Explanations play a central role in shaping users’ trust and acceptance of autonomous vehicles (AVs). While existing AV explanation research has emphasized cognitive elements such as content, timing, and presentation fidelity, it offers limited guidance on how explanations might incorporate affective elements or adjust to varying driving contexts. To address this gap, we introduce a stance-strategy-tone framework for designing affective explanations, supported by scenario-specific guidelines and illustrative example utterances. Through interviews with seven domain experts and six co-design workshops involving 27 prospective AV users, we identified the components that influence how affective explanations are constructed and mapped them onto key driving scenarios. Our findings reveal design opportunities such as tailoring emotional framing to situational demands, combining empathy with informational clarity, and calibrating tone to balance warmth with directive precision. The study provides practical guidance for creating emotionally responsive explanation systems for AVs.

Medical image segmentation plays a crucial role in clinical diagnosis, lesion quantification, and preoperative planning. However, existing Mamba-based architectures, which rely on fixed-direction sequence modeling and flatten images into one-dimensional (1D) sequences, struggle to capture hierarchical anatomical features and spatial dependencies, thereby limiting their representational capacity for complex medical structures. To address these limitations, we propose EccoMamba (Enhanced Cross-hierarchical Continuity Orthogonal Mamba), a U-shaped encoder--decoder framework designed for medical image segmentation. In the encoder's downsampling path, we introduce a Hierarchical Aggregation Enhancement (HAE) module that integrates multi-scale convolutions with hierarchical attention mechanisms. The attention branch further incorporates cross-channel interactions, allowing the model to selectively enhance semantically relevant features while suppressing irrelevant background responses. For skip connections, we design a Structural Continuity Orthogonal (SCO) module to preserve spatial continuity by modeling cross-dimensional dependencies via orthogonal Axial Shifts (AS), thereby mitigating directional bias and improving anatomical consistency. Extensive experiments on four benchmark datasets---ISIC 2018, ISIC 2017, Synapse, and ACDC---show that EccoMamba consistently outperforms state-of-the-art methods in both segmentation accuracy and structural fidelity.

Accurate prediction of compound protein interactions (CPIs) is crucial for drug discovery. However, existing deep learning-based methods suffer from hidden biases and poor cross-domain generalization, leading to spurious correlations and inadequate representation of unseen compound-protein pairs. To address these limitations, we propose FuseMine, a multimodal deep learning framework that jointly leverages molecular structures and biological sequences for reliable CPI prediction. Specifically, FuseMine adopts a dual-representation strategy for each molecule. It employs a convolutional encoder to capture structural features, combined with pretrained large language models for extracting semantic information from sequences. We propose a novel Multi-modal Feature Orchestration Aggregation (MFOA) module that enables deep and synergistic fusion between the structural features and the sequential semantics of molecules, effectively capturing the complementary patterns across modalities. Additionally, we design a Reduction Differential Feature Mining (RDFM) module to further enhance the representation of discriminative features, thereby improving the model’s generalization capability. Extensive experiments on multiple benchmark datasets demonstrate that our framework consistently outperforms state-of-the-art methods in both intra-domain and cross-domain scenarios. These results highlight the synergistic value of combining structural and sequential data for CPIs.

Single-cell multi-omics data reveal complex cellular states and deepen our understanding of tissue cell phenotypes and functions. However, data analysis remains challenging due to the discrete nature and high noise level of the data, as well as the lack of modality. Here, we propose scMultiNet, a multi-task deep adversarial neural network that can integrate different tasks to analyze single-cell multi-modal data. In particular, we achieve joint training of multi-modal integration and cross-modal prediction tasks by introducing a cross-modal bi-prediction module and a multi-head self-attention module. Data denoising is further enhanced by integrating an indicator matrix that constrains and precisely reconstructs the original expression values. Extensive simulations and real data experiments demonstrate that scMultiNet outperforms existing state-of-the-art methods in dimensionality reduction, visualization, clustering, batch elimination, data denoising, multi-modal integration, single-cell cross-modality translation, and in revealing cell type-specific biological insights. In addition, we demonstrate that scMultiNet can effectively transfer the complex relationships between modalities from one batch to another. In summary, scMultiNet stands as a comprehensive end-to-end framework, ideally suited for analyzing single-cell multi-omics data.

MOTIVATION: Accurately identifying compound-protein interactions (CPIs) is critical for accelerating drug discovery. Recent deep learning methods have achieved impressive results, yet they primarily focus on local structures and neighborhood information, often overlooking high-order interaction patterns shared among similar molecules. RESULTS: In this paper, we propose HKD-CPI, a high-order knowledge-enhanced inductive framework designed to improve generalization to unseen compound-protein pairs. Specifically, HKD-CPI introduces a molecular graph tokenization mechanism that aligns compound molecular graph features with token embeddings from sequence-pretrained large language models (LLMs), effectively infusing sequence-derived semantics into structural representations. To capture shared interaction patterns among functionally similar biomolecules, we construct a hypergraph-based representation to model high-order relationships between feature-similar compound/protein groups and their binding partners. Furthermore, a knowledge distillation strategy is further adopted to transfer high-order interaction knowledge from the hypergraph to a lightweight student model, enabling efficient and robust CPI prediction. Extensive experiments demonstrate that HKD-CPI outperforms existing state-of-the-art methods in inductive CPI prediction tasks. In particular, it achieves an average improvement of 4.94% in AUROC and 3.64% in AUPRC over the best-performing baseline across five benchmark datasets. AVAILABILITY AND IMPLEMENTATION: Our code and data are available at https://github.com/Hezy618/HKD-CPI.

MOTIVATION: Drug combinations are crucial for overcoming resistance in cancer therapy. Although deep learning has achieved strong performance in synergy prediction, existing models often treat cell-specific features and paired drugs as a static background and fail to capture how the specific cell-drug environment dynamically modulates drug representations, thereby hindering the modeling of environment-specific synergistic effects. RESULTS: We propose Env-Syn, a framework for modeling drug-drug-cell interactions through Environment-Conditioned Feature Modulation, which incorporates a Residual Feature-wise Linear Modulation (R-FiLM) module to perform precise affine transformations on drug representations conditioned on paired drugs and cellular environments. Benchmark evaluations show that Env-Syn consistently outperforms state-of-the-art methods. Notably, the model exhibits exceptional generalization performance in rigorous inductive scenarios. It maintains high predictive accuracy for unseen drugs with AUROC and AUPRC exceeding 0.81 in the Leave-drug-out setting and further demonstrates strong cross-dataset reliability by surpassing a recall of 0.7 on independent test set. Furthermore, among 15 novel predicted drug combinations, 8 are directly supported by literature evidence. These results demonstrate that Env-Syn is an effective computational tool for drug synergy discovery. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://github.com/AnQi-87/Env-Syn.

Leveraging the wealth of biomedical data available, we can derive insights into the relationships between biological entities from various angles. This underscores the complexity and significance of developing a dynamic approach for integrating data from multiple sources, a critical endeavor in drug recommendation. In this study, we introduce an innovative deep learning approach termed "Multi-View Fusion Strategy with Self-Attention" (FusionMVSA), designed to predict associations between drugs and diseases. To effectively amalgamate data from diverse sources and extract representative features, we have developed a feature extraction mechanism that capitalizes on similarities. This mechanism computes self-attention across multiple perspectives using shared group parameters, thereby highlighting common characteristics. Simultaneously, we utilize biomedical similarities among multi-source data as guiding factors for calculating similarity, enabling the capture of more nuanced features. Subsequently, we integrate these features through a feature fusion process, where known associations between drugs and diseases act as guiding terms. This strategy allows us to uncover the complementary aspects of different viewpoints. Ultimately, we predict potential drug-disease associations using a multi-layer perceptron neural network. Our methodology has undergone rigorous testing through various cross-validation experiments and case studies. We are confident that FusionMVSA will prove to be a valuable tool in drug recommendation, offering new avenues for exploration and discovery in the quest to combat diseases.

Designing new molecules with ideal properties is a critical task in drug discovery. In recent years, the accumulation of available molecular datasets has facilitated the widespread application of deep generative models in drug design. Nonetheless, a significant challenge remains in developing highperformance generative models that not only need to produce chemically valid molecular structures but also optimize the chemical properties of the generated molecules. In this study, we introduce TransFVAE, a graph Transformer-based flow Variational AutoEncoder tailored for molecular graph generation. Our approach employs VAE as the encoder and integrates a lightweight flow model as the decoder. The encoder is strategically designed to expedite the training process of the decoder, while the decoder reciprocally enhances the performance of the encoder. Unlike some existing models that only account for local node connections, our model leverages Transformer architecture in the encoder, ensuring comprehensive consideration of global information. This enables each atom to holistically interact with all other atoms, thereby enhancing molecular attribute constraint optimization in molecular optimization tasks. Validation of our model is conducted through three core tasks: molecule generation and reconstruction, latent space visualization, and molecular optimization. The results affirm the state-of-the-art performance of our model, underscoring its substantial potential in facilitating the generation of drug molecules endowed with desired properties. All source datasets and codes can be downloaded from: https://github.com/Biowust/TransFVAE.

Identifying synergistic drug combinations is crucial for improving cancer therapies. However, the combinatorial explosion of candidate drug pairs and the complexity of drug-cell interactions make exhaustive experimental validation infeasible. Existing deep learning approaches often rely solely on twodimensional molecular graphs or static cell line expression profiles, limiting their ability to capture spatial molecular conformations and dynamic cellular responses. To overcome these limitations, we propose ConGraSyn, a conformation enhanced graph attention framework for drug synergy prediction. For each drug, ConGraSyn integrates descriptor-based fingerprints with molecular graphs, where atom features encode interatomic distances through a scale attention mechanism. A gated fusion module adaptively combines fingerprint-level and graph-level embeddings into a unified drug representation. For cell lines, static protein-protein interaction-informed embeddings are augmented with drug-induced perturbations simulated from basal expression and drug embeddings. The resulting drug and cell features are concatenated and fed into a classifier to predict synergistic outcomes. Across benchmark datasets, ConGraSyn matches or exceeds representative deep learning baselines and consistently outperforms classical machine learning methods. Ablation and sensitivity analyses attribute the gains to its conformation-aware drug modeling and perturbation-informed cell representations. Leave-one-out validation, representation visualizations and case studies indicate robust generalization and biologically plausible novel synergies, supporting its utility for identifying effective drug combinations. The complete source code and datasets are publicly available at: https://github.com/HuazeLoong/ConGraSyn.

In real-world applications where knowledge systems continuously evolve with new relationships, current knowledge graph completion (KGC) methods face a fundamental limitation: their over-reliance on localized patterns and inability to capture structural semantic correlations lead to significant performance degradation when encountering unseen relations. This manifests particularly in their failure to effectively predict potential entity interactions for these novel relations. To bridge this gap, we propose BiRMet, a novel framework integrating bidirectional relational fusion with meta-learning. The framework innovates through relational graph structures that preserve semantic relationships, multi-head attention mechanisms modeling complex relational interactions, and bidirectional feature fusion enabling dynamic co-adaptation of entities and relations. When combined with meta-learning, these components collectively address the core challenge of generalizing to unseen relations. Experimental results across multiple benchmarks demonstrate consistent improvements over existing approaches in handling unseen relations. Furthermore, case studies on drug repurposing based on a biomedical knowledge graph further highlight the potential of our framework in accelerating real-world drug discovery. Codes are available at https://anonymous.4open.science/r/BiRMet-DE01

Therapeutic peptides show promise in targeting previously undruggable binding sites, with recent advancements in deep generative models enabling full-atom peptide co-design for specific protein receptors.However, the critical role of molecular surfaces in proteinprotein interactions (PPIs) has been underexplored.To bridge this gap, we propose an omni-design peptides generation paradigm, called SurfFlow, a novel surface-based generative algorithm that enables comprehensive co-design of sequence, structure, and surface for peptides.SurfFlow employs a multi-modality conditional flow matching (CFM) architecture to learn distributions of surface geometries and biochemical properties, enhancing peptide binding accuracy.Evaluated on the comprehensive PepMerge benchmark, SurfFlow consistently outperforms full-atom baselines across all metrics.These results highlight the advantages of considering molecular surfaces in de novo peptide discovery and demonstrate the potential of integrating multiple protein modalities for more effective therapeutic peptide discovery.

MOTIVATION: Accurate prediction of protein-ligand binding (PLB) relationships plays a crucial role in drug discovery, which helps identify drugs that modulate the activity of specific targets. Traditional biological assays for measuring PLB relationships are time consuming and costly. In addition, models for predicting PLB relationships have been developed and widely used in drug discovery tasks. However, learning more accurate PLB representations is essential to meet the stringent standards required for drug discovery. RESULTS: We propose an image-based PLB representation learning framework, called ImagePLB, which equips ligand representation learner (LRL) and protein representation learner (PRL) to accept 3D multi-view ligand images and protein graphs as input, respectively, and learns rich interaction information between ligand and protein through a binding representation learner (BRL). Considering the scarcity of protein-ligand pairs, we further propose a multi-level next trajectory prediction (MLNTP) task to pre-train ImagePLB on the 4D flexible dynamics trajectory of 16 972 complexes, including ligand level, protein level, and complex level, to learn information related to trajectories. Besides, by introducing trajectory regularization (TR), we effectively alleviate the problem of high (even almost identical) feature similarity caused by adjacent trajectories. Compared with the current state-of-the-art methods, ImagePLB has achieved competitive improvements on PLB-related prediction tasks, including protein-ligand affinity and efficacy prediction tasks. This study opens the door to the image-based PLB learning paradigm. AVAILABILITY AND IMPLEMENTATION: All data and implementation details of code can be obtained from https://github.com/HongxinXiang/ImagePLB.

Accurate prediction of the drug binding between proteins and ligands can significantly advance the development of structure-based drug design. Recent advances have shown great potential in applying equivariant graph neural network (EGNN) -based methods to learn representations of protein-ligand (PL) complexes. However, most of them typically focus on atom-level graph representations and omit the residue-level information in PL complexes, which are considered essential for understanding the binding mechanism. In this article, we develop a SO(3)-equivariant hierarchical graph neural network (EHGNN) that effectively captures the intrinsic hierarchy of biomolecular structures to enhance the predictive performance of PL interactions. Based on the SO(3)-EHGNN, we further propose a molecular dynamics-powered and energy-guided deep learning framework, called Dynamics-PLI, to capture the spatial structures and energetic information inside molecular dynamic (MD) trajectories. Extensive experimental results show significant improvements over current state-of-the-art methods, with a decrease of 4.03% in RMSE for the binding affinity problem and an average increase of 3.95% in AUROC and AUPRC for the ligand efficacy problem, demonstrating the superiority of Dynamics-PLI for PL interaction prediction. Our findings indicate that the SO(3)-EHGNN exhibits enhanced performance without the necessity of pre-training, emphasizing the inherent analytical strength of SO(3)-EHGNN.

Machine learning catalyzes a revolution in chemical and biological science. However, its efficacy is heavily dependent on the availability of labeled data, and annotating biochemical data is extremely laborious. To surmount this data sparsity challenge, we present an instructive learning algorithm named InstructMol to measure pseudo-labels' reliability and help the target model leverage large-scale unlabeled data. InstructMol does not require transferring knowledge between multiple domains, which avoids the potential gap between the pretraining and fine-tuning stages. We demonstrated the high accuracy of InstructMol on several real-world molecular datasets and out-of-distribution (OOD) benchmarks.

Current methods of molecular image-based drug discovery face two major challenges: (1) work effectively in absence of labels, and (2) capture chemical structure from implicitly encoded images. Given that chemical structures are explicitly encoded by molecular graphs (such as nitrogen, benzene rings and double bonds), we leverage self-supervised contrastive learning to transfer chemical knowledge from graphs to images. Specifically, we propose a novel Contrastive Graph-Image Pre-training (CGIP) framework for molecular representation learning, which learns explicit information in graphs and implicit information in images from large-scale unlabeled molecules via carefully designed intra- and inter-modal contrastive learning. We evaluate the performance of CGIP on multiple experimental settings (molecular property prediction, cross-modal retrieval and distribution similarity), and the results show that CGIP can achieve state-of-the-art performance on all 12 benchmark datasets and demonstrate that CGIP transfers chemical knowledge in graphs to molecular images, enabling image encoder to perceive chemical structures in images. We hope this simple and effective framework will inspire people to think about the value of image for molecular representation learning.

The powerful combination of large-scale drug-related interaction networks and deep learning provides new opportunities for accelerating the process of drug discovery. However, chemical structures that play an important role in drug properties and high-order relations that involve a greater number of nodes are not tackled in current biomedical networks. In this study, we present a general hypergraph learning framework, which introduces Drug-Substructures relationship into Molecular interaction Networks to construct the micro-to-macro drug centric heterogeneous network (DSMN), and develop a multi-branches HyperGraph learning model, called HGDrug, for Drug multi-task predictions. HGDrug achieves highly accurate and robust predictions on 4 benchmark tasks (drug-drug, drug-target, drug-disease, and drug-side-effect interactions), outperforming 8 state-of-the-art task specific models and 6 general-purpose conventional models. Experiments analysis verifies the effectiveness and rationality of the HGDrug model architecture as well as the multi-branches setup, and demonstrates that HGDrug is able to capture the relations between drugs associated with the same functional groups. In addition, our proposed drug-substructure interaction networks can help improve the performance of existing network models for drug-related prediction tasks.

Accumulated studies discovered that various microbes in human bodies were closely related to complex human diseases and could provide new insight into drug development. Multiple computational methods were constructed to predict microbes that were potentially associated with diseases. However, most previous methods were based on single characteristics of microbes or diseases, that lacked important biological information related to microorganisms or diseases. Therefore, we constructed a knowledge graph centered on microorganisms and diseases from several existed databases to provide knowledgeable information for microbes and diseases. Then, we adopted a graph neural network method to learn representations of microbes and diseases from the constructed knowledge graph. After that, we introduced the Gaussian kernel similarity features of microbes and diseases to generate final representations of microbes and diseases. At last, we proposed a score function on final representations of microbes and diseases to predict scores of microbe-disease associations. Comprehensive experiments on the Human Microbe-Disease Association Database (HMDAD) dataset had demonstrated that our approach outperformed baseline methods. Furthermore, we implemented case studies on two important diseases (asthma and inflammatory bowel disease), the result demonstrated that our proposed model was effective in revealing the relationship between diseases and microbes. The source code of our model and the data were available on https://github.com/ChangzhiJiang/KGNMDA_master.

Drug-target interactions (DTIs) prediction research presents important significance for promoting the development of modern medicine and pharmacology. Traditional biochemical experiments for DTIs prediction confront the challenges including long time period, high cost and high failure rate, and finally leading to a low-drug productivity. Chemogenomic-based computational methods can realize high-throughput prediction. In this study, we develop a deep collaborative filtering prediction model with multiembeddings, named DCFME (deep collaborative filtering prediction model with multiembeddings), which can jointly utilize multiple feature information from multiembeddings. Two different representation learning algorithms are first employed to extract heterogeneous network features. DCFME uses the generated low-dimensional dense vectors as input, and then simulates the drug-target relationship from the perspective of both couplings and heterogeneity. In addition, the model employs focal loss that concentrates the loss on sparse and hard samples in the training process. Comparative experiments with five baseline methods show that DCFME achieves more significant performance improvement on sparse datasets. Moreover, the model has better robustness and generalization capacity under several harder prediction scenarios.

Identifying new lead molecules to treat cancer requires more than a decade of dedicated effort. Before selected drug candidates are used in the clinic, their anti-cancer activity is generally validated by in vitro cellular experiments. Therefore, accurate prediction of cancer drug response is a critical and challenging task for anti-cancer drugs design and precision medicine. With the development of pharmacogenomics, the combination of efficient drug feature extraction methods and omics data has made it possible to use computational models to assist in drug response prediction. In this study, we propose DeepTTA, a novel end-to-end deep learning model that utilizes transformer for drug representation learning and a multilayer neural network for transcriptomic data prediction of the anti-cancer drug responses. Specifically, DeepTTA uses transcriptomic gene expression data and chemical substructures of drugs for drug response prediction. Compared to existing methods, DeepTTA achieved higher performance in terms of root mean square error, Pearson correlation coefficient and Spearman's rank correlation coefficient on multiple test sets. Moreover, we discovered that anti-cancer drugs bortezomib and dactinomycin provide a potential therapeutic option with multiple clinical indications. With its excellent performance, DeepTTA is expected to be an effective method in cancer drug design.

The interaction between microribonucleic acid and long non-coding ribonucleic acid plays a very important role in biological processes, and the prediction of the one is of great significance to the study of its mechanism of action. Due to the limitations of traditional biological experiment methods, more and more computational methods are applied to this field. However, the existing methods often have problems, such as inadequate acquisition of potential features of the sequence due to simple coding and the need to manually extract features as input. We propose a deep learning model, preMLI, based on rna2vec pre-training and deep feature mining mechanism. We use rna2vec to train the ribonucleic acid (RNA) dataset and to obtain the RNA word vector representation and then mine the RNA sequence features separately and finally concatenate the two feature vectors as the input of the prediction task. The preMLI performs better than existing methods on benchmark datasets and has cross-species prediction capabilities. Experiments show that both pre-training and deep feature mining mechanisms have a positive impact on the prediction performance of the model. To be more specific, pre-training can provide more accurate word vector representations. The deep feature mining mechanism also improves the prediction performance of the model. Meanwhile, The preMLI only needs RNA sequence as the input of the model and has better cross-species prediction performance than the most advanced prediction models, which have reference value for related research.

MOTIVATION: Drug-drug interaction (DDI) prediction is a challenging problem in pharmacology and clinical applications. With the increasing availability of large biomedical databases, large-scale biological knowledge graphs containing drug information have been widely used for DDI prediction. However, large knowledge graphs inevitably suffer from data noise problems, which limit the performance and interpretability of models based on the knowledge graph. Recent studies attempt to improve models by introducing inductive bias through an attention mechanism. However, they all only depend on the topology of entity nodes independently to generate fixed attention pathways, without considering the semantic diversity of entity nodes in different drug pair links. This makes it difficult for models to select more meaningful nodes to overcome data quality limitations and make more interpretable predictions. RESULTS: To address this issue, we propose a Link-aware Graph Attention method for DDI prediction, called LaGAT, which is able to generate different attention pathways for drug entities based on different drug pair links. For a drug pair link, the LaGAT uses the embedding representation of one of the drugs as a query vector to calculate the attention weights, thereby selecting the appropriate topological neighbor nodes to obtain the semantic information of the other drug. We separately conduct experiments on binary and multi-class classification and visualize the attention pathways generated by the model. The results prove that LaGAT can better capture semantic relationships and achieves remarkably superior performance over both the classical and state-of-the-art models on DDI prediction. AVAILABILITYAND IMPLEMENTATION: The source code and data are available at https://github.com/Azra3lzz/LaGAT. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The discovery and confirmation o f d rug-target interactions plays an important role for multiple pharmacology, drug discovery, drug repositioning, side effect prediction and drug resistance. In this paper, we propose a meta-path-based collaborative attention prediction model that effectively learns the explicit representation of drugs, targets, and meta-path contexts to predict the potential relationships between drug and target. Experimental results prove that the model has an excellent performance in predictability and interpretability.

The increase in biological data and the formation of various biomolecule interaction databases enable us to obtain diverse biological networks. These biological networks provide a wealth of raw materials for further understanding of biological systems, the discovery of complex diseases and the search for therapeutic drugs. However, the increase in data also increases the difficulty of biological networks analysis. Therefore, algorithms that can handle large, heterogeneous and complex data are needed to better analyze the data of these network structures and mine their useful information. Deep learning is a branch of machine learning that extracts more abstract features from a larger set of training data. Through the establishment of an artificial neural network with a network hierarchy structure, deep learning can extract and screen the input information layer by layer and has representation learning ability. The improved deep learning algorithm can be used to process complex and heterogeneous graph data structures and is increasingly being applied to the mining of network data information. In this paper, we first introduce the used network data deep learning models. After words, we summarize the application of deep learning on biological networks. Finally, we discuss the future development prospects of this field.

Researches have found that tumor immunotherapy can only work for some patients, and the intestinal microbiota is one of the important factors affecting the responses of patients with cancer to immune checkpoint blockade therapy. It is highly desirable to develop computational methods that can predict whether a patient with cancer will have positive effects on cancer immunotherapy by analyzing intestinal microorganisms of the patient. In this study, a multi-task model is introduced to predict the efficacy of cancer immunotherapy on a patient who suffered from non-small cell lung cancer or renal cell carcinoma. The results demonstrate the multi-task model outperforms several single-task methods. Therefore, we believe the multi-task idea can be used to predict the efficacy of cancer immunotherapy based on the gut microbe, which would be important to cancer patients.