Robust dynamic equivalents of large power networks are essential for fast and reliable stability analysis of bulk power systems. This is because the dimensionality of modern power systems raises convergence issues in modern stability-analysis programs. However, even with modern computational power, it is challenging to find reduced-order models for power systems due to the following factors: the tedious mathematical analysis involved in the classical reduction techniques requires large amounts of computational power; inadequate information sharing between geographical areas prohibits the execution of model-dependent reduction techniques; and frequent fluctuations in the operating conditions (OPs) of power systems necessitate updates to reduced models. This paper focuses on a measurement-based approach that uses a deep artificial neural network (DNN) to estimate the dynamics of an external system (ES) of a power system, enabling stability analysis of a study system (SS). This DNN technique requires boundary measurements only between the SS and the ES. However, machine learning-based techniques like this DNN are known for their extensive training requirements. In particular, for power systems that undergo continuous fluctuations in operating conditions due to the use of renewable energy sources, the applications of this DNN technique are limited. To address this issue, a Deep Transfer Learning (DTL)-based technique is proposed in this paper. This approach accounts for variations in the OPs such as time-to-time variations in loads and intermittent power generation from wind and solar energy sources. The proposed technique adjusts the parameters of a pretrained DNN model to a new OP, leveraging symmetry in the balanced adaptation of model layers to maintain consistent dynamics across operating conditions. The experimental results were obtained by representing the Queensland (QLD) system in the simplified Australian 14 generator (AU14G) model as the SS and the rest of AU14G as the ES in five scenarios that represent changes to the OP caused by variations in loads and power generation.

The constitutive expression of EhLINE1_ORF1p in Entamoeba histolytica cells in the absence of retrotransposition suggests that this protein may serve roles beyond retrotransposition. To delve into this possibility, we generated and analyzed transcriptomic data of EhLINE1_ORF1p partial-knockdown cell lines. This analysis unveiled the significance of EhLINE1_ORF1p in the growth of E. histolytica cells. Our investigation of the consequences of EhLINE1_ORF1p partial knockdown highlighted its impact on the genes involved in ribosome biogenesis, pre-rRNA processing, RNA helicase, and oxidoreductase activity, particularly those associated with amoebiasis. These affected genes are categorized as having high expression and are essential for optimal cell growth. Validation through western blotting confirmed a substantial and pronounced decline in EhLINE1_ORF1p level, and growth kinetics confirmed decline in cell growth. These findings provide compelling evidence supporting the plausible role of EhLINE1_ORF1p in regulating the growth of E. histolytica.

BACKGROUND:High-resolution typing of human leukocyte antigen (HLA) may revolutionize the field of kidney transplantation by selection of low immunogenic grafts. The new 250-nautical mile circle allocation system offers a unique opportunity to find low HLA immunogenic donors for eligible recipients. METHODS:501 transplant candidates from the University of Toledo Medical Center (UTMC) between 2015 and 2019, registered at the Scientific Registry of Transplant Recipients (SRTR) were virtually matched to 4812 donors procured within 250-nautical miles using an in-house-developed simulation algorithm. Immunogenicity of HMS (hydrophobic mismatch score) ≤10 was measured based on imputed high-resolution HLAs. Simulated "optimal" matches with a KDPI≤50 % were compared with the transplant cohort between 2000 and 2010 with their kidney allograft survivals. RESULTS:Out of 501 recipients 500 (99.8 %) were matched with donors ≤10 HMS and KDPI ≤50 %. The average HMS value for simulated transplants was 1.4 (range 0-10) versus 6.3 (range 0-75) in the retrospective cohort (p < 0.001). The simulated model had a median mismatch number of 3/6, while the reference cohort 4/6 among HLA-A/B/DR antigens (p < 0.001). The estimated median graft survival was 18.2 years for the simulated cohort vs. 13.4 years in the real-life cohort (p < 0.001), gaining 4.9 years per transplant and 2450 survival years for all patients. For year 2014, out of 98 patients and 659 donors, each recipient had a median number of 141 donors (HMS < 10; range 8-378). Similar values were found for patients between 2015 and 2019. CONCLUSION:Donors within 250-nautical miles proffers excellent and multiple options for finding well-matched low immunogenic HLA kidney donors for UTMC patients, thus significantly improving their chances for long-term allograft survival.

HLA matching improves long-term outcomes of kidney transplantation, yet implementation challenges persist, particularly within the African American (Black) patient demographic due to donor scarcity. Consequently, kidney survival rates among Black patients significantly lag behind those of other racial groups. A refined matching scheme holds promise for improving kidney survival, with prioritized matching for Black patients potentially bolstering rates of HLA-matched transplants.To facilitate quantity, quality and equity in kidney transplants, we propose two matching algorithms based on quantification of HLA immunogenicity using the hydrophobic mismatch score (HMS) for prospective transplants. We mined the national transplant patient database (SRTR) for a diverse group of donors and recipients with known racial backgrounds. Additionally, we use novel methods to infer survival assessment in the simulated transplants generated by our matching algorithms, in the absence of actual target outcomes, utilizing modified unsupervised clustering techniques.Our allocation algorithms demonstrated the ability to match 87.7% of Black and 86.1% of White recipients under the HLA immunogenicity threshold of 10. Notably, at the lowest HMS threshold of 0, 4.4% of Black and 12.1% of White recipients were matched, a marked increase from the 1.8% and 6.6% matched under the prevailing allocation scheme. Furthermore, our allocation algorithms yielded similar or improved survival rates, as illustrated by Kaplan–Meier (KM) curves, and enhanced survival prediction accuracy, evidenced by C-indices and Integrated Brier Scores.

In recent years deep learning has become the desired approach to analyse long term dependencies in data to build predictive models to forecast future outcome. They have been extensively used for natural language processing, computer vision, image processing, and for analysing sequential data. In this paper, deep variants of Recurrent Neural Network (RNN) called Long Short-Term Memory Network (LSTM), have been used to build and predict the outcomes of multivariate time series data with reference to multivariate stock market prediction data of Meta, Ford, and FedEx. Comparative performance analysis of three variants of LSTM viz., Vanilla LSTM, Stacked LSTM, and Bi-directional LSTM was conducted. Four input features were chosen, viz., opening stock price, high and low prices for the stock and adjusted closing price for a given day. The predicted output was the closing price for the day or the opening price for the next day. Traditional time series prediction methods often fail to capture the nonlinear and dynamic patterns present in stock price data, leading to inaccurate predictions. Deep learning, a subset of machine learning, has shown promising results in various fields, including time series prediction. Our results demonstrate that the proposed multivariate deep learning model Bi-directional LSTM outperforms the other two LSTM models in terms of accuracy and performance.

With the integration of distributed energy resources such as roof-top solar panels and wind turbines into the grid, power generation can surpass demand-generation and thus, giving rise to the negative pricing, especially in the summer months. In this regard, a scientific case study is conducted in this paper to analyse and predict the increasing instances of negative energy prices against demand-generation in Australian energy markets (AEMs) using real-time energy data from the Hornsdale power reserve, South Australia. A robust machine learning method, Light gradient boosting machine (LightGBM) is utilised to detect and predict negative prices at different quantiles to quantity the outliers in the pricing data. The implementation results demonstrate that predicting the prices at different quantiles can tackle outliers (negative prices) effectively with the help of extracted upper and lower bounds using quantile regression-based approach. The case study is further extended to learn the complex statistical relationships between different data features using Naive-Bayes Tree Augmented (NB-TAN) algorithm considering ‘price’ as the dependent feature against the independent features such as demand-generation, battery charging/discharging, and frequency control ancillary services.

The advancements in distributed generation (DG) technologies such as solar panels have led to a widespread integration of renewable power generation in modern power systems. However, the intermittent nature of renewable energy poses new challenges to the network operational planning with underlying uncertainties. This paper proposes a novel probabilistic scheme for renewable solar power generation forecasting by addressing data and model parameter uncertainties using Bayesian bidirectional long short-term memory (BiLSTM) neural networks, while handling the high dimensionality in weight parameters using variational auto-encoders (VAE). The forecasting performance of the proposed method is evaluated using various deterministic and probabilistic evaluation metrics such as root-mean square error (RMSE), Pinball loss, etc. Furthermore, reconstruction error and computational time are also monitored to evaluate the dimensionality reduction using the VAE component. When compared with benchmark methods, the proposed method leads to significant improvements in weight reduction, i.e., from 76,4224 to 2,022 number of weight parameters, quantifying to 97.35% improvement in weight parameters reduction and 37.93% improvement in computational time for 6 months of solar power generation data.

Convolutional neural networks (CNN) have transformed the field of computer vision by enabling the automatic extraction of features, obviating the need for manual feature engineering. Despite their success, identifying an optimal architecture for a particular task can be a time-consuming and challenging process due to the vast space of possible network designs. To address this, we propose a novel neural architecture search (NAS) framework that utilizes the clonal selection algorithm (CSA) to automatically design high-quality CNN architectures for image classification problems. Our approach uses an integer vector representation to encode CNN architectures and hyperparameters, combined with a truncated Gaussian mutation scheme that enables efficient exploration of the search space. We evaluated the proposed method on six challenging EMNIST benchmark datasets for handwritten digit recognition, and our results demonstrate that it outperforms nearly all existing approaches. In addition, our approach produces state-of-the-art performance while having fewer trainable parameters than other methods, making it low-cost, simple, and reusable for application to multiple datasets.

The detection of cell death and identification of its mechanism underpins many of the biological and medical sciences. A scattering microscopy based method is presented here for quantifying cell motility and identifying cell death in breast cancer cells using a label-free approach. We identify apoptotic and necrotic pathways by analyzing the temporal changes in morphological features of the cells. Moreover, a neural network was trained to identify the cellular morphological changes and classify cell death mechanisms automatically, with an accuracy of over 95%. A pre-trained network was tested on images of cancer cells treated with a different chemotherapeutic drug, which was not used for training, and it correctly identified cell death mechanism with ∼100% accuracy. This automated method will allow for quantification during the incubation steps without the need for additional steps, typically associated with conventional technique like fluorescence microscopy, western blot and ELISA. As a result, this technique will be faster and cost effective.

The emergence of cyber-physical smart grid (CPSG) systems has revolutionized the traditional power grid by enabling the bidirectional energy flow between consumers and utilities. However, due to escalated information exchange between the end-users, it has posed a greater challenge to the cyber security mechanisms for the communication networks at the cyber and physical planes. To address these challenges, we propose a Bayesian approach integrated with deep convolutional neural networks (CNN-Bayesian). While, the Bayesian component is used to discriminate cyber-physical intrusions from the normal events in the binary and multi-class events. CNN layers are utilized to handle the high-dimensional feature space prior to the intrusions classification task. The proposed method is validated using real-time Industrial control systems (ICS) dataset against the standard deep learning-based classification methods such as recurrent neural networks (RNN) and long-short term memory (LSTM). From the experimental results, it can be inferred that the proposed CNN-Bayesian method outperforms the existing benchmark classification methods to discriminate intrusions in CPSG systems using evaluation metrics such as accuracy, precision, recall, and $F1$ -score.

Identification of cell death mechanisms, particularly distinguishing between apoptotic versus nonapoptotic pathways, is of paramount importance for a wide range of applications related to cell signaling, interaction with pathogens, therapeutic processes, drug discovery, drug resistance, and even pathogenesis of diseases like cancers and neurogenerative disease among others. Here, we present a novel high-throughput method of identifying apoptotic versus necrotic versus other nonapoptotic cell death processes, based on lensless digital holography. This method relies on identification of the temporal changes in the morphological features of mammalian cells, which are unique to each cell death processes. Different cell death processes were induced by known cytotoxic agents. A deep learning-based approach was used to automatically classify the cell death mechanism (apoptotic vs necrotic vs nonapoptotic) with more than 93% accuracy. This label free approach can provide a low cost (<$250) alternative to some of the currently available high content imaging-based screening tools.

Cell death pathways in multicellular organisms play an important role in maintaining homeostasis and any irregularities can cause diseases like cancer. Studying cell death mechanism will aid in understanding its role in different diseases and their translational implications. It is particularly important for screening new drugs. Here we present a high-throughput and label-free method of identifying the two most common forms of cells death, programmed (apoptosis) and unprogrammed (necrosis), using lensless holographic microscopy. Breast cancer cells BT-20 were treated with a series of drugs, that induce different types of cell death, and imaged continuously during the incubation step. The cells were imaged by illuminating them with partially coherent light and the in-line holograms of the cells, resulting from the interference between the transmitted wave and scattered wave, were recorded in a CMOS imaging chip. The holograms were digitally backpropagated to reconstruct the phase and amplitude images of the cells. The absence of lenses enables imaging at unit magnification over an area <10 mm2, which includes over a thousand mammalian cells. The temporal changes in cell morphology, such as membrane blebbing, shrinking, swelling, and membrane rupture, which are reflected in the phase image was used to identify the cell death mechanism. This process was further automated using deep learning, which enabled the classification of the cell death process with <93% accuracy. This label-free approach of identifying cell death mechanisms enables high-throughput toxicity studies unlike the conventional biochemical assays, e.g., western blot, and can be useful for a variety of biomedical applications.

Convolutional neural networks (CNN) are one of the most widely used and powerful deep learning algorithms. They achieve results that were never possible before and exceed human-level performance in solving computer vision tasks like image classification. The success of CNN algorithms, on the other hand, is in designing the architecture of these networks so that they can be efficiently trained to provide the best performance for a specific problem. Because of the numerous architectural design options, their designs have largely been tuned and optimized manually by human professionals or by using time-consuming heuristics such as grid search. As a result, there is a growing need for deep CNN models to be designed automatically. To carry out the difficult task of automating CNN architectural design, this work proposed a simple yet efficient methodology based on clonal selection algorithms (CSA), a class of optimization methods inspired by the adaptive immunity of the human body. By employing selection, cloning, and mutation operators, CSA can effectively explore a complex and large space to find values close to the global optimum. Our proposed methodology is validated and evaluated using the EMNIST-Digits dataset. The results show that CSA can efficiently generate high-performance, low-cost CNN architectures. Furthermore, the discovered architecture of the EMNIST Digits dataset is tested on other EMNIST datasets with more data instances and classes. The results are impressive, demonstrating that the CSA effectively discovers less expensive and highly accurate CNN architectures for a specific dataset, as well as a reusable architecture across domains with similar properties. Furthermore, our proposed method outperforms previous work that used the same datasets. The proposed method allows for the efficient design of CNN architectures so that the best model can be found. This makes immune-based CNN optimization a promising future approach.

Evaporation residue (ER) cross-sections have been measured for the 48Ti + 140,142Ce reactions to understand the influence of neutron shell closure on compound nucleus (CN) formation. The measured ER cross-sections for 48Ti + 140Ce (NT = 82) system are found to be very close to those of 48Ti + 142Ce (NT = 84) near the Coulomb barrier. The results show that the effect of shell closure in the target nucleus on fusion cross-sections is negligible for the present systems which can be attributed to the small values of the shell correction energies in the entrance channel. Statistical model calculations are performed to interpret the experimental results. Further, comparison of the present results with those from a reaction with higher mass asymmetry in the entrance channel shows no evidence of quasifission in the present systems.

Generally, university academics' knowledge sharing behavior and social intelligence are said to be significant in improving their performance through their grasp of competencies Nevertheless, the association has yet to be thoroughly researched. Thus, this study applied the proposed mediation model among university academics in the Business School of seven Malaysian private universities. The salient findings of this research are that the competencies mediate the association among (1) knowledge sharing behavior and innovative teaching and (2) social intelligence and innovative teaching. Apart from that, it was observed that knowledge sharing behavior impacted innovative teaching and supervision, while social intelligence impacted innovative teaching, research and publication and supervision. Implications were discussed.

Entamoeba histolytica is responsible for dysentery and extraintestinal disease in humans. To establish successful infection, it must generate adaptive response against stress due to host defense mechanisms. We have developed a robust proteomics workflow by combining miniaturized sample preparation, low flow-rate chromatography, and ultra-high sensitivity mass spectrometry, achieving increased proteome coverage, and further integrated proteomics and RNA-seq data to decipher regulation at translational and transcriptional levels. Label-free quantitative proteomics led to identification of 2344 proteins, an improvement over the maximum number identified in E. histolytica proteomic studies. In serum-starved cells, 127 proteins were differentially abundant and were associated with functions including antioxidant activity, cytoskeleton, translation, catalysis, and transport. The virulence factor, Gal/GalNAc-inhibitable lectin subunits, was significantly altered. Integration of transcriptomic and proteomic data revealed that only 30% genes were coordinately regulated at both transcriptional and translational levels. Some highly expressed transcripts did not change in protein abundance. Conversely, genes with no transcriptional change showed enhanced protein abundance, indicating post-transcriptional regulation. This multi-omics approach enables more refined gene expression analysis to understand the adaptive response of E. histolytica during growth stress.

Academic staff’s social intelligence plays an important part in increasing their teaching and learning performance owing to their magnitude of competency. However, empirical research that associates both constructs is limited. Thus, this research aspires to scrutinise the influence of social intelligence on teaching and learning performance among academic staff with the moderating effect of competency. A self-administered questionnaire survey was used to collect data from 318 academic staff in the business schools of seven Malaysian private higher education institutions located in Klang Valley. SPSS version 26 were used to analyse data. The findings revealed the following: (1) social intelligence has a significant positive influence on competency, (2) competency has a significant positive influence on teaching and learning performance, (3) social intelligence has a significant positive influence on teaching and learning performance and (4) competency significantly moderates the association between social intelligence and teaching and learning performance. Theoretical and practical implications are discussed.

Although the multi-layer perceptron (MLP) neural networks provide a lot of flexibility and have proven useful and reliable in a wide range of classification and regression problems, they still have limitations. One of the most common is associated with the optimization algorithm used to train them. The most commonly used training method is stochastic gradient descent with backpropagation (or backpropagation for short) because it is mathematically tractable (given that the activation functions are differentiable). However, backpropagation is not guaranteed to find the globally optimal set of weights and biases. As a result, the MLP is often incapable of obtaining a desirable solution to the problem. Clonal selection algorithms (CSA) are optimization procedures that effectively explore a complex and large space to find values near the global optimum. Consequently, CSA can be used to solve the problem of training MLP networks. This paper presents a novel implementation of CSA for training MLP architectures to solve real-world problems such as breast cancer diagnosis, active sonar target classification, wheat classification, and flower classification. The CSA is used to find the optimal weights and biases that will significantly increase the classification accuracy of the MLP. The performance of our proposed approach is compared with other popular training methods: genetic algorithm (GA), ant colony optimization (ACO), particle swarm optimization (PSO), Harris hawks optimization (HHO), moth-flame optimization (MFO), flower pollination algorithm (FPA), and backpropagation (BP). The comparison is benchmarked using five classification datasets: Iris Flower, Sonar, Wheat Seeds, Breast Cancer Wisconsin, and Haberman's Survival. Comparative study results illustrate the improvements in MLP performance gained by using CSA over other training methods, and hence it can be considered a competitive approach to training MLP networks when solving real-world applications in various disciplines.