How groundwater time series and aquifer property data explain heterogeneity in the Permo-Triassic sandstone aquifers of the Eden Valley, Cumbria, UK

A novel investigation of the impact of meteorological and geological heterogeneity within the Permo-Triassic Sandstone aquifers of the River Eden catchment, Cumbria (UK), is described. Quantifying the impact of heterogeneity on the water cycle is increasingly important to sustainably manage water resources and minimise flood risk. Traditional investigations on heterogeneity at the catchment scale require a considerable amount of data, and this has led to the analysis of available time series to interpret the impact of heterogeneity. The current research integrated groundwater-level and meteorological time series in conjunction with aquifer property data at 11 borehole locations to quantify the impact of heterogeneity and inform the hydrogeological conceptual understanding. The study visually categorised and used seasonal trend decomposition by LOESS (STL) on 11 groundwater and meteorological time series. Decomposition components of the different time series were compared using variance ratios. Though the Eden catchment exhibits highly heterogeneous rainfall distribution, comparative analysis at borehole locations showed that (1) meteorological drivers at borehole locations are broadly homogeneous and (2) the meteorological drivers are not sufficient to generate the variation observed in the groundwater-level time series. Three distinct hydrogeological regimes were identified and shown to coincide with heterogeneous features in the southern Brockram facies, which is the northern silicified region of the Penrith Sandstone and the St Bees Sandstone. The use of STL analysis in combination with detailed aquifer property data is a low-impact insightful investigative tool that helps guide the development of hydrogeological conceptual models.

Short-Term Power Load Forecasting Under COVID-19 Based on Graph Representation Learning With Heterogeneous Features

Since 2020, the COVID-19 has spread globally at an extremely rapid rate. The epidemic, vaccination, and quarantine policies have profoundly changed economic development and human activities worldwide. As many countries start to resume economic activities aiming at a “living with COVID” new normal, a short-term load forecasting technique incorporating the epidemic’s effects is of great significance to both power system operation and a smooth transition. In this context, this paper proposes a novel short-term load forecasting method under COVID-19 based on graph representation learning with heterogeneous features. Unlike existing methods that fit power load data to time series, this study encodes heterogeneous features relevant to electricity consumption and epidemic status into a load graph so that not only the features at each time moment but also the inherent correlations between the features can be exploited; Then, a residual graph convolutional network (ResGCN) is constructed to fit the non-linear mappings from load graph to future loads. Besides, a graph concatenation method for parallel training is introduced to improve the learning efficiency. Using practical data in Houston, the annual, monthly, and daily effects of the crisis on power load are analyzed, which uncovers the strong correlation between the pandemic and the changes in regional electricity utilization. Moreover, the forecasting performance of the load graph-based ResGCN is validated by comparing with other representative methods. Its performance on MAPE and RMSE increased by 1.3264 and 15.03%, respectively. Codes related to all the simulations are available on https://github.com/YoungY6/ResGCN-for-Short-term-power-load-forecasting-under-COVID-19.

Integration of Borehole Acoustic Reflectivity Survey and Fracture Pressure Analysis to Determine Properties of Far-Field Heterogeneities During Stimulation in Tight Reservoirs

The presence of conductive and extended heterogeneous features not connected to the wellbore and located beyond the investigation depths of standard characterization tools can be the reason for unexpected loss of net pressure during stimulation treatments due to the hydraulic fracture breakthrough into these heterogeneous areas. In current field practice, if such breakthrough occurs, it is considered as bad luck without the possibility of the quantitative analysis. This mindset can be changed in favor of the stimulation and reservoir management success using an approach that ties the thorough fracture pressure analysis with the output of the specific acoustic reflectivity survey capable of identifying position, shape, and orientation of far-field heterogeneous features. The approach consists of four steps and is applicable to cases when the hydraulic fracture experiences breakthrough into the heterogeneity. First, before the stimulation treatments, at the reservoir characterization stage, a borehole acoustic reflectivity survey is run. Gathered data are interpreted and visualized according to a specific workflow that yields the image of the heterogeneous areas located around the wellbore in the radius of several tens of meters. Second, the hydraulic fracturing treatment is performed, and fracture pressure analysis is performed, which identifies the pressure drops typical for the breakthrough. Third, after the breakthrough into the heterogeneity is confirmed, the distance to this heterogeneity is used as a marker for calibration of the fracture properties and geometry. Finally, the post-stimulation pressure and production data are used to define the properties of the heterogeneous features, such as conductivity and approximate dimensions. The comprehensive field application example of the suggested approach confirmed its effectiveness. For the tight carbonate formations, the heterogeneity in a form of fracture corridor was revealed by the acoustic reflectivity survey at least 20 m away from the wellbore. The breakthrough into this heterogeneity was observed during the acid fracturing treatment. The distance to the heterogeneity and observed pumping time to breakthrough were used as markers characterizing fracture propagation; reservoir and rock properties were adjusted using a fracturing simulator to obtain this fracture propagation. Finally, the post-stimulation production data were analyzed to determine infinite conductivity of the fracture corridor and quantify its extent downward. Data gathered during reservoir and hydraulic fracture properties calibration allowed for optimization of stimulation strategy of the target layer throughout the field; the information about the heterogeneity’s properties allowed for optimization of the completion and reservoir development strategy.

Distinct and interchangeable growing patterns in colorectal cancer stem-like cells are regulated by Musashi-1

The dynamic and heterogeneous features of cancer stem-like cells (CSCs) have been widely recognized, but their nongenetic cellular plasticity mechanisms remain elusive. By using colorectal cancer organoids, we phenotypically tracked their spheroid formation and growth capacity to a single-cell resolution, and we discovered that the spheroid-forming cells exhibit a heterogeneous growth pattern, consisting of slow- and fast-growing spheroids. The isolated fast-growing spheroids seem to preserve a dual-growing pattern through multiple passages, whereas the isolated slow-growing spheroids are restricted to a slow-growing pattern. Notably, the spheroids of both patterns were tumorigenic. Moreover, the expression of CSC markers varied among the subpopulations with different growth patterns. The isolated slow-growing spheroids adopted the dual-growing pattern by various extrinsic triggers, in which Musashi-1 plays a key role. The slow-growing fraction was resistant to chemotherapy, and its successful isolation can provide an in vitro platform allowing us to elucidate their role in drug resistance.

New Prognostic Biomarkers and Drug Targets for Skin Cutaneous Melanoma via Comprehensive Bioinformatic Analysis and Validation

Skin cutaneous melanoma (SKCM) is the most aggressive and fatal type of skin cancer. Its highly heterogeneous features make personalized treatments difficult, so there is an urgent need to identify markers for early diagnosis and therapy. Detailed profiles are useful for assessing malignancy potential and treatment in various cancers. In this study, we constructed a co-expression module using expression data for cutaneous melanoma. A weighted gene co-expression network analysis was used to discover a co-expression gene module for the pathogenesis of this disease, followed by a comprehensive bioinformatics analysis of selected hub genes. A connectivity map (CMap) was used to predict drugs for the treatment of SKCM based on hub genes, and immunohistochemical (IHC) staining was performed to validate the protein levels. After discovering a co-expression gene module for the pathogenesis of this disease, we combined GWAS validation and DEG analysis to identify 10 hub genes in the most relevant module. Survival curves indicated that eight hub genes were significantly and negatively associated with overall survival. A total of eight hub genes were positively correlated with SKCM tumor purity, and 10 hub genes were negatively correlated with the infiltration level of CD4+ T cells and B cells. Methylation levels of seven hub genes in stage 2 SKCM were significantly lower than those in stage 3. We also analyzed the isomer expression levels of 10 hub genes to explore the therapeutic target value of 10 hub genes in terms of alternative splicing (AS). All 10 hub genes had mutations in skin tissue. Furthermore, CMap analysis identified cefamandole, ursolic acid, podophyllotoxin, and Gly-His-Lys as four targeted therapy drugs that may be effective treatments for SKCM. Finally, IHC staining results showed that all 10 molecules were highly expressed in melanoma specimens compared to normal samples. These findings provide new insights into SKCM pathogenesis based on multi-omics profiles of key prognostic biomarkers and drug targets. GPR143 and SLC45A2 may serve as drug targets for immunotherapy and prognostic biomarkers for SKCM. This study identified four drugs with significant potential in treating SKCM patients.

Paddy seed variety identification using T20-HOG and Haralick textural features

The seed is an inevitable element for agricultural and industrial production. The non-destructive paddy seed variety identification is essential to assure paddy purity and quality. This research is aimed at developing a computer vision-based system to identify paddy varieties using multiple heterogeneous features, exploiting textural, external, and physical properties. We captured the paddy seed images without any fixed setup to make the system user friendly at both industry and farmer levels, which can lead to illumination problems in the images. To overcome this problem, we introduced a modified histogram oriented gradient (T20-HOG) feature that can describe the illumination, scale, and rotational variations of a paddy image. We also utilized the existing Haralick and traditional features and the dimensionality of the features is reduced by the Lasso feature selection technique. The selected features are used to train the feed-forward neural network (FNN) to predict the paddy variety. The experiments conducted on two different datasets: BDRICE, and VNRICE. Results of our method are shown in terms of four standard evaluation metrics, namely, accuracy, precision, recall, and F_1 score, and achieved 99.28%, 98.64%, 98.48%, and 98.56% score, respectively. We also compared our system efficiency with existing studies. The experimental results demonstrate that our proposed features are effective to identify paddy variety and achieved a new state-of-the-art performance. And we also observed that our newly proposed T20-HOG features have a major impact on overall system performance.

Modeling virus pandemics in a globally connected world a challenge towards a mathematics for living systems

This editorial paper presents the papers published in a special issue devoted to the modeling and simulation of mutating virus pandemics in a globally connected world. The presentation is proposed in three parts. First, motivations and objectives are presented according to the idea that mathematical models should go beyond deterministic population dynamics by considering the multiscale, heterogeneous features of the complex system under consideration. Subsequently, the contents of the papers in this issue are presented referring to the aforementioned complexity features. Finally, a critical analysis of the overall contents of the issue is proposed, with the aim of providing a forward look to research perspectives.

Predicting Traffic Congestion Evolution: A Deep Meta Learning Approach

Many efforts are devoted to predicting congestion evolution using propagation patterns that are mined from historical traffic data. However, the prediction quality is limited to the intrinsic properties that are present in the mined patterns. In addition, these mined patterns frequently fail to sufficiently capture many realistic characteristics of true congestion evolution (e.g., asymmetric transitivity, local proximity). In this paper, we propose a representation learning framework to characterize and predict congestion evolution between any pair of road segments (connected via single or multiple paths). Specifically, we build dynamic attributed networks (DAN) to incorporate both dynamic and static impact factors while preserving dynamic topological structures. We propose a Deep Meta Learning Model (DMLM) for learning representations of road segments which support accurate prediction of congestion evolution. DMLM relies on matrix factorization techniques and meta-LSTM modules to exploit temporal correlations at multiple scales, and employ meta-Attention modules to merge heterogeneous features while learning the time-varying impacts of both dynamic and static features. Compared to all state-of-the-art methods, our framework achieves significantly better prediction performance on two congestion evolution behaviors (propagation and decay) when evaluated using real-world dataset.

UIBert: Learning Generic Multimodal Representations for UI Understanding

To improve the accessibility of smart devices and to simplify their usage, building models which understand user interfaces (UIs) and assist users to complete their tasks is critical. However, unique challenges are proposed by UI-specific characteristics, such as how to effectively leverage multimodal UI features that involve image, text, and structural metadata and how to achieve good performance when high-quality labeled data is unavailable. To address such challenges we introduce UIBert, a transformer-based joint image-text model trained through novel pre-training tasks on large-scale unlabeled UI data to learn generic feature representations for a UI and its components. Our key intuition is that the heterogeneous features in a UI are self-aligned, i.e., the image and text features of UI components, are predictive of each other. We propose five pretraining tasks utilizing this self-alignment among different features of a UI component and across various components in the same UI. We evaluate our method on nine real-world downstream UI tasks where UIBert outperforms strong multimodal baselines by up to 9.26% accuracy.