Accurate identification of protein active sites at the residue level is crucial for understanding protein function and advancing drug discovery. However, current methods face two critical challenges: vulnerability in single-instance prediction due to sparse training data, and inadequate modality reliability estimation that leads to performance degradation when unreliable modalities dominate fusion processes. To address these challenges, we introduce Multimodal Mixtureof-Experts with Retrieval Augmentation (MERA), the first retrieval-augmented framework for protein active site identification. MERA employs hierarchical multi-expert retrieval that dynamically aggregates contextual information from chain, sequence, and active-site perspectives through residuelevel mixture-of-experts gating. To prevent modality degradation, we propose a reliability-aware fusion strategy based on Dempster–Shafer evidence theory that quantifies modality trustworthiness through belief mass functions and learnable discounting coefficients, enabling principled multimodal integration. Extensive experiments on ProTAD-Gen and TS125 datasets demonstrate that MERA achieves state-of-the-art performance, with 90% AUPRC on active site prediction and significant gains on peptide-binding site identification, validating the effectiveness of retrieval-augmented multi-expert modeling and reliability-guided fusion

Accurate drug-target affinity (DTA) prediction holds significant potential in the field of artificial intelligence (AI)-based drug discovery. However, existing methods primarily operate at a single scale, specifically at the macro (residue) scale for target proteins and the micro (atom) scale for drugs, which limits their ability to provide information at micro (atom) scale for targets and macro (functional group, FG) scale for drugs. This limitation hinders a comprehensive understanding of the binding patterns and properties of drug-target pairs. In this paper, we propose a progressive Macro-to-Micro 3D Modeling Network (M²N) that enables macro (residue/FG) to micro (atom) scale unified modeling, termed cross-scale, to predict DTA. Specifically, M²N operates drugs by learning their chemical properties and structural characteristics from a 3D FG graph to a 3D atom graph. Correspondingly, M²N encodes proteins from a 3D residue graph to a 3D atom graph to exploit their sequence, evolutionary, and geometric representations. Such cross-scale 3D modeling scheme allows for coarse-to-fine embedding optimization, followed by an adaptive fusion module to dynamically integrate the refined features by end-to-end learning. Extensive experiments on two datasets indicate that M²N not only outperforms state-of-the-art methods under various conditions, but also provides a new paradigm for target and drug unified modeling.

While metabolic imaging can facilitate early diagnosis by revealing physiological changes of lesions, it is limited by high cost, high radiation risk, and potential renal impairment. Thus, developing an effective approach for Anatomical-to-Metabolic Image Synthesis (A2MIS) is highly required. However, existing methods are heavily hindered by the gap between distinct domains, and fail to provide a confidence score for the synthesized images, severely restricting their clinical applications. Here, we propose a novel Residual-guided Latent Brownian-bridge Co-Diffusion (RLBCD) model for A2MIS. Specifically, RLBCD starts with a co-diffusion process that leverages a residual diffusion branch to capture inter-domain differences, which are injected into an enhanced diffusion branch to maximally reconstruct modality-specific details. Furthermore, to explore desired residual guidance, we investigate the encoder and decoder features in diffusion models, and accordingly design a Hybrid-Granularity Fusion to integrate consistent semantics and complementary information for fine-grained reconstruction. Additionally, a latent consistency score is developed to enhance the restoration of modality-specific information, which also serves as an indicator of the inherent confidence of the synthesized images. Extensive experiments conducted on five public and in-house datasets demonstrate that RLBCD not only outperforms state-of-the-art methods for A2MIS, but also is valuable for downstream clinic applications.

Predicting drug-target interactions (DTI) is a fundamental step in drug discovery, where deep learning methods show promising performance. However, current approaches encounter challenges in interpretable generalizing the drug-target interactions and adapting to out-of-distribution data. To solve them, we propose a Multi-domain Awareness Drug-Target Interaction (MADTI) framework, which captures the overlooked shallow concrete features and intrinsic biological traits. Specifically, our design enhances the model’s understanding of shallow features in drug graphs and target sequences while retaining the advantages of existing methods that emphasize deep features. This multi-level domain comprehension, combined with the ability of the Category Awareness Domain Adaption(CADA) module to understand biological patterns in positive and negative samples, improves predictive accuracy in cross-domain scenarios. Additionally, by employing bilinear attention and gated attention to learn the multi-level interaction patterns of drug-target pairs explicitly, we further enhance the model’s biological interpretability. Through extensive experiments, MADTI demonstrates significant improvements over state-of-the-art methods, achieving higher AUC-ROC and PR-AUC in various tasks, including conventional prediction, missing data prediction, single-modality prediction, and clustering-based out-of-distribution prediction. The code are available at https://github.com/lian-xiao/MADTI.

For the purpose of improving the user experience of mobile smart terminals for the elderly, accessibility tools have become an indispensable component of intelligent system frameworks to meet the varying needs of senior citizens for assistance. Nonetheless, the complexity of accessibility assistive tools complicates the lives of some elderly with a lack of cognitive experience to use these tools independently. Our purpose is to study the relationship between the accessibility design elements of smartphones, the emotions, and touch behaviors of elderly users, and to optimize the accessibility design methods of mobile smart terminals. First, we obtained the four interface elements with the most influence on users through a Semantic Differential evaluation involving 50 participants. Second, in a touch event collection experiment, 20 participants’ touch data was collected, and an SVM-based correlation model between interface elements, user emotions, and touch behaviors was established. Third, the prototype of the assistive recommender was iterated employing the combination of the correlation model and Genetic Algorithms. Ultimately, the effectiveness of the six touch media recommendations was compared through usability testing. The experimental results indicates that: (1) Emotion-aware accessibility assistive recommendation systems based on haptic recognition can enhance the elderly’s ability to access information through mobile terminals. (2) Middle-aged and elderly users have a more robust negative emotion reflection for dragging and swiping on touch screens. (3) Negative emotions of elderly users can assist identify design defects of accessibility tools. Our research work distilled a set of design suggestions for digital accessibility improvement, thereby enhancing the usability and inclusiveness of assistive tools in diverse contexts, and reducing the psychological pressure caused by unfamiliar interfaces. This study supplies a reference for improving the emotional experience of elderly people employing mobile terminals and extending the approach to accessible design.

In the field of drug discovery, a proliferation of pre-trained models has surfaced, exhibiting exceptional performance across a variety of tasks. However, the extensive size of these models, coupled with the limited interpretative capabilities of current fine-tuning methods, impedes the integration of pre-trained models into the drug discovery process. This paper pushes the boundaries of pre-trained models in drug discovery by designing a novel fine-tuning paradigm known as the Head Feature Parallel Adapter (HFPA), which is highly interpretable, high-performing, and has fewer parameters than other widely used methods. Specifically, this approach enables the model to consider diverse information across representation subspaces concurrently by strategically using Adapters, which can operate directly within the model's feature space. Our tactic freezes the backbone model and forces various small-size Adapters' corresponding subspaces to focus on exploring different atomic and chemical bond knowledge, thus maintaining a small number of trainable parameters and enhancing the interpretability of the model. Moreover, we furnish a comprehensive interpretability analysis, imparting valuable insights into the chemical area. HFPA outperforms over seven physiology and toxicity tasks and achieves state-of-the-art results in three physical chemistry tasks. We also test ten additional molecular datasets, demonstrating the robustness and broad applicability of HFPA.

Whole Slide Images (WSIs) are paramount in the medical field, with extensive applications in disease diagnosis and treatment. Recently, many deep-learning methods have been used to classify WSIs. However, these methods are inadequate for accurately analyzing WSIs as they treat regions in WSIs as isolated entities and ignore contextual information. To address this challenge, we propose a novel Dual-Granularity Cooperative Diffusion Model (DCDiff) for the precise classification of WSIs. Specifically, we first design a cooperative forward and reverse diffusion strategy, utilizing fine-granularity and coarse-granularity to regulate each diffusion step and gradually improve context awareness. To exchange information between granularities, we propose a coupled U-Net for dual-granularity denoising, which efficiently integrates dual-granularity consistency information using the designed Fine- and Coarse-granularity Cooperative Aware (FCCA) model. Ultimately, the cooperative diffusion features extracted by DCDiff can achieve cross-sample perception from the reconstructed distribution of training samples. Experiments on three public WSI datasets show that the proposed method can achieve superior performance over state-of-the-art methods. The code is available at https://github.com/hemo0826/DCDiff.

Chest X-ray (CXR) is commonly performed as an initial investigation in COVID-19, whose fast and accurate diagnosis is critical. Recently, deep learning has a great potential in detecting people who are suspected to be infected with COVID-19. However, deep learning resulting with black-box models, which often breaks down when forced to make predictions about data for which limited supervised information is available and lack inter-pretability, still is a major barrier for clinical integration. In this work, we hereby propose a semantic-powered explainable model-free few-shot learning scheme to quickly and precisely diagnose COVID-19 with higher reliability and transparency. Specifically, we design a Report Image Explanation Cell (RIEC) to exploit clinically indicators derived from radiology reports as interpretable driver to introduce prior knowledge at training. Meanwhile, multi-task collaborative diagnosis strategy (MCDS) is developed to construct N-way K-shot tasks, which adopts a cyclic and collaborative training approach for producing better generalization performance on new tasks. Extensive experiments demonstrate that the proposed scheme achieves competitive results (accuracy of 98.91%, precision of 98.95%, recall of 97.94% and F1-score of 98.57%) to diagnose COVID-19 and other pneumonia infected categories, even with only 200 paired CXR images and radiology reports for training. Furthermore, statistical results of comparative experiments show that our scheme provides an interpretable window into the COVID-19 diagnosis to improve the performance of the small sample size, the reliability and transparency of black-box deep learning models. Our source codes will be released on https://github.com/AI-medical-diagnosis-team-of-JNU/SPEMFSL-Diagnosis-COVID-19.

More attention has been paid to the precision and personalized treatment of breast cancer, which is a primary risk factor that threatens the females lives. It is momentous for diagnosis, analysis and therapy of tumors to lucubrate breast intratumor heterogeneity. We propose a DCE-MRI dynamic mode based self-supervised dual attention deep clustering network (DADCN) which is utilized to achieve the individual precise segmentation of breast intratumor heterogeneity region in this paper. The specific representations learned by the graph attention network are consciously combined with the deep abstract features extracted from the deep convolutional neural network. Then the structural information of the voxel in breast tumor is mined by spreading on the graph. The model is self-supervised by dual relative loss and residual loss and the clustering graph is measured by graph cut loss. We also employ Pearson, Spearman and Kendall analysis to evaluate degree of correlation between clustering results and intratumor heterogeneity represented by molecular typing. We ultimately detect that the degree of intratumor heterogeneity is automatically determined via segmentation of the heterogeneity region, to accomplish the noninvasive individual molecular typing prediction of breast cancer. The number of clusters in breast intratumor heterogeneity region is an independent biomarker for the diagnosis of benign and malignant tumors and prediction of basal-like molecular typing.