Tezpur University

The process of community detection in a network uncovers groups of closely connected nodes, known as communities. In the context of gene correlation networks and neuro-degenerative diseases, this study introduces a systematic pipeline for centrality-based community detection using scRNA-Seq data. Comparisons with existing methods demonstrate its superior performance in terms of modularity. Furthermore, the resulting communities undergo biological validation and hub-gene analysis, which reveal disease-specific pathways and gene ontology associated with the genes within these communities.

This paper highlights the importance of deep learning-based litchi segmentation in precision agriculture using machine vision. The proposed method involves preparing a mixed UAV litchi and MinneApple database, consisting of 2000 images of the same size \(256\times 256\). This paper introduces a modified Mask-RCNN-based instance segmentation model; incorporating a spatial attention block in the backbone network ResNet101, to mitigate one of the significant challenges in litchi counting, i.e., occlusion. The results demonstrate that the proposed model achieves a mean Average Precision (mAP), recall, and F1-score of 90.81%, 89.00%, and 90.35%, respectively, for separated and unoccluded litchis, and an mAP, recall, and F1-score of 81.41%, 82.42%, and 81.91%, respectively, for occluded litchis. The proposed model provides better detection accuracy while minimizing computational burden, showing its potential for efficient and accurate litchi detection and counting in precision agriculture.

This paper analyses the difference between parental cells and cells that acquired radioresistance using scRNA-seq data and investigates the dynamic changes of the transcriptome of cells in response to fractionated irradiation (FIR) towards the identification of potential biomarkers for Esophageal Squamous Cell Carcinoma (ESCC). The divergence of gene expressions is analyzed in response to FIR and the dynamic changes in differentially expressed genes (DEGs) of KYSE-180 cells with two different cumulative doses of FIR (12-Gy and 30-Gy). We construct several biological networks and observe relative to control (0-Gy), 30-Gy induced higher variability of genes. We identified four hub genes TXN, IER2, PCNA, and CENPF involved in ESCC progression.

In this work, starch, itaconic acid, and acrylic acid-based hydrogel was prepared with incorporation of wastepaper derived modified cellulose nanofiber (mCNF) and used as an efficient adsorbent for the removal of cationic dye such as methylene blue (MB), methyl violet (MV), malachite green (MG) and cresol red (CR) from their aqueous media. Moreover, the structure and stability of mCNFs were analyzed using various analytical techniques, while reinforced hydrogel nanocomposite was characterized by different spectroscopic and microscopic techniques. The water swelling study showed that the prepared hydrogel nanocomposite possessed maximum water absorption capacity of 754 g/g due to the incorporation of mCNF within the hydrogel matrix. Besides, mCNF embedded hydrogel possessed a maximum dye adsorption capacity of 414, 405, 377 and 323 mg/g for MB, MG, MV and CR, respectively. The adsorption kinetic and isotherms were found to be fitted well with pseudo-second-order kinetics and Freundlich isotherm model, respectively. Further, reusability test and adsorption of dyes from mixed dye-solution were carried out indicating the prepared hydrogel could effectively remove cationic dyes from the mixture and possessed excellent recyclable properties up to three repeated cycles. Thus, the present study highlights the application of the prepared hydrogel as an effective adsorbent for the removal of cationic dyes from the aqueous media.

The Internet of Medical Things (IoMT), also known as Smart Healthcare, has seen incredible progress in the Smart Environment industry. A significant part of Industry 4.0 is Healthcare 4.0, which is transforming the medical industry to monitor patients’ health remotely and perform other health-related activities. The IoMT includes many wearable sensors that either work on batteries or are rechargeable. In this paper, we work to harvest ambient sources of energy and utilize them for powering medical sensor batteries. Energy Harvesting is one such technique of capturing energy from the surroundings (ambient energy or non-ambient energy) and then converting it to usable electrical energy. We designed a model for Energy Harvesting and IoMT where we take radio frequency as an ambient energy source for harvesting to power the sensors. As medical sensors are mostly used indoors, they are more exposed to radio frequency, which is an ideal source of energy. Finally, simulation results have been obtained showing that the antenna can absorb radio frequency energy up to 50 ohms depending on the impedance of the circuit. Also, the rectifier could convert the AC signal to a DC signal which could be boosted by a voltage regulator to 47 V meeting the requirement to charge a lithium-ion battery.

In this paper, we apply a two-grid scheme to the DG formulation of the 2D transient Navier–Stokes model. The two-grid algorithm consists of the following steps: Step 1 involves solving the nonlinear system on a coarse mesh with mesh size 𝐻, and Step 2 involves linearizing the nonlinear system by using the coarse grid solution on a fine mesh of mesh size ℎ and solving the resulting system to produce an approximate solution with desired accuracy. We establish optimal error estimates of the two-grid DG approximations for the velocity and pressure in energy and L 2 L^{2} -norms, respectively, for an appropriate choice of coarse and fine mesh parameters. We further discretize the two-grid DG model in time, using the backward Euler method, and derive the fully discrete error estimates. Finally, numerical results are presented to confirm the efficiency of the proposed scheme.

Traditional craft commercialization in Assam and its dynamics have been specifically examined in this paper with the help of a theoretical model framed out of concepts such as spontaneous and sponsored commercialization. It is found that craft commercialization in Assam is varied, vibrant, and dynamic as it depends upon the specific role played by artisans, intermediaries, and external agents. The spontaneous effort of artisans in adapting craft objects to various uses for pure artistic pleasure as well as sale and the proximity of producers of the crafts to the market and the customers have been explored. Most of the craftworks were found to have undergone transformation due to the artistic ingenuity of the craftsmen, while many others metamorphosed due to external influence. Agents essayed the role of design and sales agent bringing to light the degree of interface between the producers of the crafts and different types of customers. The commercialization butterfly minutely captures the distinct processes, viz., spontaneous, sponsored, pure-spontaneous, and pure-sponsored, taking place in the craft sector of Assam.

Action/gesture representation especially modeling of actions/gestures has a special role in the recognition process. Here, in this paper, we would primarily look for motion-based hand gesture representations which are widely used but less talked about topics. Model-based and appearance-based methods are the two primary techniques for hand gesture representation. Apart from these two, motion-based approaches have gained quite impressive performance in various applications. Many researchers generally include motion-based methods in appearance-based methods. But here we want to discuss the motion-based methods separately with special attention representing hand gestures. Most of the representations generally depend on the shape, size, and color of the body/body part. But these may vary depending on many factors, e.g., illumination variation, image resolution, skin color, clothing, etc. But motion estimation should be independent of these factors. Optical flow and motion templates are the two major motion-based representation schemes that can be used directly to describe human gestures/actions. The main benefits of these techniques are basically their simplicity, ease of implementation, competitive performance, and efficiency.

Sparse representation classifiers (SRCs) that consider spatial contextual information have outperformed the traditional pixel-based classifiers. Fixed-size square windows have been the most popular choice to capture spatial information. However, they do not represent the actual spatial neighbourhood of the image. To overcome this, many superpixel-based SRC techniques were developed that generate adaptive windows to capture spatial information for classification purposes. The superpixels were generated using segmentation algorithms and hence require prior information. In this paper, we use image’s connected components to define superpixels, which can better capture spatial information. Using two hyperspectral images, the success of the proposed technique was assessed with other similar techniques. The proposed technique produces better accuracy with OA of 75.89% for Pavia University and OA of 89.48% for Indian Pines dataset.

Feature selection (FS) is the problem of finding the most informative features that lead to optimal classification accuracy. In high-dimensional data classification, FS can save a significant amount of computation time as well as can help improve classification accuracy. An important issue in many applications is handling the situation where new instances arrive dynamically. A traditional approach typically handles this situation by recomputing the whole feature selection process on all instances, including new arrivals, an approach that is computationally very expensive and not feasible in many real-life applications. An incremental approach to feature selection is meant to address this issue. In this paper, we propose an effective feature selection method that incrementally scans the data once and computes credibility scores for the features with respect to the class labels. The effectiveness of the proposed method is evaluated on high-dimensional gene expression datasets using different machine learning classifiers.

Bishnupriya Manipuri is an endangered language. Due to very limited resources, it becomes very challenging for carrying out any computational tasks for this type of endangered language. Bishnupriya Manipuri is also among the less computationally explored languages. In this paper, we present an automated part-of-speech tagging approach based on conditional random field using the CRF-Suite library, which is an important and basic task for any natural language processing task. We have carried out experiments on a corpus annotated by us using gold standard tags and achieved a satisfactory tagging accuracy of 86.21%. Further, the results are compared with some existing state-of-the-art approaches.

Tea, the major beverage worldwide, is one of the oldest commercial commodities traded from ancient times. Apart from many of its advantages, including health, socio-economic, climatic, and agro-ecological values, FAO has recognized that the tea value chain covering its growth in the field, processing and marketing, and finally, the hot cup at the user’s hand needs to be made sustainable during all these stages. Tea generates a lot of waste in different forms in different stages of its growth and processing, and these wastes, if not managed properly, may cause environmental pollution. A planned utilization of these wastes as feedstocks for various processes can generate more income, create rural livelihood opportunities, help grow tea environmentally sustainable, avoid GHG emissions, and make a real contribution to SDGs. Thermochemical and biological conversion of tea wastes generates value-added products. This review provides an overview on the impacts of the tea wastes on the environment, tea waste valorization processes, and applications of value-added products. The application of value-added products for energy generation, wastewater treatment, soil conditioners, adsorbents, biofertilizers, food additives, dietary supplements, animal feed bioactive chemicals, dye, colourant, and phytochemicals has been reviewed. Further, the challenges in sustainable utilization of tea wastes and opportunities for commercial exploitation of value-added products from tea wastes have been reviewed.

We construct an S4 flavour symmetric minimal inverse seesaw model where the standard model is extended by adding two right-handed and two standard model gauge singlet neutrinos to explain the origin of tiny neutrino masses. The resulting model describes the lepton mass spectra and flavour mixing quite well for the case of the normal hierarchy of neutrino masses. The prediction of the model on the Dirac CP-violating phase is centered around 370.087◦. Furthermore, using the allowed region for the model parameters, we have calculated the value of the effective Majorana neutrino mass, |⟨mee⟩|, which characterizes neutrinoless double beta decay.

In this paper, we have realized the left–right symmetric model (LRSM) with modular symmetry. Most of the previous works on LRSM have been done considering discrete flavor symmetry, but this work has been carried out using [Formula: see text](3) modular group which is isomorphic to nonabelian discrete symmetry group [Formula: see text]. The advantage of using modular symmetry is the nonrequirement for the use of extra particles called “flavons”. In this model, the Yukawa couplings are expressed in terms of modular forms [Formula: see text]. In this work, we have studied minimal LRSM for both type-I and type-II dominances. Here, we have calculated the values for the Yukawa couplings and then plotted it against the sum of the neutrino masses. The results obtained are well within the experimental limits for the desired values of sum of neutrino masses. We have also briefly analyzed the effects of the implications of modular symmetry on neutrinoless double beta decay with the new physics contributions and the correlation of lepton flavor violation and lightest neutrino mass within the framework of modular symmetric LRSM.

The uses of highly luminescent perovskite quantum dots in real analytical detection were limited by their supersensitive nature. Here, we have designed CsPbBr3 perovskite based fluorometric sensor by integrating them...

With the advent of lab-on-a-droplet based strategies for various applications, the droplet generation process in miniaturized devices has received notable attention from the research community. For long, researchers have developed varied strategies of tuning the size and frequency of droplets so as to reduce reagent consumption, production of waste and increase controllability and portability for microfluidic usage. Hence, in the interest of minimizing droplet size, a novel technique of incorporating baffle in a microfluidic T-junction for generating monodisperse droplets of smaller size than the conventional T-junction is proposed in this study. Through comprehensive 3-D numerical simulations using the Level-Set method, it has been found that the inclusion of baffle in the T-junction reduces the droplet size by 22.5%, thus meeting demands of minimized size monodispersed droplet generation without encountering fabrication complexity issues.

Cadmium (Cd) is a non-essential highly toxic element that poses a potential health threat for plants, humans, and animals at considerably smaller concentrations. Plants can uptake Cd from soil and water due to its high mobility. In this review, the plant responses toward Cd phytotoxicity and the mechanism of Cd tolerance are summarized. Prevalent responses of Cd toxicity in plants are DNA damage, alteration in gene expression, cell division, and cell death which lead to metabolic, anatomic, and morphological modifications ranging from protein degradation to lower uptake of water and nutrients, chlorosis, inhibition of photosynthesis, and crop yield losses. Plants possess an array of mechanisms like cell wall binding, reduced transport, compartmentation in the vacuole, and chelation with metallothioneins, or phytochelatins in the detoxification and thus tolerance to Cd. Some plants develop structural and genetic adaption to achieve tolerance to Cd. Activation of plant antioxidative defence system and modulation of hormonal levels, alteration of secondary metabolites and higher mineral nutrition triggers alleviation of Cd stress. The current state of knowledge on the effect of Cd stress and possible management strategies to avert damage and develop Cd-resistant crops have also been discussed.

Molecular-iodine catalyzed access to an important class of bio-relevant indole derivative- cyclopenta[b]indoles have been achieved via cascade addition/intramolecular cyclization of indoles and acetone. Explorations on diverse substitution pattern revealed an...

This paper presents the design analysis and measurement of a linearly polarized microstrip patch antenna resonating in the X-band regime with a self-compensating mechanism to mitigate the detuning effects due to bending. The antenna is studied for its variation in the resonant frequency for bending radii up to 20mm and a compensation of ≈ 50% is achieved as compared to its counterpart. The compensation technique is realized by altering the effective dielectric constant of the substrate by injecting dielectric fluids into the grooves incorporated in the substrate, supplemented by a shape variation of the grooves due to bending. The antennas demonstrated in the current work exhibit a gain of more than 7 dBi throughout along with a consistent -10 dB impedance bandwidth.