Non-negative Matrix Factorization Research Articles

Dual semi-supervised hypergraph regular multi-view NMF with anchor graph embedding

Graph regularized nonnegative matrix factorization (GNMF) has been widely used in multi-view clustering tasks due to its good clustering properties. However, it uses a simple graph to describe the complex data relationship of multiple views and tries to obtain a consistent low-dimensional representation, which undoubtedly brings challenges to its clustering performance. In addition, clustering a large amount of high-dimensional data from multiple views undoubtedly faces a huge computational burden. In order to effectively improve the performance and efficiency of GNMF based multi-view clustering algorithm, this paper proposes a dual semi-supervised hypergraph regular multi-view clustering method with anchor graph embedding (DSSHMNMFAE). Specifically, DSSHMNMFAE develops a new anchor selection method to generate anchors and the anchor bipartite graph is constructed to embed the matrix factorization process. DSSHMNMFAE constructs a hypergraph to effectively learn the high-order relationship between data from multiple views. In order to perform semi-supervised learning more efficiently, DSSHMNMFAE integrates pairwise constraint information and label constraint information into the clustering process as dual label information. In addition, DSSHMNMFAE considers the learning of both consistency information and complementarity information, and adopts adaptive measures to distinguish the contributions of different views. We use the alternating iterative algorithm to optimize the objective function of DSSHMNMFAE. The experimental results on eight real datasets show that the performance of DSSHMNMFAE is comparable to other algorithms.

An exploratory Q-matrix estimation method based on sparse non-negative matrix factorization.

Cognitive diagnostic assessment (CDA) is widely used because it can provide refined diagnostic information. The Q-matrix is the basis of CDA, and can be specified by domain experts or by data-driven estimation methods based on observed response data. The data-driven Q-matrix estimation methods have become a research hotspot because of their objectivity, accuracy, and low calibration cost. However, most of the existing data-driven methods require known prior knowledge, such as initial Q-matrix, partial q-vector, or the number of attributes. Under the G-DINA model, we propose to estimate the number of attributes and Q-matrix elements simultaneously without any prior knowledge by the sparse non-negative matrix factorization (SNMF) method, which has the advantage of high scalability and universality. Simulation studies are carried out to investigate the performance of the SNMF. The results under a wide variety of simulation conditions indicate that the SNMF has good performance in the accuracy of attribute number and Q-matrix elements estimation. In addition, a set of real data is taken as an example to illustrate its application. Finally, we discuss the limitations of the current study and directions for future research.

Heavy metal contamination assessment and source attribution in the Vicinity of an iron slag pile in Hechi, China: Integrating multi-medium analysis

Heavy metals, such as mercury, cadmium, and nickel, may contaminate human inhabited environments, with critical consequences for human health. This study examines the health impacts of heavy metal pollution from an iron slag pile in Hechi, China, by analyzing heavy metal contamination in water, sediment, soil, and crops. Here, the Nemerow pollution index (NI) indicated severe pollution at most sampling sites, the mean NI of groundwater, and surface water had reached 594.13 and 26.79, respectively. Bioaccumulation of mercury (Hg), cadmium (Cd), and nickel (Ni) was noted in crops, cucumbers showed comparatively lower risk levels. Logarithmic surface water-sediment partition coefficient calculations indicated that heavy metals such as chromium (Cr), ferrum (Fe), zinc (Zn), copper (Cu), Ni, arsenic (As), and lead (Pb) tend to accumulate in sediments. There was a high risk in groundwater (67.48–6590.54) and surface water (13.73–2500.85). Variably influenced by rainfall, these metals can be diluted and mobilized from surface water and sediments, thereby changing the contamination levels and ecological risks. Probabilistic health risk assessments indicated that health risks were higher in children than in adults, the mean total carcinogenic risk values of soil, groundwater, and surface water, were 6.79E-04, 4.20E-06, and 1.15E-6 for children, respectively. Moderate soil pollution is the main health hazard. A Positive Matrix Factorization model attributed over 60% of the pollution to slag stacking. Biotechnologies, solidification/stabilization techniques, field management, and institutional controls, driven by principles of green, low-carbon, and economic efficiency may mitigate. These findings contribute to the management of heavy metal pollution in iron slag pile areas.

Molecular characterization of oxidized organic nitrogen in the polluted urban atmosphere of Beijing

Oxidized organic nitrogen (OON) serves as a crucial link between volatile organic compounds (VOCs), nitrogen oxides, ozone, and secondary organic aerosol (SOA). However, the comprehension of the molecular composition, formation mechanism, and implications of OON remains limited, particularly in urban environments influenced heavily by anthropogenic emissions. This study presents the field measurements of OON conducted at an urban site in Beijing using chemical ionization mass spectrometry with nitrate as the reagent ion. The molecular characteristics, dominant species, and oxidation states of OON were investigated. Positive matrix factorization analysis disentangled the diverse OON into factors characterized by unique fingerprint features and formation pathways. A modified workflow was used to classify the majority of OON as terpene (3.0 %), isoprene (15.2 %), aliphatic (38.1 %, containing dinitrate), and aromatic (36.0 %, containing aromatic ring retaining) OON. This highlights the significant impact of anthropogenic sources and underscores the need for stringent controls on anthropogenic VOC emissions to mitigate OON formation. Volatility estimates further indicate that aromatic and aliphatic OON, with relatively low volatility, are expected to be the primary contributors to SOA formation by condensation or gas-particle partitioning in urban Beijing. In addition, hazy weather conditions may facilitate multi-generation reactions, leading to the production of large amounts of semi-volatile dinitrate OON and promoting its conversion to SOA.

Characterization and source apportionment of ambient VOC concentrations: Assessing ozone formation potential in the Barnett shale oil and gas region

Atmospheric concentration and distribution of volatile organic compounds (VOC) are influenced by localized emissions and the fate and transport of secondary products formed through photochemical reactions. This study identifies the sources of VOC and its contribution to ozone formation in the Barnett Shale region, specifically in Denton, Texas, during the ozone seasons of 2015 through 2022. A total of 31 critical ozone precursor VOC compounds were observed at this site with an average total VOC concentration of 146 ppb-C, with n-alkanes from extensive oil and gas activities in the area contributing to 96.64% of the total VOC. Source apportionment using dispersion normalized positive matrix factorization (DN-PMF) identified eight VOC sources, with natural gas emissions (72%) being the dominant source followed by a combined source of natural gas combustion and aviation emissions (8.3%). Ozone formation potential (OFP) by VOC species was calculated using the propylene-equivalent concentration (propy-equiv) method. OFP calculated for all the apportioned sources highlighted that natural gas sources contributed significantly to the ozone formation in this region besides isoprene from biogenic sources and reactive alkenes from urban and industrial activities. This study underscores the critical role of slow-reacting n-alkanes along with other highly reactive VOC from urban and industrial sources influencing ambient air quality and it identifies strategies for mitigating ground-level ozone in urban areas affected by oil and gas extraction and production.

Pollution characteristics and probabilistic risk assessment of heavy metal(loid)s in agricultural soils across the Yellow River Basin, China

Heavy metal(loid)s (HMs) contamination in farmland soils, an environmental issue of global concern, has undergone extensive research across local and regional scales in the Yellow River Basin (YRB), China, but comprehensive studies at the basin scale remain scarce. This study, utilizing published data from 2000 to 2023, analyzed the concentrations, distribution patterns, potential sources, probabilistic ecological risks, and source-specific health risks of eight HMs (As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn) in farmland soils across the YRB. Results indicated that while most agricultural soils are safe, 21.7% of samples for Cd, 5.6% for As, and 1.2% for Zn exceed China’s risk screening levels for agricultural soil pollution, with Hg and Cd showing the highest pollution levels. Ecological risks were generally high, with Cd and Hg identified as primary risk factors. Correlation analysis and the positive matrix factorization model identified four potential origins, contributing in descending order: traffic-industrial atmospheric deposition (37.0%), mixed agricultural-natural sources (26.7%), industrial-natural sources (14.6%), and mixed agricultural-industrial sources (14.6%). The probabilistic health risk assessment identified negligible noncarcinogenic threats to the population; however, carcinogenic hazards were concerning, particularly due to industrial activities and As. Children are more vulnerable to hazards than adults. These findings enhance our insight into soil HM pollution risks in the YRB and offer a scientific foundation for targeted pollution management and control strategies at the basin scale.

Decomposition-Estimation-Reconstruction: An Automatic and Accurate Neuron Extraction Paradigm.

The extraction of spatiotemporal neuron activity from calcium imaging videos plays a crucial role in unraveling the coding properties of neurons. While existing neuron extraction approaches have shown promising results, disturbing and scattering background and unused depth still impede their performance. To address these limitations, we develop an automatic and accurate neuron extraction paradigm, dubbed as decomposition-estimation-reconstruction (DER), consisting of D-procedure, E-procedure, and R-procedure. Specifically, the D-procedure first decomposes the raw data into a low-rank background and a sparse neuron signal, and regularizes L0 -norm priors of intensity and gradient of the neuron signal to suppress blurring and artifact effects. Then, the E-procedure estimates the depth-dependent transmission of the neuron signal based on its bright and dark channel priors. The R-procedure finally integrates the depth estimation of the neuron signal as a content-importance weight into a constrained non-negative matrix decomposition framework, which facilitates accurate neuron locations to boost the quality of extracted neurons. These three procedures are coupled in a cascade manner, where the former copes with calcium imaging data to facilitate the subsequent one. Comprehensive experiments on neuron extraction from calcium imaging videos demonstrate the superiority of our DER paradigm in both qualitative results and quantitative assessments over state-of-the-art methods.

Enhancing visual seismocardiography in noisy environments with adaptive bidirectional filtering for Cardiac Health Monitoring

BackgroundWearable sensors have revolutionized cardiac health monitoring, with Seismocardiography (SCG) at the forefront due to its non-invasive nature. However, the substantial motion artefacts have hindered the translation of SCG-based medical applications, primarily induced by walking. In contrast, our innovative technique, Adaptive Bidirectional Filtering (ABF), surpasses these challenges by refining SCG signals more effectively than any motion-induced noise. ABF leverages a noise-cancellation algorithm, operating on the benefits of the Redundant Multi-Scale Wavelet Decomposition (RMWD) and the bidirectional filtering framework, to achieve optimal signal quality.MethodologyThe ABF technique is a two-stage process that diminishes the artefacts emanating from motion. The first step by RMWD is the identification of the heart-associated signals and the isolating samples with those related frequencies. Subsequently, the adaptive bidirectional filter operates in two dimensions: it uses Time-Frequency masking that eliminates temporal noise while engaging in non-negative matrix Decomposition to ensure spatial correlation and dorsoventral vibration reduction jointly. The main component that is altered from the other filters is the recursive structure that changes to the motion-adapted filter, which uses vertical axis accelerometer data to differentiate better between accurate SCG signals and motion artefacts.OutcomeOur empirical tests demonstrate exceptional signal improvement with the application of our ABF approach. The accuracy in heart rate estimation reached an impressive r-squared value of 0.95 at − 20 dB SNR, significantly outperforming the baseline value, which ranged from 0.1 to 0.85. The effectiveness of the motion-artifact-reduction methodology is also notable at an SNR of − 22 dB. Consequently, ECG inputs are not required. This method can be seamlessly integrated into noisy environments, enhancing ECG filtering, automatic beat detection, and rhythm interpretation processes, even in highly variable conditions. The ABF method effectively filters out up to 97% of motion-related noise components within the SCG signal from implantable devices. This advancement is poised to become an integral part of routine patient monitoring.

Spatial interaction and risk zoning of compound pollutants in farmland soils: Insights from heavy metals and polycyclic aromatic hydrocarbons in Hezhang County, China

The accurate identification and assessment of comprehensive risks associated with compound pollution in agricultural ecosystems remain significant challenges due to the complexity of pollution sources, soil heterogeneity, and spatial variability. In this study, bivariate local indicators of spatial association (LISA) were applied to analyze the spatial interaction between heavy metals (HMs) and polycyclic aromatic hydrocarbons (PAHs) in farmland soils in Hezhang County. The results revealed distinct clusters with elevated concentrations of both HMs and PAHs, predominantly in areas affected by long-standing lead-zinc mining and smelting activities. Positive matrix factorization (PMF) was utilized to identify mining and smelting activities, and associated coal consumption as common sources of both pollutants, contributing 53 % and 28 %, respectively. Ecological health risk assessment results indicated that the combined pollution in this area has led to particularly severe ecological and cancer risks, with the pollution coefficient (Pc) exceeding 3.0, and risk values for both adults and children surpassing the threshold of 10−4. Through the integration of advanced bivariate LISA mapping and thorough risk assessment, this study precisely delineated ecological risk zones (33.1 %) and more refined health risk zones (40.1 %) associated with combined pollution. The southwest of Hezhang was identified as a critical hotspot for combined pollution risks, primarily due to intensive mining and smelting activities in the region. Overall, this study underscores the utility of bivariate LISA as a robust approach for delineating spatial clustering patterns caused by combined pollutants. It provides crucial insights for identifying regions with heightened human health and ecological risks in rural settings.

Predicting the potential associations between circRNA and drug sensitivity using a multisource feature-based approach.

The unique non-coding RNA molecule known as circular RNA (circRNA) is distinguished from conventional linear RNA by having a longer half-life, greater degree of conservation and inherent solidity. Extensive research has demonstrated the profound impact of circRNA expression on cellular drug sensitivity and therapeutic efficacy. There is an immediate need for the creation of efficient computational techniques to anticipate the potential correlations between circRNA and drug sensitivity, as classical biological research approaches are time-consuming and costly. In this work, we introduce a novel deep learning model called SNMGCDA, which aims to forecast the relationships between circRNA and drug sensitivity. SNMGCDA incorporates a diverse range of similarity networks, enabling the derivation of feature vectors for circRNAs and drugs using three distinct calculation methods. First, we utilize a sparse autoencoder for the extraction of drug characteristics. Subsequently, the application of non-negative matrix factorization (NMF) enables the identification of relationships between circRNAs and drugs based on their shared features. Additionally, the multi-head graph attention network is employed to capture the characteristics of circRNAs. After acquiring the characteristics from these three separate components, we combine them to form a unified and inclusive feature vector for each cluster of circRNA and drug. Finally, the relevant feature vectors and labels are inputted into a multilayer perceptron (MLP) to make predictions. The outcomes of the experiment, obtained through 5-fold cross-validation (5-fold CV) and 10-fold cross-validation (10-fold CV), demonstrate SNMGCDA outperforms five other state-of-art methods in terms of performance. Additionally, the majority of case studies have predominantly confirmed newly discovered correlations by SNMGCDA, thereby emphasizing its reliability in predicting potential relationships between circRNAs and drugs.

Development and validation of a novel liquid-liquid phase separation gene signature for bladder cancer

Bladder carcinoma (BLCA) represents a common urinary tract malignancy, characterized by aggressive behavior and high recurrence rates. The biological response regulation during tumor proliferation and metastasis is intimately associated with liquid-liquid phase separation (LLPS). For the purpose of enhancing early detection and treatment, this study employed transcriptomic data to examine the prognostic implications of LLPS-associated genes and formulate a predictive model. Clinical and transcriptomic data of bladder cancer patients were sourced from the GEO and TCGA databases. This study applied a clustering algorithm using non-negative matrix factorization (NMF) to classify samples, which were systematically compared based on their liquid-liquid phase separation characteristics. Prognostic models were developed using multivariate Cox regression and the Least Absolute Shrinkage Selection Operator (LASSO) algorithm to establish risk formulas for nine genes. The gene signature’s validity was tested across the entire TCGA cohort (406 cases), the TCGA testing cohort (120 cases), and the external validation dataset GSE13507. The predictive accuracy of the signature system was assessed using receiver operating characteristic (ROC) and Kaplan-Meier curves. Additionally, decision curve analysis incorporating clinicopathological parameters and the genetic signature was employed to predict individual survival. This study identified two distinct molecular subtypes, C1 and C2. Patients with the C1 subtype exhibited significantly better prognoses than those with the C2 subtype. Low-risk group patients consistently showed superior prognoses compared to high-risk groups across the entire TCGA, GEO, and TCGA training cohorts. Furthermore, the LLPS-related gene model demonstrated prognostic value independent of other clinical traits. This study identifies LLPS-associated gene clusters and establishes an independent, accurate prognostic model for BLCA. The model holds potential for clinical application in BLCA prognosis assessment.

Reproducible Sex Differences in Personalized Functional Network Topography in Youth.

A key step towards understanding psychiatric disorders that disproportionately impact female mental health is delineating the emergence of sex-specific patterns of brain organization at the critical transition from childhood to adolescence. Prior work suggests that individual differences in the spatial organization of functional brain networks across the cortex are associated with psychopathology and differ systematically by sex. We aimed to evaluate the impact of sex on the spatial organization of person-specific functional brain networks. We leveraged person-specific atlases of functional brain networks defined using nonnegative matrix factorization in a sample of n = 6437 youths from the Adolescent Brain Cognitive Development Study. Across independent discovery and replication samples, we used generalized additive models to uncover associations between sex and the spatial layout ("topography") of personalized functional networks (PFNs). Next, we trained support vector machines to classify participants' sex from multivariate patterns of PFN topography. Finally, we leveraged transcriptomic data from the Allen Human Brain Atlas to evaluate spatial correlations between sex differences in PFN topography and gene expression. Sex differences in PFN topography were greatest in association networks including the fronto-parietal, ventral attention, and default mode networks. Machine learning models trained on participants' PFNs were able to classify participant sex with high accuracy. Brain regions with the greatest sex differences in PFN topography were enriched in expression of X-linked genes as well as genes expressed in astrocytes and excitatory neurons. Sex differences in PFN topography are robust, replicate across large-scale samples of youth, and are associated with expression patterns of X-linked genes. These results suggest a potential contributor to the female-biased risk in depressive and anxiety disorders that emerge at the transition from childhood to adolescence.

Characterization and Sources of VOCs during PM2.5 Pollution Periods in a Typical City of the Yangtze River Delta

To investigate the characteristics and sources of volatile organic compounds (VOCs) as well as their impacts on secondary organic aerosols (SOAs) formation during high-incidence periods of PM2.5 pollution, a field measurement was conducted in December 2019 in Hefei, a typical city of the Yangtze River Delta (YRD). During the whole process, the mixing ratios of VOCs were averaged as 21.1 ± 15.9 ppb, with alkanes, alkenes, alkyne, and aromatics accounting for 59.9%, 15.3%, 15.0%, and 9.8% of the total VOCs, respectively. It is worth noting that the contributions of alkenes and alkyne increased significantly during PM2.5 pollution periods. Based on source apportionment via the positive matrix factorization (PMF) model, vehicle emissions, liquefied petroleum gas/natural gas (LPG/NG), and biomass/coal burning were the main sources of VOCs during the research in Hefei. During pollution periods, however, the contribution of biomass/coal burning to VOCs increased significantly, reaching as much as 47.6%. The calculated SOA formation potential (SOAFP) of VOCs was 0.38 ± 1.04 µg m−3 (range: 0.04–7.30 µg m−3), and aromatics were the dominant contributors, with a percentage of 96.8%. The source contributions showed that industrial emissions (49.1%) and vehicle emissions (28.3%) contributed the most to SOAFP during non-pollution periods, whereas the contribution of biomass/coal burning to SOA formation increased significantly (32.8%) during PM2.5 pollution periods. These findings suggest that reducing VOCs emissions from biomass/coal burning, vehicle, and industrial sources is a crucial approach for the effective control of SOA formation in Hefei, which provides a scientific basis for controlling PM2.5 pollution and improving air quality in the YRD region.

Source apportionment of Volatile Organic Compounds (VOCs) in the South Coast Air Basin (SoCAB) During RECAP-CA

Ozone (O3) concentrations in the South Coast Air Basin (SoCAB) surrounding Los Angeles remain at unhealthy levels despite multiple decades of control programs designed to reduce emissions of precursor Volatile Organic Compounds (VOCs). Here we report on comprehensive VOC measurements made at Redlands, which has the highest measured O3 concentrations in SoCAB, as part of the Re-Evaluating the Chemistry of Air Pollutants in California (RECAP-CA) field campaign (July–October 2021). Positive matrix factorization (PMF) analysis was applied to identify nine VOC factors. A photochemical chamber model initialized by field measurements was configured with a tagging technique to quantify the VOC factor contributions to O3 formation in Redlands. Biogenic VOCs (BVOCs) made the largest contribution (26.6%) to O3 formation, followed by traffic VOCs (21.2%), volatile chemical products (VCPs) (19%), and plant decomposition (14.9%). High O3 episodes were not driven by increased VOC emissions from any single source, but rather were associated with stagnation events that concentrated VOCs from all sources and high temperature days that enhanced O3 formation efficiency. This implies that VOC controls optimized to reduce O3 concentrations would look similar in both the NOx-limited and VOC-limited regimes that can occur at Redlands. These results suggest that control strategies that reduce VOC and NOx emissions from the on-road vehicle fleet, such as increasing electrification, may yield O3 reductions on days in both the NOx-limited and VOC-limited chemical regimes at Redlands.

Exploring copper metabolism-induced cell death in gastric cancer: a single-cell RNA sequencing study and prognostic model development

BackgroundGastric cancer (GC) is the third leading cause of cancer-related deaths globally. Despite advancements in treatment, the overall 5-year survival rate remains below 30%, particularly in advanced stages. Copper metabolism, vital for various cellular processes, has been linked to cancer progression, but its role in GC, especially at the single-cell level, is not well understood.ObjectiveThis study aims to investigate copper metabolism in GC by integrating single-cell RNA sequencing (scRNA-seq) data and developing a prognostic model based on copper metabolism-related gene (CMRG) expression. The study explores how copper metabolism affects the tumor microenvironment and identifies potential therapeutic targets.MethodsscRNA-seq data from gastric cancer and normal tissues were analyzed using the Seurat package. Principal Component Analysis (PCA) and Uniform Manifold Approximation and Projection (UMAP) were used for dimensionality reduction and clustering. Non-negative matrix factorization (NMF) was employed for T cell subpopulation analysis. A high-dimensional weighted gene co-expression network analysis (HdWGCNA) identified key molecular features. LASSO regression and Random Survival Forest (RSF) techniques were used to create and validate a prognostic model. Survival analysis, immune microenvironment assessment, and drug sensitivity analysis were conducted.ResultsSixteen cell clusters and nine distinct cell types were identified, with T cells showing significant roles in cell communication. The NMF analysis of CD8 +T cells revealed five copper metabolism-related subtypes. The prognostic model based on nine CMRGs indicated significant survival differences between high- and low-risk groups. High-risk patients showed shorter survival times, increased immune cell infiltration, and altered immune responses. Drug sensitivity analysis suggested higher efficacy of certain drugs in high-CMRG patients.

Source-risk and uncertainty assessment of trace metals in surface sediments of a human-dominated seaward catchment in eastern China

Current total concentration-based methods for source attribution and risk assessment often overestimate metal risks, thereby impeding the formulation of effective risk management strategies. This study aims to develop a framework for source-specific risk assessment based on metal bioavailability in surface river sediments from a human-dominated seaward catchment in eastern China. Metal bioavailability was quantified using chemical fractionation results, and source apportionment was conducted using the positive matrix factorization (PMF) model. Risk assessment integrated these findings using two indices: the Potential Ecological Risk Index (PERI) and the Mean Probable Effect Concentration Quotient (mPEC-Q), with uncertainty addressed via Monte Carlo simulations. Results indicated that average total concentrations of Cu, Pb, Zn, Cr, Hg, Cd, and As exceeded their respective background levels by 1.63 to 15.00 times. The residual fraction constituted the majority, accounting for 53.84 % to 77.79 % of total concentrations, suggesting significant natural origins. However, source apportionment revealed a predominant contribution from anthropogenic activities, including industrial smelting, agricultural practices, and atmospheric deposition. The contributions were found to vary between 5.35 % and 40.03 % when the total concentration was adjusted to bioavailable content. Total concentration-based PERI/mPEC-Q assessments indicated high/moderate risk levels, decreasing to considerable/low risk levels with bioavailability adjustment. Hg and Cd were identified as priority metals. Further incorporating source appointment parameters into the risk assessment, industrial smelting was identified as the primary contributor, accounting for 66.06 % of total risk by total concentration and 65.63 % by bioavailability. This underscores the role of bioavailability in mitigating risk overestimation. Monte Carlo simulations validated industrial smelting as a major risk contributor. This study emphasizes the importance of considering bioavailability in the source-risk assessment of sediment-metals, crucial for targeted risk management in urbanized catchment areas.

Machine learning-driven assessment of heavy metal contamination in the impounded lakes of China's South‐to‐North Water Diversion Project: Identifying spatiotemporal patterns and ecological risks

The Eastern Route of China's South-to-North Water Diversion Project (SNWDP-ER) traverses through impounded lakes that are potentially vulnerable to heavy metals (HMs) contamination although the understanding remains elusive. This study employed machine learning approaches, including super-clustering of Self-Organizing Map (SOM) and Robust Principal Component Analysis (RPCA), to elucidate the spatiotemporal patterns and assess ecological risks associated with HMs in the surface sediments of Gao-Bao-Shaobo Lake (GBSL) and Dongping Lake (DPL). We collected 184 surface sediments from 47 stations across the two important impounded lakes over four seasons. The results revealed higher HMs concentrations in the south-central GBSL and west-central DPL, with a notable increase in contamination in autumn. The comprehensive risk assessment, utilizing various indicators such as the Sediment Quality Guidelines (SQGs), Improved Potential Ecological Risk Index (IPERI), Geo-accumulation Index (Igeo), Contamination Factor (CF), and Enrichment Factor (EF), identified arsenic (As), cadmium (Cd), nickel (Ni), and chromium (Cr) as primary contaminants of concern. Positive Matrix Factorization (PMF) model, coupled with Spearman analysis, attributed over 70 % of HMs pollution to anthropogenic activities. This research provides a nuanced understanding of HMs pollution in the context of large-scale water diversion projects and offers a scientific basis for targeted pollution mitigation strategies.

Chemical composition, mutagenicity, and cytotoxicity of urban submicron particulate matter (PM1) in Agra, India

This study, conducted in Agra, India, examined the mass concentrations, chemical compositions, and seasonal variations of submicron particles (PM1). The concentrations of metals, water-soluble inorganic ions (WSIs) including anions (F−, Cl−, NO₃−, SO₄2−) and cations (Ca2+, K+, Mg2+, NH₄+, Na+), organic carbon (OC) and elemental carbon (EC) and polycyclic aromatic hydrocarbons (PAHs) in PM1 extract were determined using Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES), Ion Chromatography, Thermogravimetric Analysis and Gas Chromatography-Mass Spectrometry (GC–MS) respectively. For morphological observation of PM1 particles, Field Emission Scanning Electron Microscopy-Energy Dispersive X-ray spectrometry (FESEM-EDS) was used. The annual average concentration of PM1 was 82.9 ± 33.4 μg/m3, which exceeds the World Health Organisation's (WHO) safe limit for PM2.5 of 5 μg/m3 by a factor of 17. The PM1 mass composition included metals (31 %), WSIs (28 %), OC and EC (9.8 %), and PAHs (0.4 %). Winter recorded the highest PM1 concentration (96.1 ± 25.8 μg/m3), followed by post-monsoon, summer, and monsoon seasons. The average concentration of PAHs was 364.6 ± 226.6 ng/m3. Positive Matrix Factorization (PMF) identified traffic, emissions from biomass/coal and wood combustion, industrial/stationary sources, and secondary aerosols as potential contributors. The Ames test revealed the presence of frameshift mutations and base pair substitutions, especially in winter and post-monsoon. Additionally, PM1 exhibited cytotoxic effects on V-79 cells, with heightened toxicity during winter and prolonged exposure in other seasons. This study underscores the urgent need to address local emission sources and establish regulatory standards for PM1 in urban areas.

Multi-view clustering based on pairwise co-regularization and robust dual graph non-negative matrix factorization

Multi-view clustering has been a hotspot in the field of machine learning and pattern recognition, and methods based on non-negative matrix factorization have gained attention for their simplicity and interpretability. Despite these methods achieving great clustering performance, there may also be some limitations, such as the full structural information of data and the similarity between different views not being considered. This paper proposes a novel multi-view clustering algorithm based on pairwise co-regularization and robust dual graph non-negative matrix factorization. Firstly, the l2,p-norm is applied in non-negative matrix factorization framework to enhance model robustness. Next, pairwise co-regularization is utilized to extract the inter-view information of all views. Then, graph dual regularization is applied to preserve the structure information of the data and feature spaces. Lastly, an auto-weighted strategy is introduced to assign appropriate weights to each view. In addition, an iterative updating optimization scheme for the proposed algorithm is developed, and the convergence proof of the scheme is provided. The experimental results on twelve real-world datasets show that the proposed algorithm is both effective and efficient, compared with three classical and six state-of-the-art algorithms.

Status, Sources, and Risks of Heavy Metals in Surface Sediments of Baiyangdian Lake and Inflow Rivers, North China

Baiyangdian Lake, recognized as the largest freshwater body in northern China, plays a vital role in maintaining the regional eco-environment. Prior studies have pointed out the contamination of sediments with heavy metals, raising concerns about eco-environmental challenges. Therefore, it is imperative to evaluate the current pollution levels and ecological threats related to heavy metals found in the sediments of Baiyangdian Lake as well as in its inflow rivers. In May 2022, surface sediments with a depth of less than 20 cm were analyzed for Cu, Zn, Pb, Cr, Ni, As, Cd, and Hg to determine the pollution status, identify sources of pollution, and evaluate potential ecological risks. A range of evaluation methods used by predecessors such as geo-accumulation index (Igeo), enrichment factor (EF), ecological risk index (RI), sediment quality guidelines (SQGs), positive matrix factorization (PMF), absolute principal component score-multiple linear regression model (APCS-MLR), chemical mass balance (CMB), and UNMIX model were analyzed. After comparison, multi-methods including the geo-accumulation index (Igeo), absolute principal component score-multiple linear regression model (APCS-MLR), ecological risk index (RI), and sediment quality guidelines (SQGs) were utilized this time, leading to a better result. Findings reveal that pollution levels are generally low or non-existent, with only 1.64% of sampling sites showing close to moderate pollution levels for Cu, Pb, and Zn, and 4.92% and 1.64% of sites exhibiting close to moderate and moderate pollution levels for Cd, respectively. The main contributors to heavy metal presence are pinpointed as industrial wastewater discharge, particularly Cu, Zn, Pb, Cd, and Hg. The ecological risks are also relatively low, with 4.92%, 1.64%, and 1.64% of sampling sites demonstrating close to moderate, moderate, and strong risks in the inflow rivers, respectively. Additionally, only one site shows moderate potential biological toxicity, while the rest display non-toxicity. These findings will update our cognition and offer a scientific basis for pollution treatment and ecosystem enhancement for government management.