Deep learning models for inflation forecasting

We propose a hybrid deep learning model that merges Variational Autoencoders and Convolutional LSTM Networks (VAE-ConvLSTM) to forecast inflation. Using a public macroeconomic database that comprises 134 monthly US time series from January 1978 to December 2019, the proposed model is compared against several popular econometric and machine learning benchmarks, including Ridge regression, LASSO regression, Random Forests, Bayesian methods, VECM, and multilayer perceptron. We find that VAE-ConvLSTM outperforms the competing models in terms of consistency and out-of-sample performance. The robustness of such conclusion is ensured via cross-validation and Monte-Carlo simulations using different training, validation, and test samples. Our results suggest that macroeconomic forecasting could take advantage of deep learning models when tackling nonlinearities and nonstationarity, potentially delivering superior performance in comparison to traditional econometric approaches based on linear, stationary models.     

Cytotoxic Mechanism of Sphaerodactylomelol,an Uncommon Bromoditerpene Isolated from Sphaerococcus coronopifolius

Marine natural products have exhibited uncommon chemical structures with relevant antitumor properties highlighting their potential to inspire the development of new anticancer agents. The goal of this work was to study the antitumor activities of the brominated diterpene sphaerodactylomelol, a rare example of the dactylomelane family. Cytotoxicity (10–100 µM; 24 h) was evaluated on tumor cells (A549, CACO-2, HCT-15, MCF-7, NCI-H226, PC-3, SH-SY5Y, SK-ML-28) and the effects estimated by MTT assay. Hydrogen peroxide (H₂O₂) levels and apoptosis biomarkers (membrane translocation of phosphatidylserine, depolarization of mitochondrial membrane potential, Caspase-9 activity, and DNA condensation and/or fragmentation) were studied in the breast adenocarcinoma cellular model (MCF-7) and its genotoxicity on mouse fibroblasts (L929). Sphaerodactylomelol displayed an IC₅₀ range between 33.04 and 89.41 µM without selective activity for a specific tumor tissue. The cells’ viability decrease was accompanied by an increase on H₂O₂ production, a depolarization of mitochondrial membrane potential and an increase of Caspase-9 activity and DNA fragmentation. However, the DNA damage studies in L929 non-malignant cell line suggested that this compound is not genotoxic for normal fibroblasts. Overall, the results suggest that the cytotoxicity of sphaerodactylomelol seems to be mediated by an increase of H₂O₂ levels and downstream apoptosis.     

Assessment of hydrophobic interactions and their contributions through the analysis of the methane dimer

Hydrophobic Interactions (HIs) are important in many phenomena of molecular recognition in chemistry and biology. Still, the relevance of HIs is sometimes difficult to evaluate particularly in large systems and intramolecular interactions. We put forward a method to estimate the magnitude and the different contributions of a given HI of the C···C, H? C···H, and H···H type through (i) the analysis of the electron density in the intermolecular region for eleven relative orientations of the methane dimer and (ii) the subsequent decomposition of the corresponding interaction energy in physically significant contributions using Symmetry Adapted Perturbation Theory (SAPT). Strong correlations were found between the topological properties of calculated at intermolecular bond critical points and plus its different contributions with the C···C distance of the considered orientations of (CH₄)₂. These correlations were used to construct Mollier‐like diagrams of and its components as a function of the separation between two carbons and the orientation of the groups bonded to these atoms. The ethane dimer and tert‐butylcyclohexane are used as representative examples of this new approach. Overall, we anticipate that this new method might prove useful in the study of both intramolecular and intermolecular HIs particularly of those within large systems wherein SAPT or electronic structure calculations are computationally expensive or even prohibitive. © 2014 Wiley Periodicals, Inc.     

A disposable laser print-cut-laminate polyester microchip for multiplexed PCR via infra-red-mediated thermal control

Infrared (IR)-mediated thermal cycling system, a method proven to be a effective for sub-μL scale polymerase chain reaction (PCR) on microchips, has been integrated with DNA extraction and separation on a glass microchip in a fully integrated micro Total Analysis System by Easley et al., in 2006. IR-PCR has been demonstrated on both glass and PMMA microdevices where the fabrication (bonding) is not trivial. Polyester-toner (PeT) microfluidic devices have significant potential as cost-effective, disposable microdevices as a result of the ease of fabrication (∼$0.25 USD and <10 min per device) and availability of commercial substrates. For the first time, we demonstrate here the thermal cycling in PeT microchips on the IR-PCR system. Undesirable IR absorption by the black-toner bonding layer was eliminated with a spatial filter in the form of an aluminum foil mask. The solution heating rate for a black PeT microchip using a tungsten lamp was 10.1 ± 0.7 °C s⁻¹ with a cooling rate of roughly −12 ± 0.9 °C s⁻¹ assisted by forced air cooling. Dynamic surface passivation strategies allowed the successful amplification of a 520 bp fragment of the λ-phage genome (in 11 min) and a 1500 bp region of Azospirillum brasilense. Using a centrosymmetric chamber configuration in a multichamber PeT microchip, homogenous temperature distribution over all chambers was achieved with inter-chamber temperature differences at annealing, extension and denaturing steps of less than ±2 °C. The effectiveness of the multichamber system was demonstrated with the simultaneous amplification of a 390 bp amplicon of human β-globin gene in five PeT PCR microchambers. The relative PCR amplification efficiency with a human β-globin DNA fragment ranged from 70% to 90%, in comparison to conventional thermal cyclers, with an inter-chamber standard deviation of ∼10%. Development of PeT microchips for IR-PCR has the potential to provide rapid, low-volume amplification while also integrating PCR with extraction upstream and separation/detection downstream.     

High Final Energy of Low‐Level Gallium Arsenide Laser Therapy Enhances Skeletal Muscle Recovery without a Positive Effect on Collagen Remodeling

The aim of this study was to evaluate the effects of a Gallium Arsenide (GaAs) laser, using a high final energy of 4.8 J, during muscle regeneration after cryoinjury. Thirty Wistar rats were divided into three groups: Control (C, n = 10); Injured (I, n = 10) and Injured and laser treated (Injured/LLLT, n = 10). The cryoinjury was induced in the central region of the tibialis anterior muscle (TA). The applications of the laser (904 nm, 50 mW average power) were initiated 24 h after injury, at energy density of 69 J cm^?1 for 48 s, for 5 days, to two points of the lesion. Twenty‐four hours after the final application, the TA muscle was removed and frozen in liquid nitrogen to assess the general muscle morphology and the gene expression of TNF‐α, TGF‐β, MyoD, and Myogenin. The Injured/LLLT group presented a higher number of regenerating fibers and fewer degenerating fibers (P < 0.05) without changes in the collagen remodeling. In addition, the Injured/LLLT group presented a significant decrease in the expression of TNF‐α and myogenin compared to the injured group (P < 0.05). The results suggest that the GaAs laser, using a high final energy after cryoinjury, promotes muscle recovery without changing the collagen remodeling in the muscle extracellular matrix.     

Carbon nanotube field effect transistor biosensor for the detection of toxins in seawater

Disposable field effect transistors (FET) biosensors (bio-FET) based on carbon nanotubes were fabricated for detection of domoic acid (DA), which belongs to the group of biotoxins associated with the amnesic shellfish poisoning. The analytical results obtained with the bio-FET were compared with those obtained with a traditional methodology (enzyme-linked immunosorbent assay) in order to validate the bio-FET for DA detection. Standard solutions of DA with concentrations between 10 and 500 ng L^?1 were tested in order to construct the calibration curve, where five bio-FET were used for reproducibility estimation and two analytical measurements were performed for each bio-FET for repeatability estimation. Ten spiked artificial seawater samples were used to validate the bio-FET. The obtained reproducibility (0.52–1.43%), repeatability (0.57–1.27%), limit of detection (10 ng L^?1) and recovery range (92.3–100.3%) reveal an adequate analytical performance of the bio-FET for the detection of DA in environmental samples such as seawater samples.     

Effect of Furfural on <Emphasis Type="Italic">Saccharomyces carlsbergensis</Emphasis> Growth,Physiology and Ethanol Production

This work described the effect of furfural, a product resulting from the lignocellulosic material pretreatment, on Saccharomyces carlsbergensis growth and ethanol production. Flow cytometry was used to evaluate the yeast membrane potential, membrane integrity, reactive oxygen species production and lipid content. Above 0.3 g/L of furfural, a progressive decrease in the maximal specific growth rate was observed, reaching 53% of the value obtained in the absence of toxic when the cells were grown in the presence of 4 g/L of furfural. In general, the yeast biomass concentration and yield were less affected by the furfural presence than the specific growth rate, and a maximum reduction of 25% was observed for the assay at 4 g/L. The ethanol production was even less affected by the furfural presence than the yeast growth. At 4 g/L of furfural, the maximum ethanol concentration was reduced by only 10% relatively to the maximum ethanol concentration observed in the absence of toxic. At 5 g/L of furfural, the yeast cells were barely able to keep metabolic functions and produced a final ethanol concentration of 0.87 g/L although growth was undetectable. S. carlsbergensis membrane potential was affected by the furfural presence, concomitantly with the ethanol production. However, at 4 g/L, most of the yeast cells (90%) displayed the cytoplasmic membrane depolarized. The proportion of cells with increasing reactive oxygen species (ROS) production levels increased for the experiments at 0–4 g/L. For the experiment at 4.5 g/L of furfural, ROS production was observed for only 11% of the yeast cells. The yeast lipid content was also severely affected by the furfural presence. Both polar and neutral lipids decreased in the presence of furfural, and this reduction was more notorious during the stationary phase.     

Self-Assembled Functionalized Graphene Nanoribbons from Carbon Nanotubes

Graphene nanoribbons (GNR) were generated in ethanol solution by unzipping pyrrolidine-functionalized carbon nanotubes under mild conditions. Evaporation of the solvent resulted in regular few-layer stacks of graphene nanoribbons observed by transmission electron microscopy (TEM) and X-ray diffraction. The experimental interlayer distance (0.49–0.56 nm) was confirmed by computer modelling (0.51 nm). Computer modelling showed that the large interlayer spacing (compared with graphite) is due to the presence of the functional groups and depends on their concentration. Stacked nanoribbons were observed to redissolve upon solvent addition. This preparation method could allow the fine-tuning of the interlayer distances by controlling the number and/or the nature of the chemical groups in between the graphene layers.     

Numerical Godeaux surfaces with an involution

Minimal algebraic surfaces of general type with the smallest possible invariants have geometric genus zero and and are usually called numerical Godeaux surfaces. Although they have been studied by several authors, their complete classification is not known.

In this paper we classify numerical Godeaux surfaces with an involution, i.e. an automorphism of order 2. We prove that they are birationally equivalent either to double covers of Enriques surfaces or to double planes of two different types: the branch curve either has degree 10 and suitable singularities, originally suggested by Campedelli, or is the union of two lines and a curve of degree 12 with certain singularities. The latter type of double planes are degenerations of examples described by Du Val, and their existence was previously unknown; we show some examples of this new type, also computing their torsion group.

     

Vector-tensor and vector-vector decay amplitude analysis of B0-->phiK*0

We perform an amplitude analysis of the decays B(0)-->phiK*(2)(1430)(0), phiK*(892)(0), and phi(Kpi)(0)(S-wave) with a sample of about 384x10(6) BB[over ] pairs recorded with the BABAR detector. The fractions of longitudinal polarization f(L) of the vector-tensor and vector-vector decay modes are measured to be 0.853(-0.069+0.061)+/-0.036 and 0.506+/-0.040+/-0.015, respectively. Overall, twelve parameters are measured for the vector-vector decay and seven parameters for the vector-tensor decay, including the branching fractions and parameters sensitive to CP violation.