MOTIVATION: Single-cell multi-omics technologies enable the simultaneous profiling of gene expression and chromatin accessibility, providing complementary insights into cellular identity and gene regulatory mechanisms. However, integrating paired scRNA-seq and scATAC-seq data (i.e. profiles from the same single cell) remains challenging due to inherent sparsity, technical noise, and the limited availability of high-quality paired measurements. In contrast, large-scale unpaired scRNA-seq datasets often exhibit robust and biologically meaningful cell cluster structures. RESULTS: We introduce Guided Co-clustering Transfer (GuidedCoC), a novel framework that transfers structural knowledge from unpaired scRNA-seq source data to improve both cell clustering and feature alignment in paired scRNA-seq/scATAC-seq target data. GuidedCoC jointly co-cluster cells and features across modalities and domains via a unified information-theoretic objective, aligning gene expression modules with regulatory elements while implicitly performing cross-modal dimensionality reduction to reduce noise. Additionally, it automatically aligns cell populations across unpaired and paired datasets without requiring explicit annotations. Extensive experiments on multiple benchmark datasets demonstrate that GuidedCoC achieves superior clustering accuracy and biological interpretability compared to existing methods. These results highlight the promise of structure-guided transfer learning for robust, scalable, and interpretable integration of single-cell multi-omics data. AVAILABILITY AND IMPLEMENTATION: GuidedCoC is available as open-source code at https://github.com/No-AgCl/GuidedCoC.

Defocus deblurring is a classic problem in image restoration tasks. The formation of its defocus blur is related to depth. Recently, the use of dual-pixel sensor designed according to depth-disparity characteristics has brought great improvements to the defocus deblurring task. However, the difficulty of real-time acquisition of dual-pixel images brings difficulties to algorithm deployment. This inspires us to remove defocus blur by single image with depth information. We propose a single-image depth-enhanced defocus deblurring network, which uses a depth map estimated by the monocular depth estimation network to guide the network defocus deblurring. We design a deep information fusion unit, which greatly improves the effect of deblurring. Experiments show that on the single image defocus deblurring task, the experimental results demonstrate the superiority of our method.

Visual Speech Recognition has made significant progress, with the current mainstream approach focusing on extracting lip features and using deep learning to enable the model to understand a speaker’s content from video alone. However, one might question whether visual language recognition is synonymous with lipreading. Whether we can extract additional information beyond lip movements to improve model performance. In this study, we developed the Lip-Face-Surrounding model to comprehensively extract information from video. Our findings reveal that the information beyond the lips— such as eye corner movements, jaw movements, nostril movements, throat movements, and shoulder movements—are captured by the model and serve as discriminative features for visual speech recognition. This information is also applicable to handcrafted datasets. Additionally, indirect information such as the speaker’s body language and interactions with the surrounding also impact model performance. This information provides extra context that enhances performance, while other times, it introduces noise that affects model convergence. Our model achieved promising results on the CN-CELEB and GRID datasets, with a 5% absolute performance improvement over the lip-only approach.

Modern high-throughput sequencing technologies have enabled us to profile multiple molecular modalities from the same single cell, providing unprecedented opportunities to assay cellular heterogeneity from multiple biological layers. However, the datasets generated from these technologies tend to have high level of noise and are highly sparse, bringing challenges to data analysis. In this paper, we develop a novel information-theoretic co-clustering-based multi-view learning (scICML) method for multi-omics single-cell data integration. scICML utilizes co-clusterings to aggregate similar features for each view of data and uncover the common clustering pattern for cells. In addition, scICML automatically matches the clusters of the linked features across different data types for considering the biological dependency structure across different types of genomic features. Our experiments on four real-world datasets demonstrate that scICML improves the overall clustering performance and provides biological insights into the data analysis of peripheral blood mononuclear cells.

Recent research showed that the dual-pixel sensor has made great progress in defocus map estimation and image defocus deblurring. However, extracting real-time dual-pixel views is troublesome and complex in algorithm deployment. Moreover, the deblurred image generated by the defocus deblurring network lacks high-frequency details, which is unsatisfactory in human perception. To overcome this issue, we propose a novel defocus deblurring method that uses the guidance of the defocus map to implement image deblurring. The proposed method consists of a learnable blur kernel to estimate the defocus map, which is an unsupervised method, and a single-image defocus deblurring generative adversarial network (DefocusGAN) for the first time. The proposed network can learn the deblurring of different regions and recover realistic details. We propose a defocus adversarial loss to guide this training process. Competitive experimental results confirm that with a learnable blur kernel, the generated defocus map can achieve results comparable to supervised methods. In the single-image defocus deblurring task, the proposed method achieves state-of-the-art results, especially significant improvements in perceptual quality, where PSNR reaches 25.56 dB and LPIPS reaches 0.111.

Despite supervised deep stereo matching networks have achieved impressive performance given sufficient training data, the poor generalization ability caused by the domain shifts prevents them from being applied to unseen domains. Recent progress has shown that CLIP could be a promising alternative for zero-shot visual representation task under the natural language supervision. In this paper, we present a new framework for domain generalized stereo matching by leveraging the contrastive language-image pre-training (CLIP), which distills text-guided discriminative content information rather than task-irrelevant style information. Extensive experiments show that the model generalization ability can be improved significantly in the unseen domain when transferring from SceneFlow to Middlebury, ETH3D and KITTI.

MOTIVATION: Technological advances have enabled us to profile single-cell multi-omics data from the same cells, providing us with an unprecedented opportunity to understand the cellular phenotype and links to its genotype. The available protocols and multi-omics datasets [including parallel single-cell RNA sequencing (scRNA-seq) and single-cell ATAC sequencing (scATAC-seq) data profiled from the same cell] are growing increasingly. However, such data are highly sparse and tend to have high level of noise, making data analysis challenging. The methods that integrate the multi-omics data can potentially improve the capacity of revealing the cellular heterogeneity. RESULTS: We propose an adaptively weighted multi-view learning (scAWMV) method for the integrative analysis of parallel scRNA-seq and scATAC-seq data profiled from the same cell. scAWMV considers both the difference in importance across different modalities in multi-omics data and the biological connection of the features in the scRNA-seq and scATAC-seq data. It generates biologically meaningful low-dimensional representations for the transcriptomic and epigenomic profiles via unsupervised learning. Application to four real datasets demonstrates that our framework scAWMV is an efficient method to dissect cellular heterogeneity for single-cell multi-omics data. AVAILABILITY AND IMPLEMENTATION: The software and datasets are available at https://github.com/pengchengzeng/scAWMV. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

CNNs’ strong inductive biases of locality and weight sharing provide powerful representation ability and data sample utilization efficiency. However, the weight sharing might smooth out the discrepancy between similar pixels, resulting in the wrong matching between left and right camera-image pair in thin structures region and repetitive texture region. In this paper, we propose a novel Heterogeneous Feature Fusion in MLP (HFF-MLP) for Stereo matching. It employs MLP structure and relaxes the weights sharing in the local spatial region. To this end, pixels in thin structures region and repetitive texture region are dealt with independently using the exclusive weights. Based on HFF -MLP module, we design a real-time network, i.e., MLP-Stereo. Experimental results show that our proposed HFF-MLP achieves competitive results on KITTI 2015 test dataset with the running time of 46 milliseconds. Furthermore, it performs much better than other real-time network in thin structures region and repetitive texture region.

The single-cell multiomics technologies provide an unprecedented opportunity to study the cellular heterogeneity from different layers of transcriptional regulation. However, the datasets generated from these technologies tend to have high levels of noise, making data analysis challenging. Here, we propose jointly semi-orthogonal nonnegative matrix factorization (JSNMF), which is a versatile toolkit for the integrative analysis of transcriptomic and epigenomic data profiled from the same cell. JSNMF enables data visualization and clustering of the cells and also facilitates downstream analysis, including the characterization of markers and functional pathway enrichment analysis. The core of JSNMF is an unsupervised method based on JSNMF, where it assumes different latent variables for the two molecular modalities, and integrates the information of transcriptomic and epigenomic data with consensus graph fusion, which better tackles the distinct characteristics and levels of noise across different molecular modalities in single-cell multiomics data. We applied JSNMF to single-cell multiomics datasets from different tissues and different technologies. The results demonstrate the superior performance of JSNMF in clustering and data visualization of the cells. JSNMF also allows joint analysis of multiple single-cell multiomics experiments and single-cell multiomics data with more than two modalities profiled on the same cell. JSNMF also provides rich biological insight on the markers, cell-type-specific region-gene associations and the functions of the identified cell subpopulation.

Abstract It is well known that preserving depth edges is an effective solution for achieving the accurate disparity map in stereo matching, but many state‐of‐the‐art methods do not preserve depth edges well. In order to solve it efficiently, the cell structure containing irregular and regular shape regions is designed to preserve depth edges. Based on the well‐designed cell structure, a novel disparity estimation method for stereo matching is proposed, in which a two‐layer disparity optimization method is proposed to refine the disparity plane; it includes the front‐parallel disparities computation and slanted‐surfaces disparity plane refinement. In the framework of front‐parallel disparities computation, a tree‐based cost aggregation method is presented to make full use of the segmentation information of cells and then performing semi‐global cost aggregation. In the framework of slanted‐surfaces disparity plane refinement, a new probability model is proposed that employs Bayesian inference for refining disparities in textureless, weak texture and occluded regions. Experimental results show that higher accuracy could be achieved via the proposed method compared with some known state‐of‐the‐art stereo methods on KITTI 2015 and Middlebury dataset, which are the standard benchmarks for testing the stereo matching methods. It can also be indicated that the proposed method can produce accurate disparity map and have good generalization performance.

Technological advances have enabled us to profile multiple molecular layers at unprecedented single-cell resolution and the available datasets from multiple samples or domains are growing. These datasets, including scRNA-seq data, scATAC-seq data and sc-methylation data, usually have different powers in identifying the unknown cell types through clustering. So, methods that integrate multiple datasets can potentially lead to a better clustering performance. Here we propose coupleCoC+ for the integrative analysis of single-cell genomic data. coupleCoC+ is a transfer learning method based on the information-theoretic co-clustering framework. In coupleCoC+, we utilize the information in one dataset, the source data, to facilitate the analysis of another dataset, the target data. coupleCoC+ uses the linked features in the two datasets for effective knowledge transfer, and it also uses the information of the features in the target data that are unlinked with the source data. In addition, coupleCoC+ matches similar cell types across the source data and the target data. By applying coupleCoC+ to the integrative clustering of mouse cortex scATAC-seq data and scRNA-seq data, mouse and human scRNA-seq data, mouse cortex sc-methylation and scRNA-seq data, and human blood dendritic cells scRNA-seq data from two batches, we demonstrate that coupleCoC+ improves the overall clustering performance and matches the cell subpopulations across multimodal single-cell genomic datasets. coupleCoC+ has fast convergence and it is computationally efficient. The software is available at https://github.com/cuhklinlab/coupleCoC_plus.

In recent years, there has been an increasing interest in Software Defined Networking (SDN)/OpenFlow, which is a novel network architecture splitting control and data planes. The SDN-enabled products have also been widely deployed in many production networks aiming to bypass the limitations of current Internet architecture. There are several efforts on building Software Defined Routing Platform (SDRP), which integrates SDN/OpenFlow within IP routing. However, to the best of our knowledge, there are a few successful works on realizing SDRP despite of the flexibility and rich features of SDN. In this paper, we investigate state-of-the-art works towards SDRP. We have evaluated the performance of a mainstream SDRP named RouteFlow. Our evaluations have been conducted using both the emulator Mininet and a real testbed with physical switches. Different to other works, we focus on the resilience aspect of the routing platform. The evaluation results show that RouteFlow is resilient against both single and multiple link failures. Additionally, OSPF provided by RouteFlow achieves a shorter failover time than the legacy OSPF. In the case of RIPv2, RouteFlow's protocol has a comparable performance value with the distributed one.

Open Shortest Path First (OSPF) is a broadly used routing protocol in today's data center networks. Though it can adaptively find alternative paths when link failure is detected, route recomputation in OSPF is CPU-intensive, and usually needs quite a long time to achieve routing convergence. Besides, the configuration in OSPF is rather complex. In this paper, we propose Sequoia Routing Protocol (SRP), a routing protocol with a dedicated shortest path calculation algorithm, designed for data center networks. By leveraging the observation that each data center has a fixed Clos topology, SRP largely simplifies routing configuration, and achieves light-weight computation. We show that SRP outperforms OSPF with a much faster convergence time by simulation in ns-3. We also implement SRP in real-world switches, showing that SRP is as efficient as OSPF even for a small-scale network where route recomputation of OSPF does not take a long time.