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.

With the burgeoning development of fields such as the Metaverse, Virtual Reality (VR), and Digital Twins, text-to-3D generation has emerged as a research hotspot in both academia and industry. Currently, optimization methods based on Score Distillation Sampling (SDS) utilizing 2D diffusion priors have become the mainstream technological paradigm in this field. However, due to the view bias of 2D priors and the mode-seeking ambiguity combined with gradient noise induced by high Classifier-Free Guidance (CFG), these methods still suffer from macro-topological inconsistency (e.g., the Janus problem) and micro-geometric discontinuity. To address these challenges, we propose MOC-3D, a text-to-3D generation method based on geometric manifold and semantic view-order consistency. Built upon the ScaleDreamer framework, our method incorporates a Semantic View-Order Constraint Module and a Manifold-based Feature Continuity Module. The former aims to rectify macro-topological inconsistency, while the latter focuses on eliminating micro-geometric discontinuity. Specifically, the Semantic View-Order Constraint Module leverages the prior knowledge of CLIP to impose a Monotonicity Rank Constraint on semantic score representations across different views, thereby providing effective guidance for the global topological structure of 3D objects. Meanwhile, the Manifold-based Feature Continuity Module employs the Riemannian Metric on the Symmetric Positive Definite (SPD) manifold. By measuring the distance of feature statistical distributions in the Riemannian space, it promotes the smooth evolution and continuity of micro-textures across multi-views in a statistical sense. Under the macro-micro synergistic optimization of these two modules, our model can simultaneously improve macro-structural consistency and micro-detail continuity. Experimental results demonstrate that compared with mainstream methods, our approach achieves significant advantages in terms of Semantic Consistency (CLIP Score) and Perceptual Quality (LPIPS). Furthermore, ablation studies verify the independent contributions and complementary effectiveness of the Semantic View-Order Constraint Module in rectifying macro-topological inconsistency and the Manifold-based Feature Continuity Module in eliminating micro-geometric discontinuity.

In recent years, deep learning achieves significant advancements in medical image segmentation. Research finds that integrating Transformers and CNNs effectively addresses the limitations of CNNs in managing long-distance dependencies and understanding global information.However, existing models typically employ a serial approach to combine Transformers and CNNs, which complicates the simultaneous processing of global and local information. To address this, our study proposes a parallel multi-path attention architecture, SWMA-UNET, that integrates Transformers and CNNs. This architecture deeply mines features through parallel strategies while capturing both local details and global context information, thereby enhancing the accuracy of medical image segmentation. Experimental results indicate that our method surpasses all previously reported methods in the literature on the Synapse, ACDC, ISIC 2018 and MoNuSeg datasets.

The analysis of 3D meshes with deep learning has become prevalent in computer graphics. As an essential structure, hierarchical representation is critical for mesh pooling in multiscale analysis. Existing clustering-based mesh hierarchy construction methods involve nonlinear discretization optimization operations, making them nondifferential and challenging to embed in other trainable networks for learning. Inspired by deep superpixel learning methods in image processing, we extend them from 2D images to 3D meshes by proposing a novel differentiable chart-based segmentation method named geodesic differential supervertex (GDSV). The key to the GDSV method is to ensure that the geodesic position updates are differentiable while satisfying the constraint that the renewed supervertices lie on the manifold surface. To this end, in addition to using the differential SLIC clustering algorithm to update the nonpositional features of the supervertices, a reparameterization trick, the Gumbel-Softmax trick, is employed to renew the geodesic positions of the supervertices. Therefore, the geodesic position update problem is converted into a linear matrix multiplication issue. The GDSV method can be an independent module for chart-based segmentation tasks. Meanwhile, it can be combined with the front-end feature learning network and the back-end task-specific network as a plug-in-plug-out module for training; and be applied to tasks such as shape classification, part segmentation, and 3D scene understanding. Experimental results show the excellent performance of our proposed algorithm on a range of datasets.

The liver ultrasound standard planes (LUSP) have significant diagnostic significance during ultrasonic liver diagnosis. However, the location and acquisition of LUSP could be a time-consuming and complicated mission and requires the relevant operator to have comprehensive knowledge of ultrasound diagnosis. Therefore, this study puts forward an automatic classification approach for eight types of LUSP based on a pre-trained CNN(Convolutional Neural Network). With the comparison to classification methods on the basis of conventional hand-craft characteristics, the method proposed by us can automatically catch the appearance in LUSP and classify the LUSP. The proposed model is consisted of 13 convolutional layers with little 3×3 size kernels and three completely connected layers. To address the limitation of data, we adopt the transfer learning strategy, which pre-trains the weight of convolutional layers and fine-tune the weight of fully connected layers. These extensive experiments show that the accuracy of the suggested method reaches 92.31%, as well as the performance of the suggested means outperforms previous ways, which demonstrates the suitability and effectiveness of CNN to classify LUSP for clinical diagnosis.

Tremendous differences between human sexes are universally observed. Therefore, identifying and analyzing the sex-biased genes are becoming basically important for uncovering the mystery of sex differences and personalized medicine. Here, we presented a computational method to identify sex-biased genes from public gene expression databases. We obtained 1407 female-biased genes (FGs) and 1096 male-biased genes (MGs) across 14 different tissues. Bioinformatics analysis revealed that compared with MGs, FGs have higher evolutionary rate, higher single-nucleotide polymorphism density, less homologous gene numbers and smaller phyletic age. FGs have lower expression level, higher tissue specificity and later expressed stage in body development. Moreover, FGs are highly involved in immune-related functions, whereas MGs are more enriched in metabolic process. In addition, cellular network analysis revealed that MGs have higher degree, more cellular activating signaling and tend to be located in cellular inner space, whereas FGs have lower degree, more cellular repressing signaling and tend to be located in cellular outer space. Finally, the identified sex-biased genes and the discovered biological insights together can be a valuable resource helpful for investigating sex-biased physiology and medicine, for example sex-biased disease diagnosis and therapy, which represents one important aspect of personalized and precision medicine.

In short-term load forecasting, dataset construction plays a vital role in the improvement of forecasting accuracy. This paper begins with filtering the load data of the target city, transferring and expanding the load data of the nearby cities, and analysing the influence of the periodicity of load, holiday and load growth rate on dataset construction. A dataset construction method combining compress-filtering and transfer-expanding is then proposed based on the analysis. In this method, firstly, we apply the mean compress method to compress monthly data into weekly data in which both the periodic trend and the randomness of other related factors are taken into consideration. The training set is selected according to the similarity between the load variation pattern of the predicted month and that of the historical data. Secondly, based on the analysis of the similarity of the load data between the target city and the nearby cities, we introduce the load growth rate to measure the difference of the load variation patterns between different cities. Based on the load growth rate, a transfer learning method is put forward which transfers the data of the source city to the target city. The case study on real load data shows that, compared with the mutual information filtering-based predicting method and the knowledge transfer expanding method, the mean absolute percent error is decreased by 26% and 9.5%, respectively.

The microbiota living in the human body has critical impacts on our health and disease, but a systems understanding of its relationships with disease remains limited. Here, we use a large-scale text mining-based manually curated microbe-disease association data set to construct a microbe-based human disease network and investigate the relationships between microbes and disease genes, symptoms, chemical fragments and drugs. We reveal that microbe-based disease loops are significantly coherent. Microbe-based disease connections have strong overlaps with those constructed by disease genes, symptoms, chemical fragments and drugs. Moreover, we confirm that the microbe-based disease analysis is able to predict novel connections and mechanisms for disease, microbes, genes and drugs. The presented network, methods and findings can be a resource helpful for addressing some issues in medicine, for example, the discovery of bench knowledge and bedside clinical solutions for disease mechanism understanding, diagnosis and therapy.

Long noncoding RNAs (lncRNAs) represent a big category of noncoding RNA molecules, and increasing studies have shown that they play important roles in various critical biological processes. They show a diversity of functions through diverse mechanisms, among which regulating RNA molecules is one of the most popular ones. Given the big number of lncRNAs, it becomes urgent and important to predict the RNA targets of lncRNAs in a large scale for the comprehensive understanding of lncRNA functions and action mechanisms. Although several methods have been developed to predict RNA-RNA interactions, none of them can be used to predict the RNA targets of lncRNAs in a large scale. Here we presented a tool, LncTar, which shows the ability to efficiently predict the RNA targets of lncRNAs in a large scale. To test the accuracy of LncTar, we applied it to 10 experimentally supported lncRNA-mRNA interactions. As a result, LncTar successfully predicted 8 (80%) of the 10 lncRNA-mRNA pairs, suggesting that LncTar has a reliable accuracy. Finally, we believe that LncTar could be an efficient tool for the fast identification of the RNA targets of lncRNAs. LncTar is freely available at http://www.cuilab.cn/lnctar.

We propose RoLo (Rotated Logging), a new logging architecture for RAID10 systems for enhanced energy efficiency, performance and reliability. By spreading destaging I/O activities among short idle time slots and proactively reclaiming the stale logging space, RoLo rotates loggers among a logical logging space pool formed collectively from the free storage space available among mirrored disks. Therefore, without the extra dedicated log disks and the corresponding centralized logging, RoLo eliminates the additional hardware and energy costs, potential single point of failure and performance bottleneck. Furthermore, RoLo prolongs the lifecycle of the disks and improves the system's energy efficiency by reducing the disk spin up/down frequency. We develop three flavors of RoLo, that is, RoLo-E/R/P, to emphasize energy efficiency, reliability, and performance respectively. Extensive trace-driven evaluations demonstrate the advantages of the three RoLo schemes over both a RAID10 system with centralized logging architecture and a typical RAID10 system.