Phosphorylation ranks among the most crucial post-translational modifications (PTMs), significantly influencing the conformation, activity, and functionality of proteins, and is intricately associated with numerous pathophysiological processes. Therefore, proposing a scientifically valid computational method for precise prediction of phosphorylation sites carries substantial importance. In this study, we propose and evaluate a novel method, PEGN-PSP, based on graph techniques and language models for the prediction of general phosphorylation sites. This method employs an adaptive feature fusion strategy, combining sequence embeddings and pretrained model embeddings to enhance the model's capability in representing features. The use of graph neural network attention mechanisms not only captures local patterns of protein sequences, but also effectively captures long-range dependency information between residues in protein sequences. Independent test results indicate that, in comparison to the current state-of-the-art methods for general phosphorylation site prediction, PEGN-PSP improves the Matthews correlation coefficient for S/T sites by 4.6% and for Y sites by 6.5%. Additionally, PEGN-PSP has good robustness in predicting lysine crotonylation sites, indicating that our method PEGN-PSP has strong potential in predicting other protein post-translational modification sites.

MOTIVATION: Protein function prediction is important for drug development and disease treatment. Recently, deep learning methods have leveraged protein sequence and structural information, achieving remarkable progress in the field of protein function prediction. However, existing methods ignore the complex multimodal interaction information between sequence and structural features. Since protein sequence and structural information reveal the functional characteristics of proteins from different perspectives, it is challenging to effectively fuse the information from these two modalities to portray protein functions more comprehensively. In addition, current methods have difficulty in effectively capturing long-range dependencies and global contextual information in protein sequences during feature extraction, thus limiting the ability of the model to recognize critical functional residues. RESULTS: In this study, we propose a novel framework termed Multi-stage Attention-based Extraction and Fusion model for GO prediction (MAEF-GO) based on a multistage attention mechanism to predict protein functions. MAEF-GO innovatively integrates the graph convolutional network and the graph attention network to extract protein structural features. To address the issue of modeling long-range dependencies within protein sequences, we introduce a frequency-domain attention mechanism capable of extracting global contextual relationships. Additionally, a cross-attention module is implemented to facilitate interactive fusion between protein sequence and structural modalities. Experimental evaluations demonstrate that MAEF-GO achieves superior performance compared to several state-of-the-art baseline models across standard benchmarks. Furthermore, analysis of the cross-attention weight distributions demonstrates MAEF-GO's interpretability. It can effectively identify critical functional residues of proteins. AVAILABILITY AND IMPLEMENTATION: The MAEF-GO source code can be found at https://github.com/nebstudio/MAEF-GO, an archived snapshot of the code used in this study is also available via Zenodo at https://doi.org/10.5281/zenodo.15422392.

Most existing occluded person re-identification methods use a part-based approach to extract pedestrian features. The extracted part features are isolated from each other, resulting in insufficient information exchange between part features. To address this issue, we propose an interleaved dynamic fusion network (IDFNet) for occluded person re-identification. Initially, an interleaved feature pyramid module (IFPM) was designed, which recursively transmits rich semantic information from high-level feature maps to the bottom layer through interleaved connections, achieving the extraction of multi-scale information. Secondly, a multi-scale feature dynamic fusion module (MDFM) to effectively integrate multi-scale information in IFPM and achieve cross-scale feature fusion. It allows the network to dynamically select the most suitable features for fusion based on pedestrian characteristics and size. Finally, the designed feature interaction module (FIM) uses different semantic part features as graph nodes, allowing information transfer between nodes, suppressing the transfer of meaningless feature information such as occlusion, promoting the transfer of semantic feature information, and effectively alleviating occlusion problems. Extensive experimental results on both occluded and holistic datasets demonstrate the efficacy of our approach.

Self-taught learning (STL) is a promising solution that reduces the performance gap between weakly supervised and fully supervised learning for easily accessible, label-free images. The success of traditional STL solutions relies on the assumption that the target appearance is completely visible and well-defined. In real-world facial expression recognition scenarios, however, saliency regions are often partially occluded, which significantly hampers the generalization capability of STL methods. Nevertheless, few studies have investigated the impact of occlusion on STL. In this paper, we propose an interweaved autoencoder network for weakly supervised facial expression recognition in occlusion scenarios. The key innovation of our network lies in the Residual Connection Union (RCU) blocks that can integrate the Convolutional Neural Network (CNN) and Transformer layers into a multi-scale structure. The RCU enables a progressive masking strategy to accurately identify and focus on contributive yet often overlooked image patches by analyzing the relationships among region-level target representations. In addition, we introduce a self-knowledge distillation module for the effective training of the proposed autoencoder network. Extensive experiments are conducted on four public datasets to demonstrate the superiority of our method over related works.

In the field of computer vision, remote sensing image object detection plays an important role. Although the object detection algorithm has made significant progress, there are still problems in detecting objects with multi-scale in remote sensing image. Due to the insufficient utilization of object feature information, the detection accuracy of multi-scale objects is very low. To address the aforementioned issues, this paper proposes an effective object detection algorithm for remote sensing image based on semantic fusion and scale adaptability, known as SFSANet. Firstly, in view of the problem that the existing methods ignore the semantic differences between different scale feature maps, the semantic fusion (SF) module is proposed to enrich the semantic information and improve the ability to classify and locate objects. Next, to address the issue of the objects being easily interfered in complex background and the detection performance is poor, the spatial location attention (SLA) module is constructed to suppress background information and make key objects more prominent. Additionally, the scale adaptability module (SA) is designed to enrich the expression of feature information, realize the integration of global and local information, and ensure the integrity of image structure. Finally, we adopt the SIoU loss function as the localization loss to expedite model convergence. In order to verify the effectiveness of the proposed method, we conduct experiments on the mainstream datasets DIOR and NWPU VHR-10, which fully demonstrate the superiority of the proposed method.

The space-air-ground-aqua integrated network will become the basic form of the next generation network. Various technologies, including artificial intelligence, big data, cloud computing, edge computing, etc., will be deeply integrated into the network to form an integrated intelligent network of land, sea, air and space. In this article, we will present the vision for the development of the space-air-ground-aqua integrated intelligent network and describe its main features. We put forward a network architecture which integrated sub-networks of space, air, land and sea while emphasizing network interconnection, resources sharing, cooperative control and service reuse. We also discussed several promising technologies, including the THz, free space optical communication, software defined network, network function virtualization, edge intelligent, digital twins, physical layer security and blockchains.

Image retrieval technology allows doctors to query diagnosed images which are similar to the current diagnostic image from large medical image library, and then doctors develop accurate treatment programs and reasonable clinical diagnosis. However, medical image data may be leaked and tampered with during the transmission of the Internet, resulting in data security issues. Therefore, this paper proposes a robust algorithm for encrypted medical volume data retrieval based on DWT (Discrete Wavelet Transform) and DFT (Discrete Fourier Transform). Firstly, we encrypt medical volume data. Secondly, we extract the DWT-DFT coefficients as encrypted medical images' feature vector and establish features vector database. Then, we compute the NC (The Normalized Cross-correlation) value between the feature vector of query medical volume data and each one in the features vector database. Next, automatically return the retrieved medical image. Finally, we decrypt returned medical image that we retrieved. Experimental results demonstrate that the algorithm has strong robustness against common attacks and geometric attacks.

More and more medical institutions have a large number of medical images stored in the third platform for saving cost in maintenance and storage. Cloud storage security issues arise from its characteristic of data outsourcing and service leasing. When users upload their data to the cloud computing platform, they lose direct control of these data, so users worry that their personal privacy information will be leaked and misused. In order to protect the user's personal privacy, before uploading medical images to the server, the image needs to be encrypted. But CBIR technology can't apply in cipher-text domain. In this paper, we propose an encrypted medical image retrieval algorithm based on DWT-DCT. Firstly, the medical image is encrypted in the frequency domain and the feature vector is extracted. Then, the encrypted medical image and the eigenvector are uploaded to the encrypted medical image database and the eigenvector database respectively. The normalized correlation coefficient (NC) is used to represent the Similarity of two encryption medical image. From the experimental results, we can get a conclusion that this algorithm has good robustness to against conventional attacks and geometric attacks.