DeTAL: Open-Vocabulary Temporal Action Localization with Decoupled Networks.

Pre-trained visual-language (ViL) models have demonstrated good zero-shot capability in video understanding tasks, where they were usually adapted through fine-tuning or temporal modeling. However, in the task of open-vocabulary temporal action localization (OV-TAL), such adaption reduces the robustness of ViL models against different data distributions, leading to a misalignment between visual representations and text descriptions of unseen action categories. As a result, existing methods often strike a trade-off between action detection and classification. Aiming at this issue, this paper proposes DeTAL, a simple but effective two-stage approach for OV-TAL. DeTAL decouples action detection from action classification to avoid the compromise between them, and the state-of-the-art methods for close-set action localization can be handily adapted to OV-TAL, which significantly improves the performance. Meanwhile, DeTAL can easily tackle the scenario where action category annotations are unavailable in the training dataset. In the experiments, we propose a new cross-dataset setting to evaluate the zero-shot capability of different methods. And the results demonstrate that DeTAL outperforms the state-of-the-art methods for OV-TAL on both THUMOS14 and ActivityNet1.3. Code and data are publicly available at https://github.com/vsislab/DeTAL.

IGCN: A Provably Informative GCN Embedding for Semi-Supervised Learning with Extremely Limited Labels.

Graph Neural Networks (GNNs) have gained much more attention in the representation learning for the graph-structured data. However, the labels are always limited in the graph, which easily leads to the overfitting problem and causes the poor performance. To solve this problem, we propose a new framework called IGCN, short for Informative Graph Convolutional Network, where the objective of IGCN is designed to obtain the informative embeddings via discarding the task-irrelevant information of the graph data based on the mutual information. As the mutual information for irregular data is intractable to compute, our framework is optimized via a surrogate objective, where two terms are derived to approximate the original objective. For the former term, it demonstrates that the mutual information between the learned embeddings and the ground truth should be high, where we utilize the semi-supervised classification loss and the prototype based supervised contrastive learning loss for optimizing it. For the latter term, it requires that the mutual information between the learned node embeddings and the initial embeddings should be high and we propose to minimize the reconstruction loss between them to achieve the goal of maximizing the latter term from the feature level and the layer level, which contains the graph encoder-decoder module and a novel architecture GCN Info. Moreover, we provably show that the designed GCN Info can better alleviate the information loss and preserve as much useful information of the initial embeddings as possible. Experimental results show that the IGCN outperforms the state-of-the-art methods on 7 popular datasets.

Structural basis for Retriever-SNX17 assembly and endosomal sorting.

During endosomal recycling, Sorting Nexin 17 (SNX17) facilitates the transport of numerous membrane cargo proteins by tethering them to the Retriever complex. Despite its importance, the mechanisms underlying this interaction have remained elusive. Here, we report the structure of the Retriever-SNX17 complex determined using cryogenic electron microscopy (cryo-EM). Our structure reveals that the C-terminal tail of SNX17 engages with a highly conserved interface between the VPS35L and VPS26C subunits of Retriever. Through comprehensive biochemical, cellular, and proteomic analyses, we demonstrate that disrupting this interface impairs the Retriever-SNX17 interaction, subsequently affecting the recycling of SNX17-dependent cargos and altering the composition of the plasma membrane proteome. Intriguingly, we find that the SNX17-binding pocket on Retriever can be utilized by other ligands that share a consensus acidic C-terminal tail motif. By showing how SNX17 is linked to Retriever, our findings uncover a fundamental mechanism underlying endosomal trafficking of critical cargo proteins and reveal a mechanism by which Retriever can engage with other regulatory factors.

A machine learning based method for constructing group profiles of university students.

This study facilitates university student profiling by constructing a prediction model to forecast the classification of future students participating in a survey, thereby enhancing the utility and effectiveness of the questionnaire approach. In the context of the ongoing digital transformation of campuses, higher education institutions are increasingly prioritizing student educational development. This shift aligns with the maturation of big data technology, prompting scholars to focus on profiling university student education. While earlier research in this area, particularly foreign studies, focus on extracting data from specific learning contexts and often relied on single data sources, our study addresses these limitations. We employ a comprehensive approach, incorporating questionnaire surveys to capture a diverse array of student data. Considering various university student attributes, we create a holistic profile of the student population. Furthermore, we use clustering techniques to develop a categorical prediction model. In our clustering analysis, we employ the K-means algorithm to group student survey data. The results reveal four distinct student profiles: Diligent Learners, Earnest Individuals, Discerning Achievers, and Moral Advocates. These profiles are subsequently used to label student groups. For the classification task, we leverage these labels to establish a prediction model based on the Back Propagation neural network, with the goal of assigning students to their respective groups. Through meticulous model optimization, an impressive classification accuracy of 90.22% is achieved. Our research offers a novel perspective and serves as a valuable methodological reference for university student profiling.

Switchable Enantioselectivity in Conjugate Alkyne Addition of ß,γ-Unsaturated α-Keto Esters by Asymmetric Binary Acid Catalysis.

Switchable enantioselectivity was uncovered in the enantioselective catalytic conjugate addition of ß,γ-unsaturated α-keto esters with terminal alkynes to the chiral Lewis acid complex of In(BF4)3 and chiral phosphoric acid.

A green aqueous binder to enhance the electrochemical performance of Li-rich disordered rock salt cathode material.

Li-rich disordered rock-salt oxides (DRX) are considered an attractive cathode material in the future battery field due to their excellent energy density and specific capacity. Nevertheless, anionic redox provides high capacity while causing O2 over-oxidation to O2, resulting in voltage hysteresis and capacity decay. Herein, the crystal structure of Li1.3Mn0.4Ti0.3O1.7F0.3 (LMTOF) cathode is stabilized by using sodium carboxymethylcellulose (CMC) binders replacing traditional polyvinylidene difluoride (PVDF) binders. The electrochemical impedance spectroscopy (EIS) and galvanostatic intermittent titration technique (GITT) reveal that the CMC-based LMTOF electrode has higher electronic conductivity and lithium-ion diffusion kinetics. Moreover, CMC has been demonstrated to improve the O2- reversibility, reduce the amounts of byproducts from electrolyte decomposition and suppress transition metal dissolution by Na+/Li+ exchange reaction. Furthermore, the CMC-based LMTOF electrode also exhibits less volume change upon lithiation/delithiation processes compared to the PVDF-based electrode, resulting in enhanced structural stability during cycling. Benefiting from these features, the CMC binders can effectively improve the cycling life and rate performance of the LMTOF cathode, and the CMC-based LMTOF electrode shows good capacity retention of 94.5 % after 30 cycles at 20 mA/g and 66.7 % after 100 cycles at 200 mA/g. This finding indicates that CMC as a binder can efficiently stabilize the structure and improve the electrochemical performance of Li-rich disordered rock-salt oxides cathode, making it possible for practical Li-ion battery applications.

PolypMixNet: Enhancing semi-supervised polyp segmentation with polyp-aware augmentation.

BACKGROUND: AI-assisted polyp segmentation in colonoscopy plays a crucial role in enabling prompt diagnosis and treatment of colorectal cancer. However, the lack of sufficient annotated data poses a significant challenge for supervised learning approaches. Existing semi-supervised learning methods also suffer from performance degradation, mainly due to task-specific characteristics, such as class imbalance in polyp segmentation. PURPOSE: The purpose of this work is to develop an effective semi-supervised learning framework for accurate polyp segmentation in colonoscopy, addressing limited annotated data and class imbalance challenges. METHODS: We proposed PolypMixNet, a semi-supervised framework, for colorectal polyp segmentation, utilizing novel augmentation techniques and a Mean Teacher architecture to improve model performance. PolypMixNet introduces the polyp-aware mixup (PolypMix) algorithm and incorporates dual-level consistency regularization. PolypMix addresses the class imbalance in colonoscopy datasets and enhances the diversity of training data. By performing a polyp-aware mixup on unlabeled samples, it generates mixed images with polyp context along with their artificial labels. A polyp-directed soft pseudo-labeling (PDSPL) mechanism was proposed to generate high-quality pseudo labels and eliminate the dilution of lesion features caused by mixup operations. To ensure consistency in the training phase, we introduce the PolypMix prediction consistency (PMPC) loss and PolypMix attention consistency (PMAC) loss, enforcing consistency at both image and feature levels. Code is available at https://github.com/YChienHung/PolypMix. RESULTS: PolypMixNet was evaluated on four public colonoscopy datasets, achieving 88.97% Dice and 88.85% mIoU on the benchmark dataset of Kvasir-SEG. In scenarios where the labeled training data is limited to 15%, PolypMixNet outperforms the state-of-the-art semi-supervised approaches with a 2.88-point improvement in Dice. It also shows the ability to reach performance comparable to the fully supervised counterpart. Additionally, we conducted extensive ablation studies to validate the effectiveness of each module and highlight the superiority of our proposed approach. CONCLUSION: PolypMixNet effectively addresses the challenges posed by limited annotated data and unbalanced class distributions in polyp segmentation. By leveraging unlabeled data and incorporating novel augmentation and consistency regularization techniques, our method achieves state-of-the-art performance. We believe that the insights and contributions presented in this work will pave the way for further advancements in semi-supervised polyp segmentation and inspire future research in the medical imaging domain.

Clinical Features of Li-Fraumeni Syndrome in Korea.

PURPOSE: Li-Fraumeni syndrome (LFS) is a hereditary disorder caused by germline mutation in TP53. Owing to the rarity of LFS, data on its clinical features are limited. This study aimed to evaluate the clinical characteristics and prognosis of Korean patients with LFS. MATERIALS AND METHODS: Patients who underwent genetic counseling and confirmed with germline TP53 mutation in the National Cancer Center in Korea between 2011 and 2022 were retrospectively reviewed. Data on family history with pedigree, types of mutation, clinical features, and prognosis were collected. RESULTS: Fourteen patients with LFS were included in this study. The median age at diagnosis of the first tumor was 32 years. Missense and nonsense mutations were observed in 13 and one patients, respectively. The repeated mutations were p.Arg273His, p.Ala138Val, and pPro190Leu. The sister with breast cancer harbored the same mutation of p.Ala138Val. Seven patients had multiple primary cancers. Breast cancer was most frequently observed, and other types of tumor included sarcoma, thyroid cancer, pancreatic cancer, brain tumor, adrenocortical carcinoma, ovarian cancer, endometrial cancer, colon cancer, vaginal cancer, skin cancer, and leukemia. The median follow-up period was 51.5 months. Two and four patients showed local recurrence and distant metastasis, respectively. Two patients died of leukemia and pancreatic cancer 3 and 23 months after diagnosis, respectively. CONCLUSION: This study provides information on different characteristics of patients with LFS, including types of mutation, types of cancer, and prognostic outcomes. For more appropriate management of these patients, proper genetic screening and multidisciplinary discussion are required.

Cu(OTf)2-Phosphoric Acid-Catalyzed Tandem Oxa-Nazarov Cyclization and Dibromination of Conjugated 1,2-Diketones.

A facile tandem oxa-Nazarov cyclization and dibromination has been developed. The combination of Cu(OTf)2 and diphenyl phosphate (DPP-H) was found to synergistically promote the coupling of conjugated 1,2-diketones and N-bromosuccinimide to form 2,4-dibromo-3(2H)-furanones in good yields.

Pro-carcinogenic actions of miR-155/FOXO3a in colorectal cancer development.

Colorectal cancer (CRC) ranks third in cancer incidence and second in cancer mortality globally. MicroRNAs (miRNAs) are promising biomarkers and therapeutic targets for CRC diagnosis and treatment. The miR-155 is reported to induce radiation resistance in CRC. In this study, we aimed to further clarify the role and underlying mechanism of the miR-155 in CRC cell malignancy. We found that miR-155 was significantly up-regulated in CRC tissues. The results of loss-of-function experiments revealed that miR-155 deficiency suppressed the proliferative capacity, invasion, and migration of CRC cells. Moreover, the downstream target genes of miR-155 were screened, and miR-155 was demonstrated to directly bind to FOXO3a in CRC cells to negatively regulate FOXO3a expression. FOXO3a was downregulated in CRC tissues and the expression of FOXO3a and miR-155 was in negative correlation in CRC tissues. FOXO3a overexpression alone was revealed to inhibit CRC cell growth, migration and invasion. Additionally, rescue assays showed that FOXO3a silencing significantly reversed the inhibitory effect of miR-155 deficiency on CRC cell malignant behaviors. In conclusion, miR-155 induces malignant phenotypes of CRC cells including cell proliferation, migration and invasion by targeting FOXO3a, which might provide clues for the targeted therapy of CRC.

Factors for achieving target serum uric acid levels after initiating urate-lowering therapy in patients with gout: results from the ULTRA registry.

Achieving target serum uric acid (SUA) levels is important in gout management. Guidelines recommend lowering SUA levels to < 6 mg/dL; however, many patients fail to reach this target, even with uric acid-lowering therapy (ULT). This study investigated clinical characteristics of target SUA achievers among Korean patients with gout. This study used data from the ULTRA registry, a nationwide inception cohort established in September 2021 that enrolls patients with gout who initiate ULT. Demographic, clinical, and laboratory data were collected at baseline; the 6-month follow-up. Patients were divided into two groups: target achievers (SUA level < 6 mg/dL at 6 months) and non-achievers. The mean participant (N = 117) age was 56.1 years, and 88.0% were male. At 6 months, 83 patients (70.9%) reached target SUA levels. Target achievers had better drug adherence (≥ 80%) to ULT (97.6% vs. 76.5%; p < 0.01) than non-achievers. Target non-achievers had a higher percentage of a family history of gout (32.4% vs. 10.8%; p < 0.01) and less antihypertensive agent use (38.2% vs. 59.0%; p = 0.03) than target achievers. Multivariate regression analysis revealed that good adherence to ULT, the absence of a family history of gout, and antihypertensive agent use were key factors associated with achieving target SUA levels at 6 months.

Living with particles: Disclosure of pollution information, individual responses, and health consequences.

Drawing on a panel dataset-the China Health and Retirement Longitudinal Survey (CHARLS)-and other city- and individual-level datasets, this study examines the causal impact of pollution information disclosure on individual outdoor activities and the health status of the middle-aged and elderly. Using city-level variations in disclosure timing, we found that the adoption of pollution information disclosure (PID) significantly reduces the probability of outdoor exercise, especially for those living in more polluted cities. This occurs mainly through enhanced awareness of environmental pollution, particularly for those who are more educated. However, the adoption of PID does not lead to an improvement in health status.

Silencing GMPPB Inhibits the Proliferation and Invasion of GBM via Hippo/MMP3 Pathways.

Glioblastoma multiforme (GBM) is a highly aggressive malignancy and represents the most common brain tumor in adults. To better understand its biology for new and effective therapies, we examined the role of GDP-mannose pyrophosphorylase B (GMPPB), a key unit of the GDP-mannose pyrophosphorylase (GDP-MP) that catalyzes the formation of GDP-mannose. Impaired GMPPB function will reduce the amount of GDP-mannose available for O-mannosylation. Abnormal O-mannosylation of alpha dystroglycan (α-DG) has been reported to be involved in cancer metastasis and arenavirus entry. Here, we found that GMPPB is highly expressed in a panel of GBM cell lines and clinical samples and that expression of GMPPB is positively correlated with the WHO grade of gliomas. Additionally, expression of GMPPB was negatively correlated with the prognosis of GBM patients. We demonstrate that silencing GMPPB inhibits the proliferation, migration, and invasion of GBM cells both in vitro and in vivo and that overexpression of GMPPB exhibits the opposite effects. Consequently, targeting GMPPB in GBM cells results in impaired GBM tumor growth and invasion. Finally, we identify that the Hippo/MMP3 axis is essential for GMPPB-promoted GBM aggressiveness. These findings indicate that GMPPB represents a potential novel target for GBM treatment.

Genome-Wide Analysis of MBF1 Family Genes in Five Solanaceous Plants and Functional Analysis of SlER24 in Salt Stress.

Multiprotein bridging factor 1 (MBF1) is an ancient family of transcription coactivators that play a crucial role in the response of plants to abiotic stress. In this study, we analyzed the genomic data of five Solanaceae plants and identified a total of 21 MBF1 genes. The expansion of MBF1a and MBF1b subfamilies was attributed to whole-genome duplication (WGD), and the expansion of the MBF1c subfamily occurred through transposed duplication (TRD). Collinearity analysis within Solanaceae species revealed collinearity between members of the MBF1a and MBF1b subfamilies, whereas the MBF1c subfamily showed relative independence. The gene expression of SlER24 was induced by sodium chloride (NaCl), polyethylene glycol (PEG), ABA (abscisic acid), and ethrel treatments, with the highest expression observed under NaCl treatment. The overexpression of SlER24 significantly enhanced the salt tolerance of tomato, and the functional deficiency of SlER24 decreased the tolerance of tomato to salt stress. SlER24 enhanced antioxidant enzyme activity to reduce the accumulation of reactive oxygen species (ROS) and alleviated plasma membrane damage under salt stress. SlER24 upregulated the expression levels of salt stress-related genes to enhance salt tolerance in tomato. In conclusion, this study provides basic information for the study of the MBF1 family of Solanaceae under abiotic stress, as well as a reference for the study of other plants.

Clinicopathological Factors Associated with Oncotype DX Risk Group in Patients with ER+/HER2- Breast Cancer.

Oncotype DX (ODX), a 21-gene assay, predicts the recurrence risk in early breast cancer; however, it has high costs and long testing times. We aimed to identify clinicopathological factors that can predict the ODX risk group and serve as alternatives to the ODX test. This retrospective study included 547 estrogen receptor-positive, human epidermal growth factor receptor 2-negative, and lymph node-negative breast cancer patients who underwent ODX testing. Based on the recurrence scores, three ODX risk categories (low: 0-15, intermediate: 16-25, and high: 26-100) were established in patients aged ≤50 years (n = 379), whereas two ODX risk categories (low: 0-25 and high: 26-100) were established in patients aged >50 years (n = 168). Factors selected for analysis included body mass index, menopausal status, type of surgery, and pathological and immunohistochemical features. The ODX risk groups showed significant association with histologic grade (p = 0.0002), progesterone receptor expression (p < 0.0001), Ki-67 (p < 0.0001), and p53 expression (p = 0.023) in patients aged ≤50 years. In patients aged >50 years, tumor size (p = 0.022), Ki-67 (p = 0.001), and p53 expression (p = 0.001) were significantly associated with the risk group. Certain clinicopathological factors can predict the ODX risk group and enable decision-making on adjuvant chemotherapy; these factors differ according to age.

The dual lipid desaturase/hydroxylase DEGS2 controls phytoceramide levels necessary to counter intestinal inflammation.

Intestinal immunity is dependent on barrier function to maintain quiescence. The mechanisms for the maintenance of this barrier are not fully understood. Delta 4-desaturase, sphingolipid 2 (DEGS2) is a lipid desaturase and hydroxylase that catalyzes the synthesis of ceramide and phytoceramide from dihydroceramide. Using a forward genetic approach, we found and validated a mutation in Degs2 as causative of increasing susceptibility to colitis and altering the phytoceramide balance in the colon. DEGS2 is expressed in the intestinal epithelium, and the colitis phenotype is dependent on the non-hematopoietic compartment of the mouse. In the absence of DEGS2, the colon lacks phytoceramides and accumulates large amounts of the precursor lipid dihydroceramide. In response to dextran sodium sulfate (DSS)-induced colitis, colonic epithelial cells in DEGS2-deficient mice had increased cell death and decreased proliferation compared to those in wild-type mice. These findings demonstrate that DEGS2 is needed to maintain epithelial integrity, protect against DSS-induced colitis and maintain lipid balance in vivo.

3D Visual Saliency: An Independent Perceptual Measure or a Derivative of 2D Image Saliency?

While 3D visual saliency aims to predict regional importance of 3D surfaces in agreement with human visual perception and has been well researched in computer vision and graphics, latest work with eye-tracking experiments shows that state-of-the-art 3D visual saliency methods remain poor at predicting human fixations. Cues emerging prominently from these experiments suggest that 3D visual saliency might associate with 2D image saliency. This paper proposes a framework that combines a Generative Adversarial Network and a Conditional Random Field for learning visual saliency of both a single 3D object and a scene composed of multiple 3D objects with image saliency ground truth to 1) investigate whether 3D visual saliency is an independent perceptual measure or just a derivative of image saliency and 2) provide a weakly supervised method for more accurately predicting 3D visual saliency. Through extensive experiments, we not only demonstrate that our method significantly outperforms the state-of-the-art approaches, but also manage to answer the interesting and worthy question proposed within the title of this paper.

GS-16QAM OFDM with ANN scheme combined with LFM signal for joint communication and radar sensing system.

In this Letter, an integrated waveform geometrically shaped (GS) 16 quadrature amplitude modulation (QAM) based orthogonal frequency division multiplexing (OFDM) communication signal combined with a linear frequency modulation (LFM) radar signal is proposed and experimentally demonstrated in a W-band communication and radar detection system. The proposed method can generate communication and radar signals simultaneously. The inherent error propagation and the interference of the radar signal limit the transmission performance of the joint communication and radar sensing system. Thus, an artificial neural network (ANN) scheme is proposed for the GS-16QAM OFDM signal. After 0.8-m wireless transmission, the experimental results show that the receiver sensitivity and normalized general mutual information (NGMI) of the GS-16QAM OFDM are improved compared with that of the OFDM with uniform 16QAM at the forward error correction (FEC) threshold of 3.8 × 10-3. In addition, multi-target radar detection in centimeter level radar ranging is realized.

Trans-Golgi protein TVP23B regulates host-microbe interactions via Paneth cell homeostasis and Goblet cell glycosylation.

A key feature in intestinal immunity is the dynamic intestinal barrier, which separates the host from resident and pathogenic microbiota through a mucus gel impregnated with antimicrobial peptides. Using a forward genetic screen, we have found a mutation in Tvp23b, which conferred susceptibility to chemically induced and infectious colitis. Trans-Golgi apparatus membrane protein TVP23 homolog B (TVP23B) is a transmembrane protein conserved from yeast to humans. We found that TVP23B controls the homeostasis of Paneth cells and function of goblet cells, leading to a decrease in antimicrobial peptides and more penetrable mucus layer. TVP23B binds with another Golgi protein, YIPF6, which is similarly critical for intestinal homeostasis. The Golgi proteomes of YIPF6 and TVP23B-deficient colonocytes have a common deficiency of several critical glycosylation enzymes. TVP23B is necessary for the formation of the sterile mucin layer of the intestine and its absence disturbs the balance of host and microbe in vivo.