An efficient parallel strategy for the perfect domination problem on distance-hereditary graphs

A graph is distance-hereditary if the distance stays the same between any of two vertices in every connected induced subgraph containing both. Two well-known classes of graphs, trees and cographs, both belong to distance-hereditary graphs. In this paper, we first show that the perfect domination problem can be solved in sequential linear-time on distance-hereditary graphs. By sketching some regular property of the problem, we also show that it can be easily parallelized on distance-hereditary graphs.     

Synthesis of polyurethane elastomers hypercross-linked by partially hydrated polyhydroxylated C60

Utilization of polyhydroxylated C₆₀ (fullerenols) in a condensation reaction with diisocyanated oligo(tetramethylene oxide) led to the successful fabrication of elastomeric poly (urethane-ether) networks. These polymer networks exhibit interesting thermal behavior at low temperatures, improved tensile strength and elongation at ambient temperatures, and enhanced thermal mechanical stability at high temperatures, as compared with those of the parent linear polyurethane analogues; or with the conventional oligo (tetramethylene oxide)-derived polyurethane elastomers cross-linked by trihydroxylated reagents (1,I,1-trismethylol propane) or tetrahydroxylated reagents (pentaerythritol). The presence of a limited quantity of water molecules in the condensation reaction of fullerenols with diisocyanated prepolymers modified the physical properties of the resulting elastomeric products with a notable increase in tensile strength, modulus, and Ts over those of elasotmers prepared under anhydrous conditions. These water molecules contributed effectively to the increase of the number of cross-linking centers during the reaction.     

Precipitation behavior of σ phase in fusion zone of dissimilar stainless steel welds during multi-pass GTAW process

The purpose of this study is to investigate the precipitation characteristics of σ phase in the fusion zone of stainless steel welds at various welding passes during a tungsten are welding (GTAW) process. The morphology, quantity, and chemical composition of the δ-ferrite and σ phase were analyzed using optical microscopy (OM), a ferritscope (FS), a X-ray diffractometer (XRD), scanning electron microscopy (SEM), an electron probe micro-analyzer (EPMA), and a wavelength dispersive spectrometer (WDS), respectively. Massive δ-ferrite was observed in the fusion zone of the first pass welds during welding of dissimilar stainless steels. The σ phase precipitated at the inner δ-ferrite particles and decreased δ-ferrite content during the third pass welding. The σ and δ phases can be stabilized by Si element, which promoted the phase transformation of σ→ϱ+λ₂ in the fusion zone of the third pass welds. It was found that the σ phase was a Fe−Cr−Si intermetallic compound found in the fusion zone of the third pass welds during multi-pass welding.     

Comparison of calcium phosphate cement mixture and pure calcium hydroxide as direct pulp-capping agents

This review reports the different genetic factors that have been identified either as risk factor for Alzheimer's disease (AD) or directly causing the disease. First are reviewed epidemiological data and biological mechanisms about the apoplipoprotein E gene allele epsilon 4 that is a major risk factor for Alzheimer's disease. The second part describes the mutations responsible for early-onset autosomal dominant AD found in three different genes. The gene located on chromosome 21 encodes the amyloid precusor protein (APP). The presenilin 1 and presenilin 2 genes, located on chromosome 14 and 1 respectively, encode not yet known membrane proteins.     

Air-Oxidation Behavior of a Cu60Hf25Ti15 Bulk Metallic Glass at 375–520 °C

The oxidation behavior of a Cu₆₀Hf₂₅Ti₁₅ bulk metallic glass was studied over the temperature range of 375–520 °C in dry air. The oxidation kinetics of the amorphous alloy generally followed the parabolic law at all temperatures, with an oxidation rate increasing with temperature. The oxidation rates of the amorphous alloy were much higher than those of polycrystalline pure-Cu, implying that the additions of Hf and Ti accelerated the oxidation reaction. The composition of the scales formed on the amorphous alloy was strongly temperature-dependent, since they consisted mostly of Cu₄O₃ and CuO with minor amounts of HfO₂ at T ≤ 450 °C, while mostly CuO with minor amounts of HfO₂ and Cu₂TiO₃ were detected at higher temperatures. In addition, nanocrystalline Cu₅₁Hf₁₄ and Cu₃Ti₂ phases were detected on the substrate after oxidation at T ≥ 450 °C, indicating the occurrence of phase transformation.     

The quantum dynamic capacity formula of a quantum channel

The dynamic capacity theorem characterizes the reliable communication rates of a quantum channel when combined with the noiseless resources of classical communication, quantum communication, and entanglement. In prior work, we proved the converse part of this theorem by making contact with many previous results in the quantum Shannon theory literature. In this work, we prove the theorem with an ??ab initio?? approach, using only the most basic tools in the quantum information theorist??s toolkit: the Alicki-Fannes?? inequality, the chain rule for quantum mutual information, elementary properties of quantum entropy, and the quantum data processing inequality. The result is a simplified proof of the theorem that should be more accessible to those unfamiliar with the quantum Shannon theory literature. We also demonstrate that the ??quantum dynamic capacity formula?? characterizes the Pareto optimal trade-off surface for the full dynamic capacity region. Additivity of this formula reduces the computation of the trade-off surface to a tractable, textbook problem in Pareto trade-off analysis, and we prove that its additivity holds for the quantum Hadamard channels and the quantum erasure channel. We then determine exact expressions for and plot the dynamic capacity region of the quantum dephasing channel, an example from the Hadamard class, and the quantum erasure channel.     

Epigallocatechin-3-Gallate-Loaded Liposomes Favor Anti-Inflammation of Microglia Cells and Promote Neuroprotection

Microglia-mediated neuroinflammation is recognized to mainly contribute to the progression of neurodegenerative diseases. Epigallocatechin-3-gallate (EGCG), known as a natural antioxidant in green tea, can inhibit microglia-mediated inflammation and protect neurons but has disadvantages such as high instability and low bioavailability. We developed an EGCG liposomal formulation to improve its bioavailability and evaluated the neuroprotective activity in in vitro and in vivo neuroinflammation models. EGCG-loaded liposomes have been prepared from phosphatidylcholine (PC) or phosphatidylserine (PS) coated with or without vitamin E (VE) by hydration and membrane extrusion method. The anti-inflammatory effect has been evaluated against lipopolysaccharide (LPS)-induced BV-2 microglial cells activation and the inflammation in the substantia nigra of Sprague Dawley rats. In the cellular inflammation model, murine BV-2 microglial cells changed their morphology from normal spheroid to activated spindle shape after 24 h of induction of LPS. In the in vitro free radical 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay, EGCG scavenged 80% of DPPH within 3 min. EGCG-loaded liposomes could be phagocytized by BV-2 cells after 1 h of cell culture from cell uptake experiments. EGCG-loaded liposomes improved the production of BV-2 microglia-derived nitric oxide and TNF-α following LPS. In the in vivo Parkinsonian syndrome rat model, simultaneous intra-nigral injection of EGCG-loaded liposomes attenuated LPS-induced pro-inflammatory cytokines and restored motor impairment. We demonstrated that EGCG-loaded liposomes exert a neuroprotective effect by modulating microglia activation. EGCG extracted from green tea and loaded liposomes could be a valuable candidate for disease-modifying therapy for Parkinson’s disease (PD).     

Reverse osmosis separations of polyethylene glycols in dilute aqueous solutions using porous cellulose acetate membranes

Reverse osmosis separations of eight polyethylene glycol (PEG) solutes in the average molecular weight range of 200 to 6750 in single-solute dilute aqueous solutions have been studied using porous cellulose acetate membranes at the operating pressures of 50, 75, and 100 psig. Diffusivity data for the above PEG solutes have also been obtained from experimental data on intrinsic viscosities. From an analysis of all experimental data, numerical values for the parameters representing the polar (?ΔΔG/RT), steric (δ^*ΣE_s), and nonpolar (ω^*Σs^*) forces governing reverse osmosis separations of PEG solutes have been generated. These numerical values are useful for precise characterization of cellulose acetate membranes for whose specifications sodium chloride is not the appropriate reference solute because of its low or practically negligible separation under reverse osmosis operating conditions. This work also illustrates that solute separation in reverse osmosis can predictably increase or decrease with increase in operating pressure depending on experimental conditions.     

Parallel matrix factorization for recommender systems

Matrix factorization, when the matrix has missing values, has become one of the leading techniques for recommender systems. To handle web-scale datasets with millions of users and billions of ratings, scalability becomes an important issue. Alternating least squares (ALS) and stochastic gradient descent (SGD) are two popular approaches to compute matrix factorization, and there has been a recent flurry of activity to parallelize these algorithms. However, due to the cubic time complexity in the target rank, ALS is not scalable to large-scale datasets. On the other hand, SGD conducts efficient updates but usually suffers from slow convergence that is sensitive to the parameters. Coordinate descent, a classical optimization approach, has been used for many other large-scale problems, but its application to matrix factorization for recommender systems has not been thoroughly explored. In this paper, we show that coordinate descent-based methods have a more efficient update rule compared to ALS and have faster and more stable convergence than SGD. We study different update sequences and propose the CCD++ algorithm, which updates rank-one factors one by one. In addition, CCD++ can be easily parallelized on both multi-core and distributed systems. We empirically show that CCD++ is much faster than ALS and SGD in both settings. As an example, with a synthetic dataset containing 14.6 billion ratings, on a distributed memory cluster with 64 processors, to deliver the desired test RMSE, CCD++ is 49 times faster than SGD and 20 times faster than ALS. When the number of processors is increased to 256, CCD++ takes only 16 s and is still 40 times faster than SGD and 20 times faster than ALS.