Myocardial infarction detection method based on the continuous T-wave area feature and multi-lead-fusion deep features.

OBJECTIVE: Myocardial infarction (MI) is one of the most threatening cardiovascular diseases. This paper aims to explore a method for using an algorithm to autonomously classify myocardial infarction based on the electrocardiogram (ECG). APPROACH: A detection method of MI that fuses continuous T-wave area (C_TWA) feature and ECG deep features is proposed. This method consists of three main parts: (1) The onset of MI is often accompanied by changes in the shape of the T-wave in the ECG, thus the area of the T-wave displayed on different heartbeats will be quite different. The adaptive sliding window method is used to detect the start and end of the T-wave, and calculate the C_TWA on the same ECG record. Additionally, the coefficient of variation (CV) of C_TWA is defined as the C_TWA feature of the ECG. (2) The multi lead fusion convolutional neural network (Multi-lead-fusion CNN) was implemented to extract the deep features of the ECG. (3) The C_TWA feature and deep features of the ECG were fused by soft attention, and then inputted into the multi-layer perceptron to obtain the detection result. RESULTS: According to the interpatient paradigm, the proposed method reached a 97.67% accuracy, 96.59% precision, and 98.96% recall on the PTB dataset, whereas the proposed method reached 93.15% accuracy, 93.20% precision, and 95.14% recall on the clinical dataset. SIGNIFICANCE: The proposed method accurately extracts the feature of the C_TWA, and combines the deep features of the signal, thereby improving the detection accuracy and achieving ideal results on clinical datasets.

Regulatory networks of senescence-associated gene-transcription factors promote degradation in Moso bamboo shoots.

Bamboo cultivation, particularly Moso bamboo (Phyllostachys edulis), holds significant economic importance in various regions worldwide. Bamboo shoot degradation (BSD) severely affects productivity and economic viability. However, despite its agricultural consequences, the molecular mechanisms underlying BSD remain unclear. Consequently, we explored the dynamic changes of BSD through anatomy, physiology and the transcriptome. Our findings reveal ruptured protoxylem cells, reduced cell wall thickness and the accumulation of sucrose and reactive oxygen species (ROS) during BSD. Transcriptomic analysis underscored the importance of genes related to plant hormone signal transduction, sugar metabolism and ROS homoeostasis in this process. Furthermore, BSD appears to be driven by the coexpression regulatory network of senescence-associated gene transcription factors (SAG-TFs), specifically PeSAG39, PeWRKY22 and PeWRKY75, primarily located in the protoxylem of vascular bundles. Yeast one-hybrid and dual-luciferase assays demonstrated that PeWRKY22 and PeWRKY75 activate PeSAG39 expression by binding to its promoter. This study advanced our understanding of the molecular regulatory mechanisms governing BSD, offering a valuable reference for enhancing Moso bamboo forest productivity.

TransC-ac4C: Identification of N4-acetylcytidine (ac4C) sites in mRNA using deep learning.

N4-acetylcytidine (ac4C) is a post-transcriptional modification in mRNA that is critical in mRNA translation in terms of stability and regulation. In the past few years, numerous approaches employing convolutional neural networks (CNN) and Transformer have been proposed for the identification of ac4C sites, with each variety of approaches processing distinct characteristics. CNN-based methods excels at extracting local features and positional information, whereas Transformer-based ones stands out in establishing long-range dependencies and generating global representations. Given the importance of both local and global features in mRNA ac4C sites identification, we propose a novel method termed TransC-ac4C which combines CNN and Transformer together for enhancing the feature extraction capability and improving the identification accuracy. Five different feature encoding strategies (One-hot, NCP, ND, EIIP, and K-mer) are employed to generate the mRNA sequence representations, in which way the sequence attributes and physical and chemical properties of the sequences can be embedded. To strengthen the relevance of features, we construct a novel feature fusion method. Firstly, the CNN is employed to process five single features, stitch them together and feed them to the Transformer layer. Then, our approach employs CNN to extract local features and Transformer subsequently to establish global long-range dependencies among extracted features. We use 5-fold cross-validation to evaluate the model, and the evaluation indicators are significantly improved. The prediction accuracy of the two datasets is as high as 81.42.

Effects of returning paddy field to wetland on composition and stability of soil aggregates in the Yellow River Delta.

The composition and stability of soil aggregates are important indicators for measuring soil quality, which would be affected by land use changes. Taking wetlands with different returning years (2 and 15 years) in the Yellow River Delta as the research object, paddy fields and natural wetlands as control, we analyzed the changes in soil physicochemical properties and soil aggregate composition. The results showed that soil water content, total organic carbon, dissolved organic carbon and total phosphorus of the returning soil (0-40 cm) showed an overall increasing trend with returning period, while soil pH and bulk density was in adverse. There was no significant change in clay content, electrical conductivity, and total nitrogen content. The contents of macro-aggregates and micro-aggregates showed overall increasing and decreasing trend with returning period, respectively. The stability of aggregates in the topsoil (0-10 cm) increased with returning years. Geometric mean diameter and mean weight diameter increased by 8.9% and 40.4% in the 15th year of returning, respectively, while the mass proportion of >2.5 mm fraction decreased by 10.5%. There was no effect of returning on aggregates in subsoil (10-40 cm). Our results indicated that returning paddy field to wetland in the Yellow River Delta would play a positive role in improving soil structure and aggregate stability.

GraphormerDTI: A graph transformer-based approach for drug-target interaction prediction.

The application of Artificial Intelligence (AI) to screen drug molecules with potential therapeutic effects has revolutionized the drug discovery process, with significantly lower economic cost and time consumption than the traditional drug discovery pipeline. With the great power of AI, it is possible to rapidly search the vast chemical space for potential drug-target interactions (DTIs) between candidate drug molecules and disease protein targets. However, only a small proportion of molecules have labelled DTIs, consequently limiting the performance of AI-based drug screening. To solve this problem, a machine learning-based approach with great ability to generalize DTI prediction across molecules is desirable. Many existing machine learning approaches for DTI identification failed to exploit the full information with respect to the topological structures of candidate molecules. To develop a better approach for DTI prediction, we propose GraphormerDTI, which employs the powerful Graph Transformer neural network to model molecular structures. GraphormerDTI embeds molecular graphs into vector-format representations through iterative Transformer-based message passing, which encodes molecules' structural characteristics by node centrality encoding, node spatial encoding and edge encoding. With a strong structural inductive bias, the proposed GraphormerDTI approach can effectively infer informative representations for out-of-sample molecules and as such, it is capable of predicting DTIs across molecules with an exceptional performance. GraphormerDTI integrates the Graph Transformer neural network with a 1-dimensional Convolutional Neural Network (1D-CNN) to extract the drugs' and target proteins' representations and leverages an attention mechanism to model the interactions between them. To examine GraphormerDTI's performance for DTI prediction, we conduct experiments on three benchmark datasets, where GraphormerDTI achieves a superior performance than five state-of-the-art baselines for out-of-molecule DTI prediction, including GNN-CPI, GNN-PT, DeepEmbedding-DTI, MolTrans and HyperAttentionDTI, and is on a par with the best baseline for transductive DTI prediction. The source codes and datasets are publicly accessible at https://github.com/mengmeng34/GraphormerDTI.

Algae removal and algal organic matter chemistry modulated by KMnO4-PAC in simulated karst water.

In this study, we examined the modulation of algae removal and algal organic matter (AOM) chemistry by potassium permanganate and poly-aluminum chloride (KMnO4-PAC) in simulated karst water. Specifically, we verified the compositional changes of AOM sourcing from Chlorella sp. and Pseudanabaena sp. in response to the presence of divalent ions (Ca2+ and Mg2+). Aromatic protein and soluble microbial products were identified as the primary AOM components. Divalent ions accelerated dissolved organic carbon (DOC) and UV254 removal, particularly with Pseudanabaena sp. greater than Chlorella sp. (P < 0.05). Surface morphology analysis manifested that the removal of filamentous Pseudanabaena sp. was more feasible in comparison to globular Chlorella sp.. Our results highlight the significance of divalent ions in governing chemical behaviors and subsequent removal of both algae and AOM. This study upscales the understanding of the interactions among divalent ions, algae and AOM during preoxidation and coagulation process in algae-laden karst water.

Effects of tidal hydrology on soil phosphorus forms in the Yellow River estuary wetland: A field study of soil core translocation.

Phosphorus (P) forms in soil are related to the P cycle and play an important role in maintaining the productivity and function of wetlands. Tidal hydrology is a key factor controlling soil P forms in estuary wetlands; however, the response of soil P forms to tidal hydrological changes remains unclear. A translocation experiment in the Yellow River Estuary wetland was conducted to study the effect of hydrological changes on P forms in the soil, in which freshwater marsh soils in the supratidal zone were translocated to salt marshes in different intertidal zones (up-high-tidal zone, high-tidal zone, and middle-tidal zone). Over a 23-month experiment, soil properties showed varying changes under different tidal hydrology conditions, with an increase in pH, salinity, Ca2+ and salt ions and a decrease in iron oxide and nutrients. Compared with the control, the content of different forms of phosphorus (total phosphorus, inorganic phosphorus, organic phosphorus, and calcium-bound phosphorus) in the cultured soil cores decreased from 3.3 % to 67.0 % in the intertidal zones, whereas the content of ferrum­aluminum-bound phosphorus increased from 58.9 % to 65.1 % at the end of the experiment. According to the partial least squares structural equation model, P forms are influenced by tidal hydrology mainly through the mediation of salt ions and nutrient levels. These results suggest that seawater intrusion promotes the release of P in the supratidal zone soil of estuary wetlands.

Effect of oxide interactions on chromium speciation transformation during simulated municipal solid waste incineration.

Chromium released during municipal solid waste incineration (MSWI) is toxic and carcinogenic. The removal of chromium from simulated MSWI flue gas by four sorbents (CaO, bamboo charcoal (BC), powdered activated carbon (PAC), and Al2O3) and the effects of four oxides (SiO2, Al2O3, Fe2O3, and CaO) on chromium speciation transformation were investigated. The results showed that the removal rates of total Cr by the four sorbents were Al2O3 < CaO < PAC < BC, while the removal rates of Cr(VI) by the four sorbents were Al2O3 < PAC < BC < CaO. CaO had a strong oxidizing effect on Cr(III), while BC and PAC had a better-reducing effect on Cr(VI). SiO2 was better for the reduction of Na2CrO4 and K2CrO4 above 1000°C due to its strong acidity, and the addition of CaO significantly inhibited the reduction of Cr(VI). MgCrO4 decomposed above 700°C to form MgCr2O4, and the reaction between MgCrO4 and oxides also existed in the form of a more stable trivalent spinel. Furthermore, when investigating the effect of oxides on the oxidation of Cr(III) in CrCl3, it was discovered that CaO promoted the conversion of Cr(III) to Cr(VI), while the presence of chlorine caused chromium to exist in the form of Cr(V), and increasing the content of CaO and extending the heating time facilitated the oxidation of Cr(III). In addition, silicate, aluminate, and ferrite were generated after the addition of SiO2, Al2O3, and Fe2O3, which reduced the alkalinity of CaO and had an important role in inhibiting the oxidation of Cr(III). The acidic oxides can not only promote the reduction of Cr(VI) but also have an inhibitory effect on the oxidation of Cr(III) ascribed to alkali metals/alkaline earth metals, and the proportion of acidic oxides can be increased moderately to reduce the generation of harmful substances in the hazardous solid waste heat treatment.

Cone-beam computed tomographic evaluation of mandibular incisor alveolar bone changes for the intrusion arch technique: A retrospective cohort research.

Objective: : Alveolar bone loss is a common adverse effect of intrusion treatment. Mandibular incisors are prone to dehiscence and fenestrations as they suffer from thinner alveolar bone thickness. Methods: : Thirty skeletal class II patients treated with mandibular intrusion arch therapy were included in this study. Lateral cephalograms and cone-beam computed tomography images were taken before treatment (T1) and immediately after intrusion arch removal (T2) to evaluate the tooth displacement and the alveolar bone changes. Pearson's and Spearman's correlation was used to identify risk factors of alveolar bone loss during the intrusion treatment. Results: : Deep overbite was successfully corrected (P < 0.05), accompanied by mandibular incisor proclination (P < 0.05). There were no statistically significant change in the true incisor intrusion (P > 0.05). The labial and lingual vertical alveolar bone levels showed a significant decrease (P < 0.05). The alveolar bone is thinning in the labial crestal area and lingual apical area (P < 0.05); accompanied by thickening in the labial apical area (P < 0.05). Proclined incisors, non-extraction treatment, and increased A point-nasion-B point (ANB) degree were positively correlated with alveolar bone loss. Conclusions: : While the mandibular intrusion arch effectively corrected the deep overbite, it did cause some unwanted incisor labial tipping/flaring. During the intrusion treatment, the alveolar bone underwent corresponding changes, which was thinning in the labial crestal area and thickening in the labial apical area vice versa. And increased axis change of incisors, non-extraction treatment, and increased ANB were identified as risk factors for alveolar bone loss in patients with mandibular intrusion therapy.

Improving Cancer Survival Prediction via Graph Convolutional Neural Network Learning on Protein-Protein Interaction Networks.

Cancer is one of the most challenging health problems worldwide. Accurate cancer survival prediction is vital for clinical decision making. Many deep learning methods have been proposed to understand the association between patients' genomic features and survival time. In most cases, the gene expression matrix is fed directly to the deep learning model. However, this approach completely ignores the interactions between biomolecules, and the resulting models can only learn the expression levels of genes to predict patient survival. In essence, the interaction between biomolecules is the key to determining the direction and function of biological processes. Proteins are the building blocks and principal undertakings of life activities, and as such, their complex interaction network is potentially informative for deep learning methods. Therefore, a more reliable approach is to have the neural network learn both gene expression data and protein interaction networks. We propose a new computational approach, termed CRESCENT, which is a protein-protein interaction (PPI) prior knowledge graph-based convolutional neural network (GCN) to improve cancer survival prediction. CRESCENT relies on the gene expression networks rather than gene expression levels to predict patient survival. The performance of CRESCENT is evaluated on a large-scale pan-cancer dataset consisting of 5991 patients from 16 different types of cancers. Extensive benchmarking experiments demonstrate that our proposed method is competitive in terms of the evaluation metric of the time-dependent concordance index( Ctd) when compared with several existing state-of-the-art approaches. Experiments also show that incorporating the network structure between genomic features effectively improves cancer survival prediction.

PmxPred: A data-driven approach for the identification of active polymyxin analogues against gram-negative bacteria.

The multidrug-resistant Gram-negative bacteria has evolved into a worldwide threat to human health; over recent decades, polymyxins have re-emerged in clinical practice due to their high activity against multidrug-resistant bacteria. Nevertheless, the nephrotoxicity and neurotoxicity of polymyxins seriously hinder their practical use in the clinic. Based on the quantitative structure-activity relationship (QSAR), analogue design is an efficient strategy for discovering biologically active compounds with fewer adverse effects. To accelerate the polymyxin analogues discovery process and find the polymyxin analogues with high antimicrobial activity against Gram-negative bacteria, here we developed PmxPred, a GCN and catBoost-based machine learning framework. The RDKit descriptors were used for the molecule and residues representation, and the ensemble learning model was utilized for the antimicrobial activity prediction. This framework was trained and evaluated on multiple Gram-negative bacteria datasets, including Acinetobacter baumannii, Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa and a general Gram-negative bacteria dataset achieving an AUROC of 0.857, 0.880, 0.756, 0.895 and 0.865 on the independent test, respectively. PmxPred outperformed the transfer learning method that trained on 10 million molecules. We interpreted our model well-trained model by analysing the importance of global and residue features. Overall, PmxPred provides a powerful additional tool for predicting active polymyxin analogues, and holds the potential elucidate the mechanisms underlying the antimicrobial activity of polymyxins. The source code is publicly available on GitHub (https://github.com/yanwu20/PmxPred).

Interpretable prediction models for widespread m6A RNA modification across cell lines and tissues.

MOTIVATION: RNA N6-methyladenosine (m6A) in Homo sapiens plays vital roles in a variety of biological functions. Precise identification of m6A modifications is thus essential to elucidation of their biological functions and underlying molecular-level mechanisms. Currently available high-throughput single-nucleotide-resolution m6A modification data considerably accelerated the identification of RNA modification sites through the development of data-driven computational methods. Nevertheless, existing methods have limitations in terms of the coverage of single-nucleotide-resolution cell lines and have poor capability in model interpretations, thereby having limited applicability. RESULTS: In this study, we present CLSM6A, comprising a set of deep learning-based models designed for predicting single-nucleotide-resolution m6A RNA modification sites across eight different cell lines and three tissues. Extensive benchmarking experiments are conducted on well-curated datasets and accordingly, CLSM6A achieves superior performance than current state-of-the-art methods. Furthermore, CLSM6A is capable of interpreting the prediction decision-making process by excavating critical motifs activated by filters and pinpointing highly concerned positions in both forward and backward propagations. CLSM6A exhibits better portability on similar cross-cell line/tissue datasets, reveals a strong association between highly activated motifs and high-impact motifs, and demonstrates complementary attributes of different interpretation strategies. AVAILABILITY AND IMPLEMENTATION: The webserver is available at http://csbio.njust.edu.cn/bioinf/clsm6a. The datasets and code are available at https://github.com/zhangying-njust/CLSM6A/.

SMG: self-supervised masked graph learning for cancer gene identification.

Cancer genomics is dedicated to elucidating the genes and pathways that contribute to cancer progression and development. Identifying cancer genes (CGs) associated with the initiation and progression of cancer is critical for characterization of molecular-level mechanism in cancer research. In recent years, the growing availability of high-throughput molecular data and advancements in deep learning technologies has enabled the modelling of complex interactions and topological information within genomic data. Nevertheless, because of the limited labelled data, pinpointing CGs from a multitude of potential mutations remains an exceptionally challenging task. To address this, we propose a novel deep learning framework, termed self-supervised masked graph learning (SMG), which comprises SMG reconstruction (pretext task) and task-specific fine-tuning (downstream task). In the pretext task, the nodes of multi-omic featured protein-protein interaction (PPI) networks are randomly substituted with a defined mask token. The PPI networks are then reconstructed using the graph neural network (GNN)-based autoencoder, which explores the node correlations in a self-prediction manner. In the downstream tasks, the pre-trained GNN encoder embeds the input networks into feature graphs, whereas a task-specific layer proceeds with the final prediction. To assess the performance of the proposed SMG method, benchmarking experiments are performed on three node-level tasks (identification of CGs, essential genes and healthy driver genes) and one graph-level task (identification of disease subnetwork) across eight PPI networks. Benchmarking experiments and performance comparison with existing state-of-the-art methods demonstrate the superiority of SMG on multi-omic feature engineering.

Effect of electron acceptor addition on the temperature sensitivity of soil anaerobic carbon mineralization in the Yellow River Estuary wetland, China.

The temperature sensitivity of soil carbon mineralization (Q10) is an important index to evaluate the responses of ecosystem carbon cycling to climate change. We examined the effects of three electron acceptors [SO42-, NO3- and Fe(Ⅲ)] addition on the Q10 value of anaerobic carbon mineralization of Phragmites australis community soil (0-10 cm) in the Yellow River Estuary wetland with the closed culture-gas chromatography method. The results showed that the three electron acceptors addition inhibited the production of CO2 and CH4 during the 48-day culture period, with a decrease of 17.3%-20.8% for CO2 and 29.2%-36.2% for CH4. Generally, the CO2 production differed with the concentrations of electron acceptors, while CH4 production differed with the type of electron acceptors. The CO2:CH4 ratios were significantly different with temperature, indicating an obvious temperature dependence for the anaerobic carbon mineralization pathway. The Q10 values of CO2 and CH4 production under three electron acceptor additions ranged from 1.08 to 1.11 and from 1.19 to 1.37, respectively, showing an increasing trend compared with the control. The type and concentration of electron acceptors affected the temperature dependence of CO2 production, while electron acceptors affected that of CH4 production. It is suggested that the input of reducing salts would retard the mineralization loss of organic carbon in estuary freshwater wetlands under the background of climate change, but enhance the sensitivity of carbon mineralization to increasing temperature.

DeepSecE: A Deep-Learning-Based Framework for Multiclass Prediction of Secreted Proteins in Gram-Negative Bacteria.

Proteins secreted by Gram-negative bacteria are tightly linked to the virulence and adaptability of these microbes to environmental changes. Accurate identification of such secreted proteins can facilitate the investigations of infections and diseases caused by these bacterial pathogens. However, current bioinformatic methods for predicting bacterial secreted substrate proteins have limited computational efficiency and application scope on a genome-wide scale. Here, we propose a novel deep-learning-based framework-DeepSecE-for the simultaneous inference of multiple distinct groups of secreted proteins produced by Gram-negative bacteria. DeepSecE remarkably improves their classification from nonsecreted proteins using a pretrained protein language model and transformer, achieving a macro-average accuracy of 0.883 on 5-fold cross-validation. Performance benchmarking suggests that DeepSecE achieves competitive performance with the state-of-the-art binary predictors specialized for individual types of secreted substrates. The attention mechanism corroborates salient patterns and motifs at the N or C termini of the protein sequences. Using this pipeline, we further investigate the genome-wide prediction of novel secreted proteins and their taxonomic distribution across ~1,000 Gram-negative bacterial genomes. The present analysis demonstrates that DeepSecE has major potential for the discovery of disease-associated secreted proteins in a diverse range of Gram-negative bacteria. An online web server of DeepSecE is also publicly available to predict and explore various secreted substrate proteins via the input of bacterial genome sequences.

Absolute and relative preferences for mobile phone internet content, mobile phone dependence, and depressive symptoms: a study of Chinese university students in the post-pandemic era.

Background: Due to the COVID-19 pandemic, Chinese university students may have increased mobile phone dependence, a habitual behavior in the student population and a risk factor for depressive symptoms. Therefore, this study explored the association between mobile phone dependence and depressive symptoms in Chinese university students in the post-pandemic era. It also investigated the effects of different types and categories of mobile phone Internet content preferences. In particular, this study examined whether mobile phone dependence mediates the relationship between absolute preference (AP) for mobile phone Internet content and depressive symptoms, and whether relative preference (RP) for mobile phone Internet content moderates the association between mobile phone dependence and depressive symptoms. Methods: A cross-sectional study of Chinese university students recruited through Credamo was conducted in February-March, 2023. Participants completed the Mobile Phone Internet Content Preference Questionnaire, Self-rating Questionnaire for Adolescent Problematic Mobile Phone Use, and Center for Epidemiological Survey, Depression Scale. Statistical analyses included descriptive statistics, correlation, regression, and analyses of mediation and moderation effects. The final sample comprised 1,602 students (709 males). Results: We found a positive association between mobile phone dependence and depressive symptoms. The mediating role of mobile phone dependence between AP for mobile phone Internet content and depressive symptoms differed according to the type and category of content. Meanwhile, different types and categories of RP for mobile phone Internet content moderated the association between mobile phone dependence and depressive symptoms in opposite directions. Conclusion: Our results highlight the interrelationships among mobile phone Internet content preferences, mobile phone dependence, and depressive symptoms in Chinese university students. For different types and categories of mobile phone Internet content preferences, we propose distinct preventive measures to alleviate students' depressive symptoms.

Effects of shoe collar types on ankle and knee biomechanics characteristics when performing the side-step cutting task.

This study aimed to examine the effect of football shoes with different collar types on ankle and knee kinematic and kinetics features during 45° and 135° side-step cutting tasks. Fifteen healthy college football players volunteered for the study. Each participant was instructed to perform side-step cutting tasks with high, low, and no collar football shoes. The kinematic and ground reaction force data were measured using a Vicon motion capture system and a Kistler force plate, respectively. Two-way MANOVAs with repeated measures were used to examine the effect of shoe collar type and task conditions. There were no interaction effects. The high collar football shoe showed decreased ankle range of motion in the sagittal plane (p = 0.010) and peak ankle external rotation moment (p = 0.009) compared to the no collar football shoe. The high (p = 0.025) and low (p = 0.029) collar football shoes presented greater peak ankle external rotation angles than the no collar football shoe. These results imply that football shoes with high collars made of high intensity knitted fabric could be used to restrict ankle joint movement, with potential implications for decreasing the risk of ankle sprain injuries in football players.

[Effects of Water-salt Environment on Freshwater Wetland Soil C, N, and P Ecological Stoichiometric Characteristics in the Yellow River Estuary Wetland].

Carbon (C), nitrogen (N), and phosphorus (P) are important nutrients, and their ecological stoichiometric characteristics can reflect the quality and fertility capacity of soil, which is critical to understanding the stable mechanisms of estuarine wetland ecosystems. Under global changes, the increase in salinity and flooding caused by sea level rise will lead to changes in biogeochemical processes in estuarine wetlands, which is expected to affect the ecological stoichiometric characteristics of soil C, N, and P and ultimately interfere with the stability of wetland ecosystems. However, it remains unclear how the C, N, and P ecological stoichiometric characteristics respond to the water-salt environment in estuarine wetlands. We differentiated changes in the C, N, and P ecological stoichiometric characteristics through an ex-situ culture experiment for 23 months in the Yellow River Estuary Wetland. The five sites with distinct tidal hydrology were selected to manipulate translocation of soil cores from the freshwater marsh to high-, middle-, and low-tidal flats in June 2019. The results showed that soil water content (SWC); electrical conductivity (EC); and C, N, and P ecological stoichiometric characteristics of freshwater marsh soil significantly changed after translocation for 23 months. SWC decreased on the high- and middle-tidal flats (P<0.05) and increased on the low-tidal flat (P<0.05). EC increased to different degrees on all three tidal flats (P<0.05). Soil total organic carbon (TOC) and total nitrogen (TN) were significantly lower on the high-tidal flat (P<0.05), whereas total phosphorus (TP) was significantly lower on the middle- and high-tidal flats (P<0.05). C:N was decreased on the high- and middle-tidal flats (P<0.05); C:P and N:P were lower on the high-tidal flat; and all C, N, and P ecological stoichiometric characteristics showed no change on the low-tidal flat (P>0.05). Pearson's analysis showed that the ecological stoichiometric characteristics of C, N, and P were related to some properties of soil over the culture sites. The PLS-SEM model showed that the water-salt environment had different effects on soil C:N, C:P, and N:P through the main pathways of negative effects on soil TOC and TP. The results suggest that sea level rise may impact the C, N, and P ecological stoichiometric characteristics in freshwater marsh soil, resulting in some possible changes in the nutrient cycles of estuarine wetlands.

MULGA, a unified multi-view graph autoencoder-based approach for identifying drug-protein interaction and drug repositioning.

MOTIVATION: Identifying drug-protein interactions (DPIs) is a critical step in drug repositioning, which allows reuse of approved drugs that may be effective for treating a different disease and thereby alleviates the challenges of new drug development. Despite the fact that a great variety of computational approaches for DPI prediction have been proposed, key challenges, such as extendable and unbiased similarity calculation, heterogeneous information utilization, and reliable negative sample selection, remain to be addressed. RESULTS: To address these issues, we propose a novel, unified multi-view graph autoencoder framework, termed MULGA, for both DPI and drug repositioning predictions. MULGA is featured by: (i) a multi-view learning technique to effectively learn authentic drug affinity and target affinity matrices; (ii) a graph autoencoder to infer missing DPI interactions; and (iii) a new "guilty-by-association"-based negative sampling approach for selecting highly reliable non-DPIs. Benchmark experiments demonstrate that MULGA outperforms state-of-the-art methods in DPI prediction and the ablation studies verify the effectiveness of each proposed component. Importantly, we highlight the top drugs shortlisted by MULGA that target the spike glycoprotein of severe acute respiratory syndrome coronavirus 2 (SAR-CoV-2), offering additional insights into and potentially useful treatment option for COVID-19. Together with the availability of datasets and source codes, we envision that MULGA can be explored as a useful tool for DPI prediction and drug repositioning. AVAILABILITY AND IMPLEMENTATION: MULGA is publicly available for academic purposes at https://github.com/jianiM/MULGA/.

Surformer: An interpretable pattern-perceptive survival transformer for cancer survival prediction from histopathology whole slide images.

BACKGROUND AND OBJECTIVE: High-resolution histopathology whole slide images (WSIs) contain abundant valuable information for cancer prognosis. However, most computational pathology methods for survival prediction have weak interpretability and cannot explain the decision-making processes reasonably. To address this issue, we propose a highly interpretable neural network termed pattern-perceptive survival transformer (Surformer) for cancer survival prediction from WSIs. METHODS: Notably, Surformer can quantify specific histological patterns through bag-level labels without any patch/cell-level auxiliary information. Specifically, the proposed ratio-reserved cross-attention module (RRCA) generates global and local features with the learnable prototypes (pglobal, plocals) as detectors and quantifies the patches correlative to each plocal in the form of ratio factors (rfs). Afterward, multi-head self&cross-attention modules proceed with the computation for feature enhancement against noise. Eventually, the designed disentangling loss function guides multiple local features to focus on distinct patterns, thereby assisting rfs from RRCA in achieving more explicit histological feature quantification. RESULTS: Extensive experiments on five TCGA datasets illustrate that Surformer outperforms existing state-of-the-art methods. In addition, we highlight its interpretation by visualizing rfs distribution across high-risk and low-risk cohorts and retrieving and analyzing critical histological patterns contributing to the survival prediction. CONCLUSIONS: Surformer is expected to be exploited as a useful tool for performing histopathology image data-driven analysis and gaining new insights for interpreting the associations between such images and patient survival states.