READRetro: natural product biosynthesis predicting with retrieval-augmented dual-view retrosynthesis

Plants, as a sessile organism, produce various secondary metabolites to interact with the environment. These chemicals have fascinated the plant science community because of their ecological significance and notable biological activity. However, predicting the complete biosynthetic pathways from target molecules to metabolic building blocks remains a challenge. Here, we propose retrieval-augmented dual-view retrosynthesis (READRetro) as a practical bio-retrosynthesis tool to predict the biosynthetic pathways of plant natural products. Conventional bio-retrosynthesis models have been limited in their ability to predict biosynthetic pathways for natural products. READRetro was optimized for the prediction of complex metabolic pathways by incorporating cutting-edge deep learning architectures, an ensemble approach, and two retrievers. Evaluation of single- and multi-step retrosynthesis showed that each component of READRetro significantly improved its ability to predict biosynthetic pathways. READRetro was also able to propose the known pathways of secondary metabolites such as monoterpene indole alkaloids and the unknown pathway of menisdaurilide, demonstrating its applicability to real-world bio-retrosynthesis of plant natural products. For researchers interested in the biosynthesis and production of secondary metabolites, a user-friendly website (https://readretro.net) and the open-source code of READRetro have been made available.

Engineering salt-tolerant Cas12f1 variants for gene-editing applications

CRISPR has revolutionized the field of genome editing in life sciences by serving as a versatile and state-of-the-art tool. Cas12f1 is a small nuclease of the bacterial immunity CRISPR system with an ideal size for cellular delivery, in contrast to CRISPR-associated (Cas) proteins like Cas9 or Cas12. However, Cas12f1 works best at low salt concentrations. In this study, we find that the plasticity of certain Cas12f1 regions (K196-Y202 and I452-L515) is negatively affected by increased salt concentrations. On this line, key protein domains (REC1, WED, Nuc, lid) that are involved in the DNA-target recognition and the activation of the catalytic RuvC domain are in turn also affected. We suggest that salt concentration should be taken in to consideration for activity assessments of Cas engineered variants, especially if the mutations are on the protospacer adjacent motif interacting domain. The results can be exploited for the engineering of Cas variants and the assessment of their activity at varying salt concentrations. We propose that the K198Q mutation can restore at great degree the compromised plasticity and could potentially lead to salt-tolerant Cas12f1 variants. The methodology can be also employed for the study of biomolecules in terms of their salinity tolerance.Communicated by Ramaswamy H. Sarma.

Deep Learning for Dynamic Graphs: Models and Benchmarks

Recent progress in research on deep graph networks (DGNs) has led to a maturation of the domain of learning on graphs. Despite the growth of this research field, there are still important challenges that are yet unsolved. Specifically, there is an urge of making DGNs suitable for predictive tasks on real-world systems of interconnected entities, which evolve over time. With the aim of fostering research in the domain of dynamic graphs, first, we survey recent advantages in learning both temporal and spatial information, providing a comprehensive overview of the current state-of-the-art in the domain of representation learning for dynamic graphs. Second, we conduct a fair performance comparison among the most popular proposed approaches on node- and edge-level tasks, leveraging rigorous model selection and assessment for all the methods, thus establishing a sound baseline for evaluating new architectures and approaches.

Classification of epileptic seizures in EEG data based on iterative gated graph convolution network

Introduction

The automatic and precise classification of epilepsy types using electroencephalogram (EEG) data promises significant advancements in diagnosing patients with epilepsy. However, the intricate interplay among multiple electrode signals in EEG data poses challenges. Recently, Graph Convolutional Neural Networks (GCN) have shown strength in analyzing EEG data due to their capability to describe complex relationships among different EEG regions. Nevertheless, several challenges remain: (1) GCN typically rely on predefined or prior graph topologies, which may not accurately reflect the complex correlations between brain regions. (2) GCN struggle to capture the long-temporal dependencies inherent in EEG signals, limiting their ability to effectively extract temporal features.

Methods

To address these challenges, we propose an innovative epileptic seizure classification model based on an Iterative Gated Graph Convolutional Network (IGGCN). For the epileptic seizure classification task, the original EEG graph structure is iteratively optimized using a multi-head attention mechanism during training, rather than relying on a static, predefined prior graph. We introduce Gated Graph Neural Networks (GGNN) to enhance the model's capacity to capture long-term dependencies in EEG series between brain regions. Additionally, Focal Loss is employed to alleviate the imbalance caused by the scarcity of epileptic EEG data.

Results

Our model was evaluated on the Temple University Hospital EEG Seizure Corpus (TUSZ) for classifying four types of epileptic seizures. The results are outstanding, achieving an average F1 score of 91.5% and an average Recall of 91.8%, showing a substantial improvement over current state-of-the-art models.

Discussion

Ablation experiments verified the efficacy of iterative graph optimization and gated graph convolution. The optimized graph structure significantly differs from the predefined EEG topology. Gated graph convolutions demonstrate superior performance in capturing the long-term dependencies in EEG series. Additionally, Focal Loss outperforms other commonly used loss functions in the TUSZ classification task.

A Point Cloud Graph Neural Network for Protein-Ligand Binding Site Prediction

Predicting protein-ligand binding sites is an integral part of structural biology and drug design. A comprehensive understanding of these binding sites is essential for advancing drug innovation, elucidating mechanisms of biological function, and exploring the nature of disease. However, accurately identifying protein-ligand binding sites remains a challenging task. To address this, we propose PGpocket, a geometric deep learning-based framework to improve protein-ligand binding site prediction. Initially, the protein surface is converted into a point cloud, and then the geometric and chemical properties of each point are calculated. Subsequently, the point cloud graph is constructed based on the inter-point distances, and the point cloud graph neural network (GNN) is applied to extract and analyze the protein surface information to predict potential binding sites. PGpocket is trained on the scPDB dataset, and its performance is verified on two independent test sets, Coach420 and HOLO4K. The results show that PGpocket achieves a 58% success rate on the Coach420 dataset and a 56% success rate on the HOLO4K dataset. These results surpass competing algorithms, demonstrating PGpocket's advancement and practicality for protein-ligand binding site prediction.

Spatial-Temporal Transformer Networks for Traffic Flow Forecasting Using a Pre-Trained Language Model

Most current methods use spatial-temporal graph neural networks (STGNNs) to analyze complex spatial-temporal information from traffic data collected from hundreds of sensors. STGNNs combine graph neural networks (GNNs) and sequence models to create hybrid structures that allow for the two networks to collaborate. However, this collaboration has made the model increasingly complex. This study proposes a framework that relies solely on original Transformer architecture and carefully designs embeddings to efficiently extract spatial-temporal dependencies in traffic flow. Additionally, we used pre-trained language models to enhance forecasting performance. We compared our new framework with current state-of-the-art STGNNs and Transformer-based models using four real-world traffic datasets: PEMS04, PEMS08, METR-LA, and PEMS-BAY. The experimental results demonstrate that our framework outperforms the other models in most metrics.

Federated learning for network attack detection using attention-based graph neural networks

Federated Learning is an effective solution to address the issues of data isolation and privacy leakage in machine learning. However, ensuring the security of network devices and architectures deploying federated learning remains a challenge due to network attacks. This paper proposes an attention-based Graph Neural Network for detecting cross-level and cross-department network attacks. This method enables collaborative model training while protecting data privacy on distributed devices. By organizing network traffic information in chronological order and constructing a graph structure based on log density, enhances the accuracy of network attack detection. The introduction of an attention mechanism and the construction of a Federated Graph Attention Network (FedGAT) model are used to evaluate the interactivity between nodes in the graph, thereby improving the precision of internal network interactions. Experimental results demonstrate that our method achieves comparable accuracy and robustness to traditional detection methods while prioritizing privacy protection and data security.

Distilroberta2gnn: a new hybrid deep learning approach for aspect-based sentiment analysis

In the field of natural language processing (NLP), aspect-based sentiment analysis (ABSA) is crucial for extracting insights from complex human sentiments towards specific text aspects. Despite significant progress, the field still faces challenges such as accurately interpreting subtle language nuances and the scarcity of high-quality, domain-specific annotated datasets. This study introduces the Distil- RoBERTa2GNN model, an innovative hybrid approach that combines the DistilRoBERTa pre-trained model's feature extraction capabilities with the dynamic sentiment classification abilities of graph neural networks (GNN). Our comprehensive, four-phase data preprocessing strategy is designed to enrich model training with domain-specific, high-quality data. In this study, we analyze four publicly available benchmark datasets: Rest14, Rest15, Rest16-EN, and Rest16-ESP, to rigorously evaluate the effectiveness of our novel DistilRoBERTa2GNN model in ABSA. For the Rest14 dataset, our model achieved an F1 score of 77.98%, precision of 78.12%, and recall of 79.41%. The Rest15 dataset shows that our model achieves an F1 score of 76.86%, precision of 80.70%, and recall of 79.37%. For the Rest16-EN dataset, our model reached an F1 score of 84.96%, precision of 82.77%, and recall of 87.28%. For Rest16-ESP (Spanish dataset), our model achieved an F1 score of 74.87%, with a precision of 73.11% and a recall of 76.80%. These metrics highlight our model's competitive edge over different baseline models used in ABSA studies. This study addresses critical ABSA challenges and sets a new benchmark for sentiment analysis research, guiding future efforts toward enhancing model adaptability and performance across diverse datasets.

Adaptive node feature extraction in graph-based neural networks for brain diseases diagnosis using self-supervised learning

Electroencephalography (EEG) has demonstrated significant value in diagnosing brain diseases. In particular, brain networks have gained prominence as they offer additional valuable insights by establishing connections between EEG signal channels. While brain connections are typically delineated by channel signal similarity, there lacks a consistent and reliable strategy for ascertaining node characteristics. Conventional node features such as temporal and frequency domain properties of EEG signals prove inadequate for capturing the extensive EEG information. In our investigation, we introduce a novel adaptive method for extracting node features from EEG signals utilizing a distinctive task-induced self-supervised learning technique. By amalgamating these extracted node features with fundamental edge features constructed using Pearson correlation coefficients, we showed that the proposed approach can function as a plug-in module that can be integrated to many common GNN networks (e.g., GCN, GraphSAGE, GAT) as a replacement of node feature selections module. Comprehensive experiments are then conducted to demonstrate the consistently superior performance and high generality of the proposed method over other feature selection methods in various of brain disorder prediction tasks, such as depression, schizophrenia, and Parkinson's disease. Furthermore, compared to other node features, our approach unveils profound spatial patterns through graph pooling and structural learning, shedding light on pivotal brain regions influencing various brain disorder prediction based on derived features.

Discovery of novel materials through machine learning

Experimental exploration of new materials relies heavily on a laborious trial-and-error approach. In addition to substantial time and resource requirements, traditional experiments and computational modelling are typically limited in finding target materials within the enormous chemical space. Therefore, creating innovative techniques to expedite material discovery becomes essential. Recently, machine learning (ML) has emerged as a valuable tool for material discovery, garnering significant attention due to its remarkable advancements in prediction accuracy and time efficiency. This rapidly developing computational technique accelerates the search and optimization process and enables the prediction of material properties at a minimal computational cost, thereby facilitating the discovery of novel materials. We provide a comprehensive overview of recent studies on discovering new materials by predicting materials and their properties using ML techniques. Beginning with an introduction of the fundamental principles of ML methods, we subsequently examine the current research landscape on the applications of ML in predicting material properties that lead to the discovery of novel materials. Finally, we discuss challenges in employing ML within materials science, propose potential solutions, and outline future research directions.

Machine Learning and Deep Learning Approaches in Lifespan Brain Age Prediction: A Comprehensive Review

The concept of 'brain age', derived from neuroimaging data, serves as a crucial biomarker reflecting cognitive vitality and neurodegenerative trajectories. In the past decade, machine learning (ML) and deep learning (DL) integration has transformed the field, providing advanced models for brain age estimation. However, achieving precise brain age prediction across all ages remains a significant analytical challenge. This comprehensive review scrutinizes advancements in ML- and DL-based brain age prediction, analyzing 52 peer-reviewed studies from 2020 to 2024. It assesses various model architectures, highlighting their effectiveness and nuances in lifespan brain age studies. By comparing ML and DL, strengths in forecasting and methodological limitations are revealed. Finally, key findings from the reviewed articles are summarized and a number of major issues related to ML/DL-based lifespan brain age prediction are discussed. Through this study, we aim at the synthesis of the current state of brain age prediction, emphasizing both advancements and persistent challenges, guiding future research, technological advancements, and improving early intervention strategies for neurodegenerative diseases.

Rapid prediction of wall shear stress in stenosed coronary arteries based on deep learning

There is increasing evidence that coronary artery wall shear stress (WSS) measurement provides useful prognostic information that allows prediction of adverse cardiovascular events. Computational Fluid Dynamics (CFD) has been extensively used in research to measure vessel physiology and examine the role of the local haemodynamic forces on the evolution of atherosclerosis. Nonetheless, CFD modelling remains computationally expensive and time-consuming, making its direct use in clinical practice inconvenient. A number of studies have investigated the use of deep learning (DL) approaches for fast WSS prediction. However, in these reports, patient data were limited and most of them used synthetic data generation methods for developing the training set. In this paper, we implement 2 approaches for synthetic data generation and combine their output with real patient data in order to train a DL model with a U-net architecture for prediction of WSS in the coronary arteries. The model achieved 6.03% Normalised Mean Absolute Error (NMAE) with inference taking only 0.35 s; making this solution time-efficient and clinically relevant.

Machine Learning Assists in the Design and Application of Microneedles

Microneedles (MNs), characterized by their micron-sized sharp tips, can painlessly penetrate the skin and have shown significant potential in disease treatment and biosensing. With the development of artificial intelligence (AI), the design and application of MNs have experienced substantial innovation aided by machine learning (ML). This review begins with a brief introduction to the concept of ML and its current stage of development. Subsequently, the design principles and fabrication methods of MNs are explored, demonstrating the critical role of ML in optimizing their design and preparation. Integration between ML and the applications of MNs in therapy and sensing were further discussed. Finally, we outline the challenges and prospects of machine learning-assisted MN technology, aiming to advance its practical application and development in the field of smart diagnosis and treatment.

PT-KGNN: A framework for pre-training biomedical knowledge graphs with graph neural networks

Biomedical knowledge graphs (KGs) serve as comprehensive data repositories that contain rich information about nodes and edges, providing modeling capabilities for complex relationships among biological entities. Many approaches either learn node features through traditional machine learning methods, or leverage graph neural networks (GNNs) to directly learn features of target nodes in the biomedical KGs and utilize them for downstream tasks. Motivated by the pre-training technique in natural language processing (NLP), we propose a framework named PT-KGNN (Pre-Training the biomedical KG with GNNs) to learn embeddings of nodes in a broader context by applying GNNs on the biomedical KG. We design several experiments to evaluate the effectivity of our proposed framework and the impact of the scale of KGs. The results of tasks consistently improve as the scale of the biomedical KG used for pre-training increases. Pre-training on large-scale biomedical KGs significantly enhances the drug-drug interaction (DDI) and drug-disease association (DDA) prediction performance on the independent dataset. The embeddings derived from a larger biomedical KG have demonstrated superior performance compared to those obtained from a smaller KG. By applying pre-training techniques on biomedical KGs, rich semantic and structural information can be learned, leading to enhanced performance on downstream tasks. it is evident that pre-training techniques hold tremendous potential and wide-ranging applications in bioinformatics.

Masked hypergraph learning for weakly supervised histopathology whole slide image classification

BACKGROUND AND OBJECTIVES

Graph neural network (GNN) has been extensively used in histopathology whole slide image (WSI) analysis due to the efficiency and flexibility in modelling relationships among entities. However, most existing GNN-based WSI analysis methods only consider the pairwise correlation of patches from one single perspective (e.g. spatial affinity or embedding similarity) yet ignore the intrinsic non-pairwise relationships present in gigapixel WSI, which are likely to contribute to feature learning and downstream tasks. The objective of this study is therefore to explore the non-pairwise relationships in histopathology WSI and exploit them to guide the learning of slide-level representations for better classification performance.

METHODS

In this paper, we propose a novel Masked HyperGraph Learning (MaskHGL) framework for weakly supervised histopathology WSI classification. Compared with most GNN-based WSI classification methods, MaskHGL exploits the non-pairwise correlations between patches with hypergraph and global message passing conducted by hypergraph convolution. Concretely, multi-perspective hypergraphs are first built for each WSI, then hypergraph attention is introduced into the jointed hypergraph to propagate the non-pairwise relationships and thus yield more discriminative node representation. More importantly, a masked hypergraph reconstruction module is devised to guide the hypergraph learning which can generate more powerful robustness and generalization than the method only using hypergraph modelling. Additionally, a self-attention-based node aggregator is also applied to explore the global correlation of patches in WSI and produce the slide-level representation for classification.

RESULTS

The proposed method is evaluated on two public TCGA benchmark datasets and one in-house dataset. On the public TCGA-LUNG (1494 WSIs) and TCGA-EGFR (696 WSIs) test set, the area under receiver operating characteristic (ROC) curve (AUC) were 0.9752±0.0024 and 0.7421±0.0380, respectively. On the USTC-EGFR (754 WSIs) dataset, MaskHGL achieved significantly better performance with an AUC of 0.8745±0.0100, which surpassed the second-best state-of-the-art method SlideGraph+ 2.64%.

CONCLUSIONS

MaskHGL shows a great improvement, brought by considering the intrinsic non-pairwise relationships within WSI, in multiple downstream WSI classification tasks. In particular, the designed masked hypergraph reconstruction module promisingly alleviates the data scarcity and greatly enhances the robustness and classification ability of our MaskHGL. Notably, it has shown great potential in cancer subtyping and fine-grained lung cancer gene mutation prediction from hematoxylin and eosin (H&E) stained WSIs.

Towards Inductive and Efficient Explanations for Graph Neural Networks

Despite recent progress in Graph Neural Networks (GNNs), explaining predictions made by GNNs remains a challenging and nascent problem. The leading method mainly considers the local explanations, i.e., important subgraph structure and node features, to interpret why a GNN model makes the prediction for a single instance, e.g. a node or a graph. As a result, the explanation generated is painstakingly customized at the instance level. The unique explanation interpreting each instance independently is not sufficient to provide a global understanding of the learned GNN model, leading to the lack of generalizability and hindering it from being used in the inductive setting. Besides, training the explanation model explaining for each instance is time-consuming for large-scale real-life datasets. In this study, we address these key challenges and propose PGExplainer, a parameterized explainer for GNNs. PGExplainer adopts a deep neural network to parameterize the generation process of explanations, which renders PGExplainer a natural approach to multi-instance explanations. Compared to the existing work, PGExplainer has better generalization ability and can be utilized in an inductive setting without training the model for new instances. Thus, PGExplainer is much more efficient than the leading method with significant speed-up. In addition, the explanation networks can also be utilized as a regularizer to improve the generalization power of existing GNNs when jointly trained with downstream tasks. Experiments on both synthetic and real-life datasets show highly competitive performance with up to 24.7% relative improvement in AUC on explaining graph classification over the leading baseline.

Distributed Optimization of Graph Convolutional Network Using Subgraph Variance

In recent years, distributed graph convolutional networks (GCNs) training frameworks have achieved great success in learning the representation of graph-structured data with large sizes. However, existing distributed GCN training frameworks require enormous communication costs since a multitude of dependent graph data need to be transmitted from other processors. To address this issue, we propose a graph augmentation-based distributed GCN framework (GAD). In particular, GAD has two main components: GAD-Partition and GAD-Optimizer. We first propose an augmentation-based graph partition (GAD-Partition) that can divide the input graph into augmented subgraphs to reduce communication by selecting and storing as few significant vertices of other processors as possible. To further speed up distributed GCN training and improve the quality of the training result, we design a subgraph variance-based importance calculation formula and propose a novel weighted global consensus method, collectively referred to as GAD-Optimizer. This optimizer adaptively adjusts the importance of subgraphs to reduce the effect of extra variance introduced by GAD-Partition on distributed GCN training. Extensive experiments on four large-scale real-world datasets demonstrate that our framework significantly reduces the communication overhead ( ≈ 50% ), improves the convergence speed ( ≈ 2 × ) of distributed GCN training, and obtains a slight gain in accuracy ( ≈ 0.45% ) based on minimal redundancy compared to the state-of-the-art methods.

Autophagy and machine learning: Unanswered questions

Autophagy is a critical conserved cellular process in maintaining cellular homeostasis by clearing and recycling damaged organelles and intracellular components in lysosomes and vacuoles. Autophagy plays a vital role in cell survival, bioenergetic homeostasis, organism development, and cell death regulation. Malfunctions in autophagy are associated with various human diseases and health disorders, such as cancers and neurodegenerative diseases. Significant effort has been devoted to autophagy-related research in the context of genes, proteins, diagnosis, etc. In recent years, there has been a surge of studies utilizing state of the art machine learning (ML) tools to analyze and understand the roles of autophagy in various biological processes. We taxonomize ML techniques that are applicable in an autophagy context, comprehensively review existing efforts being taken in this direction, and outline principles to consider in a biomedical context. In recognition of recent groundbreaking advances in the deep-learning community, we discuss new opportunities in interdisciplinary collaborations and seek to engage autophagy and computer science researchers to promote autophagy research with joint efforts.

Deep learning algorithms applied to computational chemistry

Recently, there has been a significant increase in the use of deep learning techniques in the molecular sciences, which have shown high performance on datasets and the ability to generalize across data. However, no model has achieved perfect performance in solving all problems, and the pros and cons of each approach remain unclear to those new to the field. Therefore, this paper aims to review deep learning algorithms that have been applied to solve molecular challenges in computational chemistry. We proposed a comprehensive categorization that encompasses two primary approaches; conventional deep learning and geometric deep learning models. This classification takes into account the distinct techniques employed by the algorithms within each approach. We present an up-to-date analysis of these algorithms, emphasizing their key features and open issues. This includes details of input descriptors, datasets used, open-source code availability, task solutions, and actual research applications, focusing on general applications rather than specific ones such as drug discovery. Furthermore, our report discusses trends and future directions in molecular algorithm design, including the input descriptors used for each deep learning model, GPU usage, training and forward processing time, model parameters, the most commonly used datasets, libraries, and optimization schemes. This information aids in identifying the most suitable algorithms for a given task. It also serves as a reference for the datasets and input data frequently used for each algorithm technique. In addition, it provides insights into the benefits and open issues of each technique, and supports the development of novel computational chemistry systems.

AI-Powered Microfluidics: Shaping the Future of Phenotypic Drug Discovery

Phenotypic drug discovery (PDD), which involves harnessing biological systems directly to uncover effective drugs, has undergone a resurgence in recent years. The rapid advancement of artificial intelligence (AI) over the past few years presents numerous opportunities for augmenting phenotypic drug screening on microfluidic platforms, leveraging its predictive capabilities, data analysis, efficient data processing, etc. Microfluidics coupled with AI is poised to revolutionize the landscape of phenotypic drug discovery. By integrating advanced microfluidic platforms with AI algorithms, researchers can rapidly screen large libraries of compounds, identify novel drug candidates, and elucidate complex biological pathways with unprecedented speed and efficiency. This review provides an overview of recent advances and challenges in AI-based microfluidics and their applications in drug discovery. We discuss the synergistic combination of microfluidic systems for high-throughput screening and AI-driven analysis for phenotype characterization, drug-target interactions, and predictive modeling. In addition, we highlight the potential of AI-powered microfluidics to achieve an automated drug screening system. Overall, AI-powered microfluidics represents a promising approach to shaping the future of phenotypic drug discovery by enabling rapid, cost-effective, and accurate identification of therapeutically relevant compounds.

Effective optimization of atomic decoration in giant and superstructurally ordered crystals with machine learning

Crystals with complicated geometry are often observed with mixed chemical occupancy among Wyckoff sites, presenting a unique challenge for accurate atomic modeling. Similar systems possessing exact occupancy on all the sites can exhibit superstructural ordering, dramatically inflating the unit cell size. In this work, a crystal graph convolutional neural network (CGCNN) is used to predict optimal atomic decorations on fixed crystalline geometries. This is achieved with a site permutation search (SPS) optimization algorithm based on Monte Carlo moves combined with simulated annealing and basin-hopping techniques. Our approach relies on the evidence that, for a given chemical composition, a CGCNN estimates the correct energetic ordering of different atomic decorations, as predicted by electronic structure calculations. This provides a suitable energy landscape that can be optimized according to site occupation, allowing the prediction of chemical decoration in crystals exhibiting mixed or disordered occupancy, or superstructural ordering. Verification of the procedure is carried out on several known compounds, including the superstructurally ordered clathrate compound Rb8Ga27Sb16 and vacancy-ordered perovskite Cs2SnI6, neither of which was previously seen during the neural network training. In addition, the critical temperature of an order-disorder phase transition in solid solution CuZn is probed with our SPS routines by sampling site configuration trajectories in the canonical ensemble. This strategy provides an accurate method for determining favorable decoration in complex crystals and analyzing site occupation at unprecedented speed and scale.

Multiview representation learning for identification of novel cancer genes and their causative biological mechanisms

Tumorigenesis arises from the dysfunction of cancer genes, leading to uncontrolled cell proliferation through various mechanisms. Establishing a complete cancer gene catalogue will make precision oncology possible. Although existing methods based on graph neural networks (GNN) are effective in identifying cancer genes, they fall short in effectively integrating data from multiple views and interpreting predictive outcomes. To address these shortcomings, an interpretable representation learning framework IMVRL-GCN is proposed to capture both shared and specific representations from multiview data, offering significant insights into the identification of cancer genes. Experimental results demonstrate that IMVRL-GCN outperforms state-of-the-art cancer gene identification methods and several baselines. Furthermore, IMVRL-GCN is employed to identify a total of 74 high-confidence novel cancer genes, and multiview data analysis highlights the pivotal roles of shared, mutation-specific, and structure-specific representations in discriminating distinctive cancer genes. Exploration of the mechanisms behind their discriminative capabilities suggests that shared representations are strongly associated with gene functions, while mutation-specific and structure-specific representations are linked to mutagenic propensity and functional synergy, respectively. Finally, our in-depth analyses of these candidates suggest potential insights for individualized treatments: afatinib could counteract many mutation-driven risks, and targeting interactions with cancer gene SRC is a reasonable strategy to mitigate interaction-induced risks for NR3C1, RXRA, HNF4A, and SP1.

ToxTeller: Predicting Peptide Toxicity Using Four Different Machine Learning Approaches

Examining the toxicity of peptides is essential for therapeutic peptide-based drug design. Machine learning approaches are frequently used to develop highly accurate predictors for peptide toxicity prediction. In this paper, we present ToxTeller, which provides four predictors using logistic regression, support vector machines, random forests, and XGBoost, respectively. For prediction model development, we construct a data set of toxic and nontoxic peptides from SwissProt and ConoServer databases with existence evidence levels checked. We also fully utilize the protein annotation in SwissProt to collect more toxic peptides than using keyword search alone. From this data set, we construct an independent test data set that shares at most 40% sequence similarity within itself and with the training data set. From a quite comprehensive list of 28 feature combinations, we conduct 10-fold cross-validation on the training data set to determine the optimized feature combination for model development. ToxTeller's performance is evaluated and compared with existing predictors on the independent test data set. Since toxic peptides must be avoided for drug design, we analyze strategies for reducing false-negative predictions of toxic peptides and suggest selecting models by top sensitivity instead of the widely used Matthews correlation coefficient, and also suggest using a meta-predictor approach with multiple predictors.

scSemiProfiler: Advancing large-scale single-cell studies through semi-profiling with deep generative models and active learning

Single-cell sequencing is a crucial tool for dissecting the cellular intricacies of complex diseases. Its prohibitive cost, however, hampers its application in expansive biomedical studies. Traditional cellular deconvolution approaches can infer cell type proportions from more affordable bulk sequencing data, yet they fall short in providing the detailed resolution required for single-cell-level analyses. To overcome this challenge, we introduce "scSemiProfiler", an innovative computational framework that marries deep generative models with active learning strategies. This method adeptly infers single-cell profiles across large cohorts by fusing bulk sequencing data with targeted single-cell sequencing from a few rigorously chosen representatives. Extensive validation across heterogeneous datasets verifies the precision of our semi-profiling approach, aligning closely with true single-cell profiling data and empowering refined cellular analyses. Originally developed for extensive disease cohorts, "scSemiProfiler" is adaptable for broad applications. It provides a scalable, cost-effective solution for single-cell profiling, facilitating in-depth cellular investigation in various biological domains.

MGLEP: Multimodal Graph Learning for Modeling Emerging Pandemics with Big Data

Accurate forecasting and analysis of emerging pandemics play a crucial role in effective public health management and decision-making. Traditional approaches primarily rely on epidemiological data, overlooking other valuable sources of information that could act as sensors or indicators of pandemic patterns. In this paper, we propose a novel framework, MGLEP, that integrates temporal graph neural networks and multi-modal data for learning and forecasting. We incorporate big data sources, including social media content, by utilizing specific pre-trained language models and discovering the underlying graph structure among users. This integration provides rich indicators of pandemic dynamics through learning with temporal graph neural networks. Extensive experiments demonstrate the effectiveness of our framework in pandemic forecasting and analysis, outperforming baseline methods across different areas, pandemic situations, and prediction horizons. The fusion of temporal graph learning and multi-modal data enables a comprehensive understanding of the pandemic landscape with less time lag, cheap cost, and more potential information indicators.

Coupling Fault Diagnosis Based on Dynamic Vertex Interpretable Graph Neural Network

Mechanical equipment is composed of several parts, and the interaction between parts exists throughout the whole life cycle, leading to the widespread phenomenon of fault coupling. The diagnosis of independent faults cannot meet the requirements of the health management of mechanical equipment under actual working conditions. In this paper, the dynamic vertex interpretable graph neural network (DIGNN) is proposed to solve the problem of coupling fault diagnosis, in which dynamic vertices are defined in the data topology. First, in the date preprocessing phase, wavelet transform is utilized to make input features interpretable and reduce the uncertainty of model training. In the fault topology, edge connections are made between nodes according to the fault coupling information, and edge connections are established between dynamic nodes and all other nodes. Second the data topology with dynamic vertices is used in the training phase and in the testing phase, the time series data are only fed into dynamic vertices for classification and analysis, which makes it possible to realize coupling fault diagnosis in an industrial production environment. The features extracted in different layers of DIGNN interpret how the model works. The method proposed in this paper can realize the accurate diagnosis of independent faults in the dataset with an accuracy of 100%, and can effectively judge the coupling mode of coupling faults with a comprehensive accuracy of 88.3%.

HGTMDA: A Hypergraph Learning Approach with Improved GCN-Transformer for miRNA-Disease Association Prediction

Accumulating scientific evidence highlights the pivotal role of miRNA-disease association research in elucidating disease pathogenesis and developing innovative diagnostics. Consequently, accurately identifying disease-associated miRNAs has emerged as a prominent research topic in bioinformatics. Advances in graph neural networks (GNNs) have catalyzed methodological breakthroughs in this field. However, existing methods are often plagued by data noise and struggle to effectively integrate local and global information, which hinders their predictive performance. To address this, we introduce HGTMDA, an innovative hypergraph learning framework that incorporates random walk with restart-based association masking and an enhanced GCN-Transformer model to infer miRNA-disease associations. HGTMDA starts by constructing multiple homogeneous similarity networks. A novel enhancement of our approach is the introduction of a restart-based random walk association masking strategy. By stochastically masking a subset of association data and integrating it with a GCN enhanced by an attention mechanism, this strategy enables better capture of key information, leading to improved information utilization and reduced impact of noisy data. Next, we build an miRNA-disease heterogeneous hypergraph and adopt an improved GCN-Transformer encoder to effectively solve the effective extraction of local and global information. Lastly, we utilize a combined Dice cross-entropy (DCE) loss function to guide the model training and optimize its performance. To evaluate the performance of HGTMDA, comprehensive comparisons were conducted with state-of-the-art methods. Additionally, in-depth case studies on lung cancer and colorectal cancer were performed. The results demonstrate HGTMDA's outstanding performance across various metrics and its exceptional effectiveness in real-world application scenarios, highlighting the advantages and value of this method.

LGGNet: Learning From Local-Global-Graph Representations for Brain-Computer Interface

Neuropsychological studies suggest that co-operative activities among different brain functional areas drive high-level cognitive processes. To learn the brain activities within and among different functional areas of the brain, we propose local-global-graph network (LGGNet), a novel neurologically inspired graph neural network (GNN), to learn local-global-graph (LGG) representations of electroencephalography (EEG) for brain-computer interface (BCI). The input layer of LGGNet comprises a series of temporal convolutions with multiscale 1-D convolutional kernels and kernel-level attentive fusion. It captures temporal dynamics of EEG which then serves as input to the proposed local- and global-graph-filtering layers. Using a defined neurophysiologically meaningful set of local and global graphs, LGGNet models the complex relations within and among functional areas of the brain. Under the robust nested cross-validation settings, the proposed method is evaluated on three publicly available datasets for four types of cognitive classification tasks, namely the attention, fatigue, emotion, and preference classification tasks. LGGNet is compared with state-of-the-art (SOTA) methods, such as DeepConvNet, EEGNet, R2G-STNN, TSception, regularized graph neural network (RGNN), attention-based multiscale convolutional neural network-dynamical graph convolutional network (AMCNN-DGCN), hierarchical recurrent neural network (HRNN), and GraphNet. The results show that LGGNet outperforms these methods, and the improvements are statistically significant ( ) in most cases. The results show that bringing neuroscience prior knowledge into neural network design yields an improvement of classification performance. The source code can be found at https://github.com/yi-ding-cs/LGG.

Detection of pre-mRNA involved in abnormal splicing using Graph Neural Network and Nearest Correlation Method

BACKGROUND

DNA is the building block of genetic information, and is composed of alternating sequences of exons with genetic information and introns without no genetic information. DNA is damaged by normal metabolic activities and environmental factors, causing base substitutions. Changes at the single nucleotide level are called Single Nucleotide Variants (SNVs), which usually result in mRNAs that are non-pathogenic or produce harmless proteins by splicing. However, when pathogenic SNVs are present in DNA, abnormal RNA splicing may cause transcription and translation of the wrong genetic information, resulting in genetic diseases such as cancer.

METHODS

We proposed a new method for detecting pre-mRNA with SNVs that cause abnormal RNA splicing, which utilizes a Graph Neural Network (GNN) and the Nearest Correlation (NC) method which is a similarity measure based on the correlation among multiple inputs. In the proposed method, multiple features were extracted from target sequences and their similarities are calculated by means of the NC method. A GNN model is trained from a graph constructed based on the similarities among the sequences and used as input of LightGBM in order to classify pre-mRNA with SNVs.

RESULTS

The specificity, recall, G-mean, ROC-AUC (area under the ROC curve), and PR-AUC (area under the PR curve) of the proposed method were 0.73±0.080, 0.69±0.13, 0.71±0.073, 0.76±0.062, and 0.42±0.10.

CONCLUSION

The proposed method will contribute to preventing misjudgment of the disease and lead to the selection of the correct treatment in the future.

Global explanation supervision for Graph Neural Networks

With the increasing popularity of Graph Neural Networks (GNNs) for predictive tasks on graph structured data, research on their explainability is becoming more critical and achieving significant progress. Although many methods are proposed to explain the predictions of GNNs, their focus is mainly on "how to generate explanations." However, other important research questions like "whether the GNN explanations are inaccurate," "what if the explanations are inaccurate," and "how to adjust the model to generate more accurate explanations" have gained little attention. Our previous GNN Explanation Supervision (GNES) framework demonstrated effectiveness on improving the reasonability of the local explanation while still keep or even improve the backbone GNNs model performance. In many applications instead of per sample explanations, we need to find global explanations which are reasonable and faithful to the domain data. Simply learning to explain GNNs locally is not an optimal solution to a global understanding of the model. To improve the explainability power of the GNES framework, we propose the Global GNN Explanation Supervision (GGNES) technique which uses a basic trained GNN and a global extension of the loss function used in the GNES framework. This GNN creates local explanations which are fed to a Global Logic-based GNN Explainer, an existing technique that can learn the global Explanation in terms of a logic formula. These two frameworks are then trained iteratively to generate reasonable global explanations. Extensive experiments demonstrate the effectiveness of the proposed model on improving the global explanations while keeping the performance similar or even increase the model prediction power.

Variational quantum algorithm for node embedding

Quantum machine learning has made remarkable progress in many important tasks. However, the gate complexity of the initial state preparation is seldom considered in lots of quantum machine learning algorithms, making them non-end-to-end. Herein, we propose a quantum algorithm for the node embedding problem that maps a node graph's topological structure to embedding vectors. The resulting quantum embedding state can be used as an input for other quantum machine learning algorithms. With O ( log ( N ) ) qubits to store the information of N nodes, our algorithm will not lose quantum advantage for the subsequent quantum information processing. Moreover, owing to the use of a parameterized quantum circuit with O ( poly ( log ( N ) ) ) depth, the resulting state can serve as an efficient quantum database. In addition, we explored the measurement complexity of the quantum node embedding algorithm, which is the main issue in training parameters, and extended the algorithm to capture high-order neighborhood information between nodes. Finally, we experimentally demonstrated our algorithm on an nuclear magnetic resonance quantum processor to solve a graph model.

Weakly Supervised AUC Optimization: A Unified Partial AUC Approach

Since acquiring perfect supervision is usually difficult, real-world machine learning tasks often confront inaccurate, incomplete, or inexact supervision, collectively referred to as weak supervision. In this work, we present WSAUC, a unified framework for weakly supervised AUC optimization problems, which covers noisy label learning, positive-unlabeled learning, multi-instance learning, and semi-supervised learning scenarios. Within the WSAUC framework, we first frame the AUC optimization problems in various weakly supervised scenarios as a common formulation of minimizing the AUC risk on contaminated sets, and demonstrate that the empirical risk minimization problems are consistent with the true AUC. Then, we introduce a new type of partial AUC, specifically, the reversed partial AUC (rpAUC), which serves as a robust training objective for AUC maximization in the presence of contaminated labels. WSAUC offers a universal solution for AUC optimization in various weakly supervised scenarios by maximizing the empirical rpAUC. Theoretical and experimental results under multiple settings support the effectiveness of WSAUC on a range of weakly supervised AUC optimization tasks.

Structure-aware deep learning model for peptide toxicity prediction

Antimicrobial resistance is a critical public health concern, necessitating the exploration of alternative treatments. While antimicrobial peptides (AMPs) show promise, assessing their toxicity using traditional wet lab methods is both time-consuming and costly. We introduce tAMPer, a novel multi-modal deep learning model designed to predict peptide toxicity by integrating the underlying amino acid sequence composition and the three-dimensional structure of peptides. tAMPer adopts a graph-based representation for peptides, encoding ColabFold-predicted structures, where nodes represent amino acids and edges represent spatial interactions. Structural features are extracted using graph neural networks, and recurrent neural networks capture sequential dependencies. tAMPer's performance was assessed on a publicly available protein toxicity benchmark and an AMP hemolysis data we generated. On the latter, tAMPer achieves an F1-score of 68.7%, outperforming the second-best method by 23.4%. On the protein benchmark, tAMPer exhibited an improvement of over 3.0% in the F1-score compared to current state-of-the-art methods. We anticipate tAMPer to accelerate AMP discovery and development by reducing the reliance on laborious toxicity screening experiments.

Deep learning for optical tweezers

Optical tweezers exploit light-matter interactions to trap particles ranging from single atoms to micrometer-sized eukaryotic cells. For this reason, optical tweezers are a ubiquitous tool in physics, biology, and nanotechnology. Recently, the use of deep learning has started to enhance optical tweezers by improving their design, calibration, and real-time control as well as the tracking and analysis of the trapped objects, often outperforming classical methods thanks to the higher computational speed and versatility of deep learning. In this perspective, we show how cutting-edge deep learning approaches can remarkably improve optical tweezers, and explore the exciting, new future possibilities enabled by this dynamic synergy. Furthermore, we offer guidelines on integrating deep learning with optical trapping and optical manipulation in a reliable and trustworthy way.

Mesh Neural Networks Based on Dual Graph Pyramids

Deep neural networks (DNNs) have been widely used for mesh processing in recent years. However, current DNNs can not process arbitrary meshes efficiently. On the one hand, most DNNs expect 2-manifold, watertight meshes, but many meshes, whether manually designed or automatically generated, may have gaps, non-manifold geometry, or other defects. On the other hand, the irregular structure of meshes also brings challenges to building hierarchical structures and aggregating local geometric information, which is critical to conduct DNNs. In this paper, we present DGNet, an efficient, effective and generic deep neural mesh processing network based on dual graph pyramids; it can handle arbitrary meshes. First, we construct dual graph pyramids for meshes to guide feature propagation between hierarchical levels for both downsampling and upsampling. Second, we propose a novel convolution to aggregate local features on the proposed hierarchical graphs. By utilizing both geodesic neighbors and euclidean neighbors, the network enables feature aggregation both within local surface patches and between isolated mesh components. Experimental results demonstrate that DGNet can be applied to both shape analysis and large-scale scene understanding. Furthermore, it achieves superior performance on various benchmarks, including ShapeNetCore, HumanBody, ScanNet and Matterport3D. Code and models will be available at https://github.com/li-xl/DGNet.

Deep learning for MRI lesion segmentation in rectal cancer

Rectal cancer (RC) is a globally prevalent malignant tumor, presenting significant challenges in its management and treatment. Currently, magnetic resonance imaging (MRI) offers superior soft tissue contrast and radiation-free effects for RC patients, making it the most widely used and effective detection method. In early screening, radiologists rely on patients' medical radiology characteristics and their extensive clinical experience for diagnosis. However, diagnostic accuracy may be hindered by factors such as limited expertise, visual fatigue, and image clarity issues, resulting in misdiagnosis or missed diagnosis. Moreover, the distribution of surrounding organs in RC is extensive with some organs having similar shapes to the tumor but unclear boundaries; these complexities greatly impede doctors' ability to diagnose RC accurately. With recent advancements in artificial intelligence, machine learning techniques like deep learning (DL) have demonstrated immense potential and broad prospects in medical image analysis. The emergence of this approach has significantly enhanced research capabilities in medical image classification, detection, and segmentation fields with particular emphasis on medical image segmentation. This review aims to discuss the developmental process of DL segmentation algorithms along with their application progress in lesion segmentation from MRI images of RC to provide theoretical guidance and support for further advancements in this field.

Knowledge Graph Convolutional Network with Heuristic Search for Drug Repositioning

Drug repositioning is a strategy of repurposing approved drugs for treating new indications, which can accelerate the drug discovery process, reduce development costs, and lower the safety risk. The advancement of biotechnology has significantly accelerated the speed and scale of biological data generation, offering significant potential for drug repositioning through biomedical knowledge graphs that integrate diverse entities and relations from various biomedical sources. To fully learn the semantic information and topological structure information from the biological knowledge graph, we propose a knowledge graph convolutional network with a heuristic search, named KGCNH, which can effectively utilize the diversity of entities and relationships in biological knowledge graphs, as well as topological structure information, to predict the associations between drugs and diseases. Specifically, we design a relation-aware attention mechanism to compute the attention scores for each neighboring entity of a given entity under different relations. To address the challenge of randomness of the initial attention scores potentially impacting model performance and to expand the search scope of the model, we designed a heuristic search module based on Gumbel-Softmax, which uses attention scores as heuristic information and introduces randomness to assist the model in exploring more optimal embeddings of drugs and diseases. Following this module, we derive the relation weights, obtain the embeddings of drugs and diseases through neighborhood aggregation, and then predict drug-disease associations. Additionally, we employ feature-based augmented views to enhance model robustness and mitigate overfitting issues. We have implemented our method and conducted experiments on two data sets. The results demonstrate that KGCNH outperforms competing methods. In particular, case studies on lithium and quetiapine confirm that KGCNH can retrieve more actual drug-disease associations in the top prediction results.

DKPE-GraphSYN: a drug synergy prediction model based on joint dual kernel density estimation and positional encoding for graph representation

Introduction: Synergistic medication, a crucial therapeutic strategy in cancer treatment, involves combining multiple drugs to enhance therapeutic effectiveness and mitigate side effects. Current research predominantly employs deep learning models for extracting features from cell line and cancer drug structure data. However, these methods often overlook the intricate nonlinear relationships within the data, neglecting the distribution characteristics and weighted probability densities of gene expression data in multi-dimensional space. It also fails to fully exploit the structural information of cancer drugs and the potential interactions between drug molecules. Methods: To overcome these challenges, we introduce an innovative end-to-end learning model specifically tailored for cancer drugs, named Dual Kernel Density and Positional Encoding (DKPE) for Graph Synergy Representation Network (DKPEGraphSYN). This model is engineered to refine the prediction of drug combination synergy effects in cancer. DKPE-GraphSYN utilizes Dual Kernel Density Estimation and Positional Encoding techniques to effectively capture the weighted probability density and spatial distribution information of gene expression, while exploring the interactions and potential relationships between cancer drug molecules via a graph neural network. Results: Experimental results show that our prediction model achieves significant performance enhancements in forecasting drug synergy effects on a comprehensive cancer drug and cell line synergy dataset, achieving an AUPR of 0.969 and an AUC of 0.976. Discussion: These results confirm our model's superior accuracy in predicting cancer drug combinations, providing a supportive method for clinical medication strategy in cancer.

Composite Graph Neural Networks for Molecular Property Prediction

Graph Neural Networks have proven to be very valuable models for the solution of a wide variety of problems on molecular graphs, as well as in many other research fields involving graph-structured data. Molecules are heterogeneous graphs composed of atoms of different species. Composite graph neural networks process heterogeneous graphs with multiple-state-updating networks, each one dedicated to a particular node type. This approach allows for the extraction of information from s graph more efficiently than standard graph neural networks that distinguish node types through a one-hot encoded type of vector. We carried out extensive experimentation on eight molecular graph datasets and on a large number of both classification and regression tasks. The results we obtained clearly show that composite graph neural networks are far more efficient in this setting than standard graph neural networks.

Code Similarity Prediction Model for Industrial Management Features Based on Graph Neural Networks

The code of industrial management software typically features few system API calls and a high number of customized variables and structures. This makes the similarity of such codes difficult to compute using text features or traditional neural network methods. In this paper, we propose an FSPS-GNN model, which is based on graph neural networks (GNNs), to address this problem. The model categorizes code features into two types, outer graph and inner graph, and conducts training and prediction with four stages-feature embedding, feature enhancement, feature fusion, and similarity prediction. Moreover, differently structured GNNs were used in the embedding and enhancement stages, respectively, to increase the interaction of code features. Experiments with code from three open-source projects demonstrate that the model achieves an average precision of 87.57% and an F0.5 Score of 89.12%. Compared to existing similarity-computation models based on GNNs, this model exhibits a Mean Squared Error (MSE) that is approximately 0.0041 to 0.0266 lower and an F0.5 Score that is 3.3259% to 6.4392% higher. It broadens the application scope of GNNs and offers additional insights for the study of code-similarity issues.

The Millennia-Long Development of Drugs Associated with the 80-Year-Old Artificial Intelligence Story: The Therapeutic Big Bang?

The journey of drug discovery (DD) has evolved from ancient practices to modern technology-driven approaches, with Artificial Intelligence (AI) emerging as a pivotal force in streamlining and accelerating the process. Despite the vital importance of DD, it faces challenges such as high costs and lengthy timelines. This review examines the historical progression and current market of DD alongside the development and integration of AI technologies. We analyse the challenges encountered in applying AI to DD, focusing on drug design and protein-protein interactions. The discussion is enriched by presenting models that put forward the application of AI in DD. Three case studies are highlighted to demonstrate the successful application of AI in DD, including the discovery of a novel class of antibiotics and a small-molecule inhibitor that has progressed to phase II clinical trials. These cases underscore the potential of AI to identify new drug candidates and optimise the development process. The convergence of DD and AI embodies a transformative shift in the field, offering a path to overcome traditional obstacles. By leveraging AI, the future of DD promises enhanced efficiency and novel breakthroughs, heralding a new era of medical innovation even though there is still a long way to go.

Improving the performance and interpretability on medical datasets using graphical ensemble feature selection

MOTIVATION

A major hindrance towards using Machine Learning (ML) on medical datasets is the discrepancy between a large number of variables and small sample sizes. While multiple feature selection techniques have been proposed to avoid the resulting overfitting, overall ensemble techniques offer the best selection robustness. Yet, current methods designed to combine different algorithms generally fail to leverage the dependencies identified by their components. Here, we propose Graphical Ensembling (GE), a graph-theory-based ensemble feature selection technique designed to improve the stability and relevance of the selected features.

RESULTS

Relying on four datasets, we show that GE increases classification performance with fewer selected features. For example, on rheumatoid arthritis patient stratification, GE outperforms the baseline methods by 9% Balanced Accuracy while relying on fewer features. We use data on sub-cellular networks to show that the selected features (proteins) are closer to the known disease genes, and the uncovered biological mechanisms are more diversified. By successfully tackling the complex correlations between biological variables, we anticipate that GE will improve the medical applications of ML.

AVAILABILITY AND IMPLEMENTATION

https://github.com/ebattistella/auto_machine_learning.

DAGCN: Dynamic and Adaptive Graph Convolutional Network for Salient Object Detection

Deep-learning-based salient object detection (SOD) has achieved significant success in recent years. The SOD focuses on the context modeling of the scene information, and how to effectively model the context relationship in the scene is the key. However, it is difficult to build an effective context structure and model it. In this article, we propose a novel SOD method called dynamic and adaptive graph convolutional network (DAGCN) that is composed of two parts, adaptive neighborhood-wise graph convolutional network (AnwGCN) and spatially restricted K-nearest neighbors (SRKNN). The AnwGCN is novel adaptive neighborhood-wise graph convolution, which is used to model and analyze the saliency context. The SRKNN constructs the topological relationship of the saliency context by measuring the non-Euclidean spatial distance within a limited range. The proposed method constructs the context relationship as a topological graph by measuring the distance of the features in the non-Euclidean space, and conducts comparative modeling of context information through AnwGCN. The model has the ability to learn the metrics from features and can adapt to the hidden space distribution of the data. The description of the feature relationship is more accurate. Through the convolutional kernel adapted to the neighborhood, the model obtains the structure learning ability. Therefore, the graph convolution process can adapt to different graph data. Experimental results demonstrate that our solution achieves satisfactory performance on six widely used datasets and can also effectively detect camouflaged objects. Our code will be available at: https://github.com/CSIM-LUT/DAGCN.git.

Early Predicting Osteogenic Differentiation of Mesenchymal Stem Cells Based on Deep Learning Within One Day

Osteogenic differentiation of mesenchymal stem cells (MSCs) is proposed to be critical for bone tissue engineering and regenerative medicine. However, the current approach for evaluating osteogenic differentiation mainly involves immunohistochemical staining of specific markers which often can be detected at day 5-7 of osteogenic inducing. Deep learning (DL) is a significant technology for realizing artificial intelligence (AI). Computer vision, a branch of AI, has been proved to achieve high-precision image recognition using convolutional neural networks (CNNs). Our goal was to train CNNs to quantitatively measure the osteogenic differentiation of MSCs. To this end, bright-field images of MSCs during early osteogenic differentiation (day 0, 1, 3, 5, and 7) were captured using a simple optical phase contrast microscope to train CNNs. The results showed that the CNNs could be trained to recognize undifferentiated cells and differentiating cells with an accuracy of 0.961 on the independent test set. In addition, we found that CNNs successfully distinguished differentiated cells at a very early stage (only 1 day). Further analysis showed that overall morphological features of MSCs were the main basis for the CNN classification. In conclusion, MSCs differentiation detection can be achieved early and accurately through simple bright-field images and DL networks, which may also provide a potential and novel method for the field of cell detection in the near future.

Self-Supervised Node Representation Learning via Node-to-Neighbourhood Alignment

Self-supervised node representation learning aims to learn node representations from unlabelled graphs that rival the supervised counterparts. The key towards learning informative node representations lies in how to effectively gain contextual information from the graph structure. In this work, we present simple-yet-effective self-supervised node representation learning via aligning the hidden representations of nodes and their neighbourhood. Our first idea achieves such node-to-neighbourhood alignment by directly maximizing the mutual information between their representations, which, we prove theoretically, plays the role of graph smoothing. Our framework is optimized via a surrogate contrastive loss and a Topology-Aware Positive Sampling (TAPS) strategy is proposed to sample positives by considering the structural dependencies between nodes, which enables offline positive selection. Considering the excessive memory overheads of contrastive learning, we further propose a negative-free solution, where the main contribution is a Graph Signal Decorrelation (GSD) constraint to avoid representation collapse and over-smoothing. The GSD constraint unifies some of the existing constraints and can be used to derive new implementations to combat representation collapse. By applying our methods on top of simple MLP-based node representation encoders, we learn node representations that achieve promising node classification performance on a set of graph-structured datasets from small- to large-scale.

HealthGAT: Node Classifications in Electronic Health Records using Graph Attention Networks

While electronic health records (EHRs) are widely used across various applications in healthcare, most applications use the EHRs in their raw (tabular) format. Relying on raw or simple data pre-processing can greatly limit the performance or even applicability of downstream tasks using EHRs. To address this challenge, we present HealthGAT, a novel graph attention network framework that utilizes a hierarchical approach to generate embeddings from EHR, surpassing traditional graph-based methods. Our model iteratively refines the embeddings for medical codes, resulting in improved EHR data analysis. We also introduce customized EHR-centric auxiliary pre-training tasks to leverage the rich medical knowledge embedded within the data. This approach provides a comprehensive analysis of complex medical relationships and offers significant advancement over standard data representation techniques. HealthGAT has demonstrated its effectiveness in various healthcare scenarios through comprehensive evaluations against established methodologies. Specifically, our model shows outstanding performance in node classification and downstream tasks such as predicting readmissions and diagnosis classifications.

Discovering dynamic laws from observations: The case of self-propelled, interacting colloids

Active matter spans a wide range of time and length scales, from groups of cells and synthetic self-propelled colloids to schools of fish and flocks of birds. The theoretical framework describing these systems has shown tremendous success in finding universal phenomenology. However, further progress is often burdened by the difficulty of determining forces controlling the dynamics of individual elements within each system. Accessing this local information is pivotal for the understanding of the physics governing an ensemble of active particles and for the creation of numerical models capable of explaining the observed collective phenomena. In this work, we present ActiveNet, a machine-learning tool consisting of a graph neural network that uses the collective motion of particles to learn active and two-body forces controlling their individual dynamics. We verify our approach using numerical simulations of active Brownian particles, active particles undergoing underdamped Langevin dynamics, and chiral active Brownian particles considering different interaction potentials and values of activity. Interestingly, ActiveNet can equally learn conservative or nonconservative forces as well as torques. Moreover, ActiveNet has proven to be a useful tool to learn the stochastic contribution to the forces, enabling the estimation of the diffusion coefficients. Therefore, all coefficients of the equation of motion of Active Brownian Particles are captured. Finally, we apply ActiveNet to experiments of electrophoretic Janus particles, extracting the active and two-body forces controlling colloids' dynamics. On the one side, we have learned that the active force depends on the electric field and area fraction. On the other side, we have also discovered a dependence of the two-body interaction with the electric field that leads us to propose that the dominant force between active colloids is a screened electrostatic interaction with a constant length scale. We believe that the proposed methodological tool, ActiveNet, might open a new avenue for the study and modeling of experimental suspensions of active particles.

A comprehensive survey on protein-ligand binding site prediction

Protein-ligand binding site prediction is critical for protein function annotation and drug discovery. Biological experiments are time-consuming and require significant equipment, materials, and labor resources. Developing accurate and efficient computational methods for protein-ligand interaction prediction is essential. Here, we summarize the key challenges associated with ligand binding site (LBS) prediction and introduce recently published methods from their input features, computational algorithms, and ligand types. Furthermore, we investigate the specificity of allosteric site identification as a particular LBS type. Finally, we discuss the prospective directions for machine learning-based LBS prediction in the near future.

Learning Graph Representations Through Learning and Propagating Edge Features

Graph convolutional networks have achieved considerable success in various graph domain tasks. Recently, numerous types of graph convolutional networks have been developed. A typical rule for learning a node's feature in these graph convolutional networks is to aggregate node features from the node's local neighborhood. However, in these models, the interrelation information between adjacent nodes is not well-considered. This information could be helpful to learn improved node embeddings. In this article, we present a graph representation learning framework that generates node embeddings through learning and propagating edge features. Instead of aggregating node features from a local neighborhood, we learn a feature for each edge and update a node's representation by aggregating local edge features. The edge feature is learned from the concatenation of the edge's starting node feature, the input edge feature, and the edge's end node feature. Unlike node feature propagation-based graph networks, our model propagates different features from a node to its neighbors. In addition, we learn an attention vector for each edge in aggregation, enabling the model to focus on important information in each feature dimension. By learning and aggregating edge features, the interrelation between a node and its neighboring nodes is integrated in the aggregated feature, which helps learn improved node embeddings in graph representation learning. Our model is evaluated on graph classification, node classification, graph regression, and multitask binary graph classification on eight popular datasets. The experimental results demonstrate that our model achieves improved performance compared with a wide variety of baseline models.

DWSSA: Alleviating over-smoothness for deep Graph Neural Networks

Graph Neural Networks (GNNs) have demonstrated great potential in achieving outstanding performance in various graph-related tasks, e.g., graph classification and link prediction. However, most of them suffer from the following issue: shallow networks capture very limited knowledge. Prior works design deep GNNs with more layers to solve the issue, which however introduces a new challenge, i.e., the infamous over-smoothness. Graph representation over emphasizes node features but only considers the static graph structure with a uniform weight are the key reasons for the over-smoothness issue. To alleviate the issue, this paper proposes a Dynamic Weighting Strategy (DWS) for addressing over-smoothness. We first employ Fuzzy C-Means (FCM) to cluster all nodes into several groups and get each node's fuzzy assignment, based on which a novel metric function is devised for dynamically adjusting the aggregation weights. This dynamic weighting strategy not only enables the intra-cluster interactions, but also inter-cluster aggregations, which well addresses undifferentiated aggregation caused by uniform weights. Based on DWS, we further design a Structure Augmentation (SA) step for addressing the issue of underutilizing the graph structure, where some potentially meaningful connections (i.e., edges) are added to the original graph structure via a parallelable KNN algorithm. In general, the optimized Dynamic Weighting Strategy with Structure Augmentation (DWSSA) alleviates over-smoothness by reducing noisy aggregations and utilizing topological knowledge. Extensive experiments on eleven homophilous or heterophilous graph benchmarks demonstrate the effectiveness of our proposed method DWSSA in alleviating over-smoothness and enhancing deep GNNs performance.