agents of the isolates, the cellular fatty acid chromatograms for the isolates, and the random amplified polymorphic DNA (RAPD) patterns generated by arbitrarily primed PCR of the isolates were studied. The antibiotypes and RAPD patterns differed among the isolates recovered from different patients. However, both antibiotypes and RAPD patterns were identical among the five isolates from one patient with multiple episodes of central venous catheter-associated bacteremia within a 1.5-month period and between the two isolates from another patient suffering from two episodes of catheter-related bacteriuria at an interval of 14 days. It is documented that the recurrent infections in each of these two patients were caused by a single F. indologenes clone, respectively. Identical antibiotypes and RAPD patterns were also demonstrated between two isolates from a patient with ventilator-associated pneumonia, one recovered from an endotracheal aspirate and the other derived from a blood specimen 10 days later, indicating the invasive nature of F. indologenes. Two cellular fatty acid chromatograms were identified among these isolates. All of the isolates showed in vitro resistance to cephalothin, cefotaxime, ceftriaxone, moxalactam, aztreonam, aminoglycosides, erythromycin, clindamycin, vancomycin, and teicoplanin. F. indologenes should be included as an etiologic agent of infections associated with the use of indwelling devices.     

BOD (Bcl-2-related ovarian death gene) is an ovarian BH3 domain-containing proapoptotic Bcl-2 protein capable of dimerization with diverse antiapoptotic Bcl-2 members

Using the yeast two-hybrid protein-protein interaction system to search for genes capable of forming dimers with the antiapoptotic protein Mcl-1, we have isolated BOD (Bcl-2-related ovarian death agonist) from an ovarian fusion cDNA library. The three variants of BOD (long, medium, and short) have an open reading frame of 196, 110, and 93 amino acids, respectively; all of them contain a consensus Bcl-2 homology 3 (BH3) domain but lack other BH domains found in channel-forming Bcl-2 family proteins. In the yeast cell assay, BOD interacts with diverse antiapoptotic Bcl-2 proteins [Mcl-1, Bcl-2, Bcl-xL, Bcl-w, Bfl-1, and Epstein-Barr virus (EBV) BHRF-1] but not with different proapoptotic Bcl-2 proteins (BAD, Bak, Bok, and Bax). After overexpression in mammalian Chinese hamster ovary (CHO) cells, BOD induces apoptosis that can be prevented by the baculoviral caspase inhibitor P35. The cell-killing activity of BOD is also antagonized in cells cotransfected with the antiapoptotic Bcl-w protein, which showed high affinity for BOD in the two-hybrid assay. Furthermore, mutagenesis studies showed that BOD mutants with alterations in the BH3 domain lose cell-killing ability, suggesting that the BH3 domain is important for the mediation of cell killing by BOD. BOD mRNA is ubiquitously expressed in ovary and multiple other tissues. The BOD gene is also conserved in diverse mammalian species. Identification of BOD expands the group of proapoptotic Bcl-2 proteins that only contains the BH3 domain and allows future elucidation of the intracellular mechanism for apoptosis regulation in ovary and other tissues.