Search

The health of the trees in the forest affects the ecological environment, so timely detection of Standing Dead Trees (SDTs) plays an important role in forest management. However, due to the large spatial scope of forests, it is difficult to find SDTs through conventional approaches such as field inventories. In recent years, the development of deep learning and Unmanned Aerial Vehicle (UAV) has provided technical support for low-cost real-time monitoring of SDTs, but the inability to fully utilize global features and the difficulty of small-scale SDTs detection have brought challenges to the detection of SDTs in visible light images. Therefore, this paper proposes a multi-scale attention mechanism detection method for identifying SDTs in UAV RGB images. This method takes Faster-RCNN as the basic framework and uses Swin-Transformer as the backbone network for feature extraction, which can effectively obtain global information. Then, features of different scales are extracted through the feature pyramid structure and feature balance enhancement module. Finally, dynamic training is used to improve the quality of the model. The experimental results show that the algorithm proposed in this paper can effectively identify the SDTs in the visible light image of the UAV with an accuracy of 95.9%. This method of SDTs identification can not only improve the efficiency of SDTs exploration, but also help relevant departments to explore other forest species in the future. [ABSTRACT FROM AUTHOR]

Student Dropout Prediction (SDP) is pivotal in mitigating withdrawals in Massive Open Online Courses. Previous studies generally modeled the SDP problem as a binary classification task, providing a single prediction outcome. Accordingly, some attempts introduce survival analysis methods to achieve continuous and consistent predictions over time. However, the volatility and sparsity of data always weaken the models' performance. Prevailing solutions rely heavily on data pre-processing independent of predictive models, which are labor-intensive and may contaminate authentic data. This paper proposes a Survival Analysis based Volatility and Sparsity Modeling Network (SAVSNet) to address these issues in an end-to-end deep learning framework. Specifically, SAVSNet smooths the volatile time series by convolution network while preserving the original data information using Long-Short Term Memory Network (LSTM). Furthermore, we propose a Time-Missing-Aware LSTM unit to mitigate the impact of data sparsity by integrating informative missingness patterns into the model. A survival analysis loss function is adopted for parameter estimation, and the model outputs monotonically decreasing survival probabilities. In the experiments, we compare the proposed method with state-of-the-art methods in two real-world MOOC datasets, and the experiment results show the effectiveness of our proposed model. [ABSTRACT FROM AUTHOR]

Polyamines are critical metabolites involved in various cellular processes and often dysregulated in cancers. Kaposi's sarcoma-associated Herpesvirus (KSHV), a defined human oncogenic virus, leads to profound alterations of host metabolic landscape to favor development of KSHV-associated malignancies. In our studies, we identified that polyamine biosynthesis and eIF5A hypusination are dynamically regulated by KSHV infection through modulation of key enzymes (ODC1 and DHPS) of these pathways. During KSHV latency, ODC1 and DHPS are upregulated along with increase of hypusinated eIF5A (hyp-eIF5A), while hyp-eIF5A is further induced along with reduction of ODC1 and intracellular polyamines during KSHV lytic reactivation. In return these metabolic pathways are required for both KSHV lytic reactivation and de novo infection. Further analysis unraveled that synthesis of critical KSHV latent and lytic proteins (LANA, RTA) depends on hypusinated-eIF5A. We also demonstrated that KSHV infection can be efficiently and specifically suppressed by inhibitors targeting these pathways. Collectively, our results illustrated that the dynamic and profound interaction of a DNA tumor virus (KSHV) with host polyamine biosynthesis and eIF5A hypusination pathways promote viral propagation, thus defining new therapeutic targets to treat KSHV-associated malignancies. Author summary: Understanding virus-host interactions is crucial to develop and improve therapies. Kaposi's sarcoma associated Herpesvirus (KSHV) is a human gamma-herpesvirus which deeply modulates the host metabolism and is associated with various cancers of endothelial and lymphoid origin. Polyamines are critical metabolites often dysregulated in cancers. In this study we demonstrated KSHV dynamically modulates polyamine metabolism to favor eIF5A hypusination and translation of critical KSHV latent and lytic proteins (LANA, RTA). Consequently, we found KSHV lytic switch from latency and de novo infection were dependent on polyamines and hypusination and pharmacological inhibition efficiently and specifically restricted KSHV infection. Our study provides new insights into KSHV alteration of the host metabolism and describe new therapeutic targets to treat KSHV-associated malignancies. [ABSTRACT FROM AUTHOR]

Both Kaposi's sarcoma-associated herpesvirus (KSHV) and Epstein-Barr virus (EBV) establish the persistent, life-long infection primarily at the latent status, and associate with certain types of tumors, such as B cell lymphomas, especially in immuno-compromised individuals including people living with HIV (PLWH). Lytic reactivation of these viruses can be employed to kill tumor cells harboring latently infected viral episomes through the viral cytopathic effects and the subsequent antiviral immune responses. In this study, we identified that polo-like kinase 1 (PLK1) is induced by KSHV de novo infection as well as lytic switch from KSHV latency. We further demonstrated that PLK1 depletion or inhibition facilitates KSHV reactivation and promotes cell death of KSHV-infected lymphoma cells. Mechanistically, PLK1 regulates Myc that is critical to both maintenance of KSHV latency and support of cell survival, and preferentially affects the level of H3K27me3 inactive mark both globally and at certain loci of KSHV viral episomes. Furthremore, we recognized that PLK1 inhibition synergizes with STAT3 inhibition to efficiently induce KSHV reactivation. We also confirmed that PLK1 depletion or inhibition yields the similar effect on EBV lytic reactivation and cell death of EBV-infected lymphoma cells. Lastly, we noticed that PLK1 in B cells is elevated in the context of HIV infection and caused by HIV Nef protein to favor KSHV/EBV latency. Author summary: KSHV is a large double-stranded DNA virus belonging to human gamma-herpesviruses and capable of establishing the long-term persistent infection, which causes cancers under immunocompromised states, such as acquired immunodeficiency syndrome (AIDS). There are currently still no effective therapies to specifically treat KSHV-associated tumors, nor eliminate its persistent infection. In this study, we illustrated that PLK1 is upregulated by KSHV, which in return suppresses KSHV lytic infection. We further demonstrated that PLK1 is a novel host target that can be inhibited pharmacologically to benefit the viral oncolysis for promotion of KSHV lytic reactivation and elimination of KSHV-positive lymphoma cells, particularly for people living with HIV (PLWH). PLK1 inhibitors also exerted the similar effect on EBV that is the other member of human gamma-herpesvirus. [ABSTRACT FROM AUTHOR]