Time series modeling and forecasting has fundamental importance to a wide range of applications. While classical time series models dominated the forecasting field for years, their applications have been limited to single time series data at low frequency such as monthly, quarterly or annually. The goal of this paper is to build multi-series time series models to forecast future daily death counts for each county in the state of Pennsylvania. The data used in this paper include JHU daily death counts and confirmed cases and CDC vaccination rates from 1/22/2020 to 1/7/2022 at the county level for Pennsylvania. Both machine learning (Extreme Gradient Boosted Tree "XGBoost") and deep learning (Keras Slim Residual Neural Network Regressor, "Keras") algorithms were explored and time series modeling related steps such as feature engineering, data partition and project setup are discussed in detail. In addition, four metrics were calculated to evaluate the algorithms’ performance. The comparison with a baseline time series model indicated that machine learning and deep learning algorithms did improve forecasting accuracy significantly and Keras has slightly better performance than XGBoost. Finally, the Keras model was utilized to forecast daily death counts for 60 days after 1/7/2022, i.e., 1/8/2022 to 3/8/2022. Based on the model forecasts, daily death counts should gradually ease off by mid-February which has been validated by the subsequent observations. (Abstract)

The eight species of bears world-wide consume a wide variety of diets. Some are specialists with extensive anatomical and physiological adaptations necessary to exploit specific foods or environments [e.g., polar bears (Ursus maritimus), giant pandas (Ailuropoda melanoleuca), and sloth bears (Melursus ursinus)], while the rest are generalists. Even though ursids evolved from a high-protein carnivore, we hypothesized that all have become low-protein macronutrient omnivores. While this dietary strategy has already been described for polar bears and brown bears (Ursus arctos), a recent study on giant pandas suggested their macronutrient selection was that of the ancestral high-protein carnivore. Consumption of diets with inappropriate macronutrient profiles has been associated with increased energy expenditure, ill health, failed reproduction, and premature death. Consequently, we conducted feeding and preference trials with giant pandas and sloth bears, a termite and ant-feeding specialist. Both giant pandas and sloth bears branched off from the ursid lineage a million or more years before polar bears and brown bears. We found that giant pandas are low-protein, high-carbohydrate omnivores, whereas sloth bears are low-protein, high-fat omnivores. The preference for low protein diets apparently occurred early in the evolution of ursids and may have been critical to their world-wide spread.

Abstract Amyotrophic lateral sclerosis (ALS) is an inexorably progressive and degenerative disorder of motor neurons with no currently-known cure. Studies to determine the mechanism of neurotoxicity and the impact of ALS-linked mutations (SOD1, FUS, TARDP, C9ORF72, PFN1, TUBA4A and others) have greatly expanded our knowledge of ALS disease mechanisms and have helped to identify potential targets for ALS therapy. Cellular pathologies (e.g., aggregation of mutant forms of SOD1, TDP43, FUS, Ubiqulin2, PFN1, and C9ORF72), mitochondrial dysfunction, neuro­inflammation, and oxidative damage are major pathways implicated in ALS. Nevertheless, the selective vulnerability of motor neurons remains unexplained. The importance of tubulins for long-axon infrastructure, and the special morphology and function of motor neurons, underscore the central role of the cytoskeleton. The recent linkage of mutations to the tubulin α chain, TUBA4A, to familial and sporadic cases of ALS provides a new investigative opportunity to shed light on both mechanisms of ALS and the vulnerability of motor neurons. In the current study we investigate TUBA4A, a structural microtubule protein with mutations causal to familial ALS, using molecular-dynamic (MD) modeling of protein structure to predict the effects of each mutation and its overall impact on GTP binding, chain stability, tubulin assembly, and aggregation propensity. These studies predict that each of the reported mutations will cause notable structural changes to the TUBA4A (α chain) tertiary protein structure, adversely affecting its physical properties and function. Molecular docking and MD simulations indicate certain α chain mutations (e.g. K430N, R215C, and W407X) will cause structural deviations that impair GTP binding, and may prevent tubulin polymerization. Furthermore, several mutations (including R320C and K430N) confer a significant increase in predicted aggregation propensity of TUBA4A mutants relative to wild-type. Taken together, these in silico modeling studies revealed structural perturbations and disruption of GTP binding, culminating in failure to form the tubulin heterocomplex, which may account for an important mechanism and a trigger in initiation of motor neuron degeneration in ALS.

Two experiments were conducted to evaluate the effects of fenugreek seeds (FS) as a potential alternative to antibiotic growth promoters in broiler chickens. In the first experiment, one-day-old Ross (n = 160) straight-run broilers were fed FS at 0 g, 2.5 g, 5 g, and 10 g/kg of diet during the starter (from 1 to 21 days) and finisher phase (from 22 to 35 days) with four replicates of ten birds each. In the second experiment, one-day-old Ross (n = 144) male broilers were fed 0 g, 5 g, and 10 g FS per kilogram of diet during the starter (from 1 to 21 days) and finisher phase (from 22 to 42 days) with six replicates of eight birds each. In addition to growth performance, hematological parameters and intestinal histomorphology were measured in the second experiment. FS linearly reduced the body weight gain (BWG) (p < 0.001), feed intake (FI) (p < 0.05), and increased feed conversion ratio (FCR) (p < 0.05) during the starter phase in both experiments. However, no significant effects on BWG, FI, and FCR were observed during the finisher phase. Moreover, the overall BWG and FI were linearly reduced (p < 0.05) with the increasing levels of FS, but BWG and FI were similar in the 5 g/kg FS group and control group. The inclusion of FS had a linear increase in white blood cell (WBC), heterophil, and lymphocyte count (p < 0.005) and the decrease in hematocrit % (p = 0.004) and total bilirubin (p = 0.001). The villus height and villus height: crypt depth ratio of jejunum and ileum were significantly lower in 5 g FS and 10 g FS treatments (p < 0.001) compared to the control. The result indicates that the dietary inclusion of FS reduces the early growth performance, increases the WBC counts, and negatively affects the intestinal morphology of broiler chickens.

In our study we aim to compare the computer resources utilized by cubic and linear splines. We will access and use publicly available data collected by NASA&#x0027;s Global Ecosystem Dynamic Investigation&#x0027;s (GEDI) Light Detection and Ranging, or LiDAR, technology to model waveforms using cubic and linear b-spline interpolation. We will compare the viability and efficiency of these two interpolation methods to determine which is more applicable when working with GEDI data, and when working with datasets similar in size, scope, and variability specifically as it relates to artificial intelligence creation. This study will assist researchers working with GEDI data and data like it to make more informed decisions about creating models and the efficiency of those models as it relates to large scale artificial intelligence deployment.

During the Trump era there was a significant rise of hate crimes, racial bullying, and violence against the LGBTQ+ community which can be explained by political polarization. Both major political parties were pushed to the far ends of the spectrum to counteract the influence of the other side. We tested how this polarization occurs at the level of individual political issues, and study the political factors under Trump that contributed to it. We used a survey research method to collect data on peoples’ beliefs over 5 different contentious political topics (abortion, climate change, gun control, healthcare, and immigration). This data was compared to data from studies performed in 2016 (pre-Trump). To ensure standardization of the data, our survey used the same questions as the previous surveys. Along with the questions gauging opinion, we also included an individual question per issue that gauged how the participant formed that opinion. Compared to 2016 there was a shift towards more government involvement and regulation in the areas of healthcare and gun control, respectively. There was a shift towards environmental protection, and less stringent immigration standards. More participants were in favor of abortion. Most participants said they formed ALL of their political opinions individually. However, social media and major news outlets had played a role in shaping opinions about abortion and environment, respectively. Compared to 2016 surveys there was a significant change in public opinion about various issues of contemporary importance, partly influenced by political polarization and by social media and news outlets.

Cancer is the common name used to categorize a collection of diseases. In the United States, there were an estimated 1.8 million new cancer cases and 600,000 cancer deaths in 2020. Though it has been proven that an early diagnosis can significantly reduce cancer mortality, cancer screening is inaccessible to much of the world’s population. Machine learning approaches are increasingly successful in image-based diagnosis, disease prognosis, and risk assessment. A literature search with the Google Scholar and PubMed databases from January 2020 to June 2021 determined that currently, no machine learning model (n=0/417) has an accuracy of 90% or higher in diagnosing multiple cancers. We propose our model HOPE, the Heuristic Oncological Prognosis Evaluator, a transfer learning diagnostic tool for the screening of patients with common cancers. By applying this approach to magnetic resonance (MRI) and digital whole slide pathology images, HOPE 2.0 demonstrates an overall accuracy of 95.52% in classifying brain, breast, colorectal, and lung cancer. HOPE 2.0 is a unique state-of-the-art model, as it possesses the ability to analyze multiple types of image data (radiology and pathology) and has an accuracy higher than existing models. HOPE 2.0 may ultimately aid in accelerating the diagnosis of multiple cancer types, resulting in improved clinical outcomes compared to previous research that focused on singular cancer diagnosis.

Global access to sanitary water is of utmost importance to human health. Presently, textile dye water pollution and cigarette pollution are both plaguing the environment. Herein, waste cigarette filters (CFs) are converted into useful carbon-based adsorbent materials via a facile, microwave-assisted carbonization procedure. The CFs are activated and co-doped with phosphorus and nitrogen simultaneously to enhance their surface characteristics and adsorbent capability by introducing chemisorptive binding sites to the surface. The doped carbonized CF (DCCF) and undoped carbonized CF (CCF) adsorbents are characterized physically to examine their surface area, elemental composition, and surface charge properties. The maximum adsorption capacity of synthesized adsorbents was determined via batch adsorption experiments and Langmuir modelling. Additionally, the influence of different parameters on the adsorption process was studied by varying the adsorption conditions such as adsorbent dosage, initial concentration, contact time, temperature, and pH. The DCCF adsorbent showed a maximum adsorption capacity of 303 mg g− 1. Adsorption of both adsorbents fit best to Langmuir model and pseudo-second order kinetics, indicating chemisorptive mechanism. Both adsorbents showed endothermic adsorption process which is indicated by increasing adsorption capacity with increased temperatures. DCCF exhibited greater adsorption capability than CCF at all temperatures from 25 to 55 °C. The pH of the solution significantly affected the adsorption capacity of CCF while DCCF adsorption is favorable at a wide pH range due to low value of the adsorbent’s point of zero charge. Reusability results showed that both adsorbents can be used over several cycles for removal of dye. Thus, results conclude that the waste DCCF-based adsorbent does not only show a profound potential as a sustainable solution to combat textile dye water pollution but also addresses the valuable use of the CF pollution simultaneously. This approach, which can target two major pollutants, is attractive due to its ease of preparation, negligible cost, and versatility in application.

Renewable resources and their byproducts are becoming of growing interest for alternative energy. Here, we have demonstrated the use of Arkansas’ most important crop, soy, as a carbon precursor for the synthesis of carbonized activated materials for supercapacitor applications. Different soy products (soymeal, defatted soymeal, soy flour and soy protein isolate) were converted into carbonized carbon and co-doped with phosphorus and nitrogen simultaneously, using a facile and time-effective microwave synthesis method. Ammonium polyphosphate was used as a doping agent which also absorbs microwave radiation. The surface morphology of the resulting carbonized materials was characterized in detail using scanning electron microscopy. X-ray photoelectron spectroscopy was also performed, which revealed the presence of a heteroelemental composition, along with different functional groups at the surface of the carbonized materials. Raman spectroscopy results depicted the presence of both a graphitic and defect carbon peak, with defect ratios of over one. The electrochemical performance of the materials was recorded using cyclic voltammetry in various electrolytes including acids, bases and salts. Among all the other materials, soymeal exhibited the highest specific capacitance value of 127 F/g in acidic electrolytes. These economic materials can be further tuned by changing the doping elements and their mole ratios to attain exceptional surface characteristics with improved specific capacitance values, in order to boost the economy of Arkansas, USA.

Abstract Global access to sanitary water is of utmost importance to human health. Presently, textile dye water pollution and cigarette pollution are both plaguing the environment. Herein, waste cigarette filters are converted into useful carbon-based adsorbent materials via a facile, microwave-assisted carbonization procedure. The cigarette filters are co-doped with phosphorus and nitrogen using ammonium polyphosphate to enhance their surface characteristics and adsorbent capability. The adsorbents are characterized physically to examine their surface area, elemental composition, and surface charge properties. Batch adsorption experiments were performed to determine the maximum adsorption capacity of the adsorbents. Additionally, the effects of various adsorption parameters— temperature, adsorbent dosage, pH, and time—on adsorption process were examined. The doped adsorbent showed a maximum adsorption capacity of 303.3 mg g− 1 respectively, which is three times that of the methylene blue adsorption capacity of commercially available activated carbon (~ 100 mg g− 1). Thus, the phosphorus and nitrogen co-doped carbonized waste cigarette filter adsorbent shows a profound potential as a sustainable solution to combat textile dye water pollution and cigarette filter pollution simultaneously, due to its low cost, simple preparation, and versatility in application.