From conceptual models to schemata: An object-process-based data warehouse construction method

Data warehouse modeling is a complex task, which involves knowledge of business processes of the domain of discourse, understanding the structural and behavioral system's conceptual model, and familiarity with data warehouse technologies. The suitability of current data warehouse modeling methods for large-scale systems is questionable, as they require multiple manual actions to discover measures and relevant dimensional entities and they tend to disregard the system's dynamic aspects. We present an Object-process-based Data Warehouse Construction (ODWC) method that overcomes these limitations of existing methods by utilizing the operational system conceptual model to construct a corresponding data warehouse schema. We specify the ODWC method, apply it on a case study, evaluate it, and compare it to existing methods.     

Learning Automata from Ordered Examples

Connectionist learning models have had considerable empirical success, but it is hard to characterize exactly what they learn. The learning of finite-state languages (FSL) from example strings is a domain which has been extensively studied and might provide an opportunity to help understand connectionist learning. A major problem is that traditional FSL learning assumes the storage of all examples and thus violates connectionist principles. This paper presents a provably correct algorithm for inferring any minimum-state deterministic finite-state automata (FSA) from a complete ordered sample using limited total storage and without storing example strings. The algorithm is an iterative strategy that uses at each stage a current encoding of the data considered so far, and one single sample string. One of the crucial advantages of our algorithm is that the total amount of space used in the course of learning for encoding any finite prefix of the sample is polynomial in the size of the inferred minimum state deterministic FSA. The algorithm is also relatively efficient in time and has been implemented. More importantly, there is a connectionist version of the algorithm that preserves these properties. The connectionist version requires much more structure than the usual models and has been implemented using the Rochester Connectionist Simulator. We also show that no machine with finite working storage can iteratively identify the FSL from arbitrary presentations.     

Antilipid a monoclonal antibody HA-1A: immune complex clearance of endotoxin reduces TNF-alpha, IL-1 beta and IL-6 production

When the projecting point of saphenous nerve in second somatosensory cortex (S II) of cat was stimulated, the evoked potentials elicited by C-fiber inputs of saphenous nerve recorded in the primary somatosensory cortex (C-CEP) might be either inhibited or facilited according to whether the superficial and/or the deeper layer of the cortex was stimulated. The inhibition was expressed as a decrease of amplitude and prolongation of latency of C-CEP; while the facilitation, as an increase of amplitude and duration of C-CEP. When the superfaicial layer of S II was stimulated by weaker current, both inhibitory and facilitatory effects could be observed, but only inhibitory effect was observed, when the deep layer was stimulated. With the same intensity of stimulation, inhibitory effect was more pronounced when the deep layer rather than the superficial layer was stimulated. It is suggested that S II may play a role in the modulation of C-CEP of S I.     

Optimal techniques for arterial gene transfer

Cardiovascular gene therapy is becoming a clinical reality due to improved vectors, delivery systems and careful experimental validation studies. Nearly all cardiovascular diseases are amenable to gene therapy, but the optimal combination of vector, delivery system and therapeutic gene is likely to be unique to each application. Currently, the most efficient vectors available are replication-defective adenoviral vectors, but transgene expression is limited in time due to a strong immune response. Conversely, non-viral vectors or plasmid DNA may be used safely but have very limited efficiency. Percutaneous, catheter-based delivery is feasible for most applications. The ultimate issues that will decide of the future of gene therapy are safety of the transfer and delivery techniques as well as cost/effectiveness comparisons with alternative therapies, including local delivery of drugs, proteins and/or mechanical devices.     

Structural Determinants of the Sign of the Pyroelectric Effect in Quasi‐Amorphous SrTiO3 Films

The magnitude and direction of the permanent electric polarization in the non‐crystalline, polar phase (termed quasi‐amorphous) of SrTiO₃ in Si\SiO₂\Me\SrTiO₃\Me, (Me = Cr or W), Si\SrRuO₃\SrTiO₃, and Si\SrTiO₃ layered structures were investigated. Three potential sources of the polarization which appears after the material is pulled through a temperature gradient were considered: a) contact potential difference; b) a flexoelectric effect due to a strain gradient caused by substrate curvature; and c) a flexoelectric effect due to the thermally induced strain gradient that develops while pulling through the steep temperature gradient. Measurements show that options a) and b) can be eliminated from consideration. In most cases studied in this (Si\SrTiO₃, Si\SiO₂\Me\SrTiO₃\Me, M = Cr or W) and previous works (Si\BaTiO₃, Si\BaZrO₃), the top surface of the quasi‐amorphous phase acquires a negative charge upon heating. However, in Si\SrRuO₃\SrTiO₃ structures the top surface acquires a positive charge upon heating. On the basis of the difference in the measured expansion of the upper and lower surfaces of the SrTiO₃ layer in the presence and absence of SrRuO₃, we contend that the magnitude and direction of the pyroelectric effect are determined by the out‐of‐plane gradient of the in‐plane strain in the SrTiO₃ layer while pulling through the temperature gradient.     

Influence of order-disorder on the vibron excitations of H2 and D2 in ortho-para mixed crystals

We present a detailed experimental and theoretical study of the vibrons in ortho-para mixed crystals of solid hydrogen and deuterium at ambient and high pressure. Experimental results were obtained at ambient pressure and T=6–7 K (e.g., for hydrogen samples having ortho fractions of 19–62%) using high-resolution Fabry-Perot techniques, and at high pressure and T=77 K (e.g., hydrogen 50–50% ortho-para samples) using dispersive spectrographic techniques with diamond-anvil cells. The numerical calculations are based on the James and Van Kranendonk theory, and were performed by exactly diagonalizing the Hamiltonian for a large supercell of randomly placed molecular “species” on a crystalline lattice. Overall, excellent agreement between theory and experiment is obtained. The calculations show that disorder leads to Anderson localized vibrons for many of the pressures and concentrations studied experimentally and that a substantial portion of the Raman intensity is derived from these localized vibrons. We also calculate the species characteristics of the individual Raman peaks, the results of which suggest an explanation for the previously noted disagreement between experimental high-pressure results and the predictions of the van Kranendonk theory. Specifically, our analysis indicates that the higher frequency peak is associated with anisotropic scattering arising from partial alignment of J=1 angular momenta with respect to the crystallographic axes. Finally, our calculations show that the observed doublet structure in the lower frequency Raman peak for deuterium at low para (J=1) concentrations is well represented by added (para-molecule) diagonal terms in the van Kranendonk Hamiltonian that are plausibly associated with electric quadrupole-quadrupole interactions.     

Platelet-reinforced polymer matrix composites by combined gel-casting and hot-pressing. Part II: Thermoplastic polyurethane matrix composites

A combined gel-casting and hot-pressing method was used to fabricate platelet-reinforced polymer matrix composites. Submicrometer thin alumina platelets were dispersed in a highly diluted polymer solution. A thermoplastic polyurethane elastomer was used as matrix for its high elasticity and excellent adhesion to the platelets. After dissolution of the polymer and casting, quick evaporation of the solvent triggered the formation of a polymer gel trapping the platelets in their well dispersed positions. The polymer–platelet gel densified during drying and the platelets were oriented horizontally due to the capillary forces and the large decrease in the thickness of the gel. The dried composites were hot-pressed to further improve the platelet orientation along the shear flow and close potential pores in the polymer. While the ultimate tensile strength of the composites gradually decreased with increasing platelet volume fractions, the increase in the elastic modulus and the stress necessary to deform the composite 10% was more than 100 and 18 times higher than the respective values of the pure polymer. The use of alumina platelets with an aspect ratio below the critical value allowed for the ductile platelet pull-out fracture mode. Since the polymer had to deform more to achieve identical deformation of the composite at higher platelet volume fraction, the strain at rupture steadily decreased. The incorporation of voids towards high platelet concentrations and the thereby triggered crack initiation and growth during straining lead to an additional decrease in the elasticity of composites with increasing platelet volume fractions. However, the extremely high extensibility of a polymer matrix allowed for the fabrication of composites that still deformed up to 162 ± 19% at platelet volume fractions as high as 0.33. When compared to other platelet-reinforced elastomers, the achieved platelet volume fraction is much higher and the relative increase in elastic modulus and stress at low strains is therefore much larger at the expense of a decrease in the strain at rupture. The fabrication method and designing principles employed in this study are transferable to other types of polymers and platelets and potentially allow the creation of new composites with tailored properties.     

Pim1 cooperates with E2a-Pbx1 to facilitate the progression of thymic lymphomas in transgenic mice

Mice transgenic for the leukemia oncogene E2A-PBX1 invariably develop lethal, high-grade T-cell lymphomas by 5 months of age. In this study, retroviral insertional mutagenesis was employed to identify oncogenes that cooperate with the E2A-PBX1 transgene in lymphomagenesis. Neonatal retroviral infection substantially reduced length of survival due to accelerated development of lymphomas (81 versus 130 days). The Pim1 gene was targeted by retroviral insertions in 48% of accelerated lymphomas whereas less than 5% contained activated c-Myc and none contained activated Pim2. However, Pim1 DNA rearrangements were frequently sub-stoichiometric and not present at all sites of involvement in an otherwise monoclonal lymphoma indicating that Pim1 activation occurred late in the course of lymphomagenesis. Tumor subpopulations containing activated Pim1 alleles displayed a substantial growth advantage over Pim1 negative cells following serial transfer to secondary, syngeneic recipients. Cooperative interactions were observed in intercrossed Pim1 and E2A-PBX1 transgenic mice in which all double transgenic progeny developed lethal, diffuse T lineage lymphomas by 3 months of age, whereas only 13% of E2A-PBX1 and none of Pim1 single transgenic intercross progeny developed lymphomas by 1 year. Tumors from double transgenic mice were monoclonal providing evidence that additional genetic events were required for transformation. Therefore, Pim1 and E2a-Pbx1 cooperate in T lineage lymphomagenesis but they are not sufficient and the role of Pim1 is more likely to be associated with tumor progression.     

1alpha,25-dihydroxyvitamin D3 actions in LNCaP human prostate cancer cells are androgen-dependent

We and others have recently shown that 1alpha,25-dihydroxyvitamin D3 [1,25-(OH)2D3] significantly inhibits cell proliferation and increases secretion of prostate-specific antigen (PSA) in LNCaP cells, an androgen-responsive human prostate cancer cell line. The present study was designed to investigate the possible interactions between 1,25-(OH)2D3 and androgens in the regulation of LNCaP cellular function. LNCaP cell growth was dose-dependently inhibited by 1,25-(OH)2D3 (60% inhibition at 10 nM) when cells were cultured in medium supplemented with FBS (FBS medium). 1,25-(OH)2D3-treated cells showed a 5-fold increase in PSA secretion, similar to the increase seen in dihydrotestosterone (DHT)-treated cells. In combination, 1,25-(OH)2D3 and DHT synergistically enhanced PSA secretion 22-fold. This synergistic effect was even greater when cells were cultured in medium supplemented with charcoal-stripped serum (CSS medium), where endogenous steroids are substantially depleted. Under these conditions, 1,25-(OH)2D3 and DHT together stimulated PSA secretion up to 50-fold over the untreated control. Radioligand binding assays and Western blot analyses showed that the androgen receptor (AR) content was increased significantly by 1,25-(OH)2D3 at 48 h. Furthermore, the steady-state mRNA level of AR was up-regulated approximately 2-fold by 1,25-(OH)2D3 at 24 h. When cells were grown in CSS medium, 1,25-(OH)2D3 alone no longer inhibited cell growth or induced PSA secretion. Titration experiments revealed that the addition of DHT at 1 nM to the medium restored the antiproliferative activity of 1,25-(OH)2D3. Conversely, an antiandrogen, Casodex, completely blocked 1,25-(OH)2D3 antiproliferative and PSA stimulation activities when cells were cultured in FBS medium. In conclusion, these results demonstrate that the antiproliferative and PSA induction activities of 1,25-(OH)2D3 in LNCaP cells are dependent upon androgen action and that AR up-regulation by 1,25-(OH)2D3 likely contributes to the synergistic actions of 1,25-(OH)2D3 and DHT in these cells.