Drug repurposing relies critically on the accurate prediction of drug-target interactions (DTIs). Conventional graph neural network approaches typically model drugs and proteins as isolated nodes, focusing solely on intrinsic attributes such as molecular structure or sequence information. As a result, they often fail to capture complex synergistic effects-such as multi-target regulation. To overcome this limitation, this paper proposes the Topology-Aggregation Enhanced Graph Interaction Network (TAGIN-DTI), a novel framework that shifts the prediction paradigm from a conventional node-level view to a subnetwork-level perspective. The model aggregates topological association features from drug-drug and protein-protein interaction networks via a dual-path Transformer encoder, integrates multi-scale global and local information through a gated fusion mechanism, and incorporates MinHash-based subgraph structure encoding to enhance neighborhood topological representation. Experimental results demonstrate that TAGIN-DTI outperforms existing methods in both prediction accuracy and generalization capability, offering valuable insights for drug repurposing and target discovery.

Drug repositioning is a strategy to identify new therapeutic uses for existing drugs, significantly reducing development costs and time. Although deep learning methods for drug-target interaction prediction have advanced, most models are limited to binary relationships and struggle to capture complex ternary associations among drugs, targets, and diseases. Additionally, limitations in graph structure modeling and node feature representation often constrain their generalization capability. To address these challenges, this paper proposes GraphTransHGN, a framework integrating graph embedding, Transformer feature extraction, and heterogeneous graph neural networks. First, Node2Vec constructs graph representations of drugs and targets, with the Transformer extracting high-level semantic features. Drug-target pairs are then combined into composite nodes (DT_node), and target-disease data are incorporated to form a heterogeneous graph. Finally, the HGTConv models multi-type relationships between DT_node and diseases, enabling end-to-end prediction of potential therapeutic associations. Experimental results demonstrate that GraphTransHGN outperforms mainstream methods across key metrics, achieving an AUC of 0.9916, Recall of 0.9959, and AUPR of 0.9849, which confirms its discriminative power and robustness. The model not only improves prediction accuracy for drug repositioning but also offers a novel technical and theoretical foundation for mechanism-based drug discovery.

Corporate credit rating serves as a crucial intermediary service in the market economy, playing a key role in maintaining economic order. Existing credit rating models rely on financial metrics and deep learning. However, they often overlook insights from non-financial data, such as corporate annual reports. To address this, this paper introduces a corporate credit rating framework that integrates financial data with features extracted from annual reports using FinBERT, aiming to fully leverage the potential value of unstructured text data. In addition, we have developed a large-scale dataset, the Comprehensive Corporate Rating Dataset (CCRD), which combines both traditional financial data and textual data from annual reports. The experimental results show that the proposed method improves the accuracy of the rating predictions by 8–12%, significantly improving the effectiveness and reliability of corporate credit ratings.

Corporate credit rating is a problem of classifying high-dimensional feature vectors. Traditional corporate credit rating methods primarily rely on statistical models, machine learning models, or neural network models, all of which face limitations in handling complex inter-company relationships and nonlinear features. Particularly in cases of insufficient sample sizes, the performance of traditional models is often compromised. This paper proposes an innovative corporate credit rating method, CoReGraph-CR, which integrates Transformer, Graph Neural Network, and contrastive learning techniques to address these limitations. The contrastive learning method effectively addresses the issue of limited labeled data, improving the model’s performance with small sample sizes. Experimental results demonstrate that CoReGraph-CR outperforms existing GNN-based corporate credit rating methods in terms of classification accuracy, recall, and F1 scores, with significant improvements across various credit rating levels. Additionally, interpretability experiments enhance the model’s transparency, making it more trustworthy for practical applications.

BACKGROUND: Drug-drug interactions (DDIs) frequently occur in combination therapy and may cause adverse effects or reduced efficacy. Existing computational approaches often fail to capture both the semantic information in drug sequences and the structural properties of drug molecules, limiting predictive power. RESULTS: We propose MDG-DDI, a deep learning framework that integrates a Frequent Consecutive Subsequence (FCS)-based Transformer encoder with a Deep Graph Network (DGN) to extract complementary semantic and structural features. These representations are fused and fed into a Graph Convolutional Network (GCN) for DDI prediction. Experiments on three benchmark datasets under transductive and inductive settings show that MDG-DDI consistently outperforms state-of-the-art methods, with particularly strong gains when predicting interactions involving unseen drugs. CONCLUSION: By jointly modeling substructure-level semantics and molecular graph structure, MDG-DDI achieves robust and accurate DDI prediction. The framework demonstrates improved generalization and offers potential for enhancing drug safety assessment and discovery.

Space-Air-Ground Integrated Network (SAGIN) has recently attracted extensive attention as a new type of network architecture, which can meet the ever-increasing demands of users for ubiquitous access. However, due to different coverage performance of various networks and the demands of huge capacity in ultra-dense regions, frequent passive group handover will occur, thereby decreasing the quality of service (QoS) and causing signaling storms. To tackle this problem, we introduce low Earth orbit (LEO) satellites, high-altitude platforms and ground base stations to cover ultra-dense regions. The mobility management functions are configured in LEO satellites, which serve as a central controller and compute the average utility based on QoS. We propose an evolutionary game-based vertical handover scheme, where the users covered by SAGINs are modeled as players to compete limited network resources. Simulation results verified the effectiveness of the proposed scheme in meeting the QoS while improving the utility of networks.

Effectively and accurately predicting the effects of interactions between proteins after amino acid mutations is a key issue for understanding the mechanism of protein function and drug design. In this study, we present a deep graph convolution (DGC) network-based framework, DGCddG, to predict the changes of protein-protein binding affinity after mutation. DGCddG incorporates multi-layer graph convolution to extract a deep, contextualized representation for each residue of the protein complex structure. The mined channels of the mutation sites by DGC is then fitted to the binding affinity with a multi-layer perceptron. Experiments with results on multiple datasets show that our model can achieve relatively good performance for both single and multi-point mutations. For blind tests on datasets related to angiotensin-converting enzyme 2 binding with the SARS-CoV-2 virus, our method shows better results in predicting ACE2 changes, may help in finding favorable antibodies. Code and data availability: https://github.com/lennylv/DGCddG.

Proteinprotein interactions are the basis of many cellular biological processes, such as cellular organization, signal transduction, and immune response. Identifying proteinprotein interaction sites is essential for understanding the mechanisms of various biological processes, disease development, and drug design. However, it remains a challenging task to make accurate predictions, as the small amount of training data and severe imbalanced classification reduce the performance of computational methods. We design a deep learning method named ctP2ISP to improve the prediction of proteinprotein interaction sites. ctP2ISP employs Convolution and Transformer to extract information and enhance information perception so that semantic features can be mined to identify proteinprotein interaction sites. A weighting loss function with different sample weights is designed to suppress the preference of the model toward multi-category prediction. To efficiently reuse the information in the training set, a preprocessing of data augmentation with an improved sample-oriented sampling strategy is applied. The trained ctP2ISP was evaluated against current state-of-the-art methods on six public datasets. The results show that ctP2ISP outperforms all other competing methods on the balance metrics: F1, MCC, and AUPRC. In particular, our prediction on open tests related to viruses may also be consistent with biological insights.

MOTIVATION: Histone modifications are epigenetic markers that impact gene expression by altering the chromatin structure or recruiting histone modifiers. Their accurate identification is key to unraveling the mechanisms by which they regulate gene expression. However, the solutions for this task can be improved by exploiting multiple relationships from dataset and exploring designs of learning models, for example jointly learning technology. RESULTS: This article proposes a deep learning-based multi-objective computational approach, iHMnBS, to identify which of the seven typical histone modifications a DNA sequence may choose to bind, and which parts of the DNA sequence bind to them. iHMnBS employs a customized dataset that allows the marking of modifications contained in histones that may bind to any position in the DNA sequence. iHMnBS tries to mine the information implicit in this richer data by means of deep neural networks. In comprehensive comparisons, iHMnBS outperforms a baseline method, and the probability of binding to modified histones assigned to a representative nucleotide of a DNA sequence can serve as a reference for biological experiments. Since the interaction between transcription factors and histone modifications has an important role in gene expression, we extracted a number of sequence patterns that may bind to transcription factors, and explored their possible impact on disease. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://github.com/lennylv/iHMnBS. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.