Search

The Internet of Things (IoT) is transforming how we live and work, and its applications are widespread, spanning smart homes, industrial monitoring, smart cities, healthcare, agriculture, and retail. Considering its wide range of applications, addressing the security challenges arising from IoT devices’ massive collection and transmission of user data is vital. Intrusion detection systems (IDS) based on deep learning techniques offer new means and research directions for resolving IoT security issues. Deep learning models can process large volumes of data and extract complex patterns, making them generally more effective than traditional rule based IDSs. While deep learning techniques are gradually gaining popularity in IDS applications, current research needs a comprehensive summary of deep learning-based IDS in IoT. This paper introduces intrusion detection technologies, followed by a detailed comparison, analysis, and discussion of deep learning models, datasets, feature extraction and classifiers, data preprocessing techniques, and experimental design of the models. It also highlights the challenges and issues associated with deep learning models and relevant techniques for IDS. Finally, it concludes by providing recommendations to assist researchers in this domain.

With global warming accelerating, the heavily populated and rapidly urbanized coastal regions of the Pearl River Delta (PRD) and the Yangtze River Delta (YRD) stand as representative areas with mounting concerns about extreme heat stress. This study analyzes differentiated effects of temperature (TAS) and relative humidity (RH) on human heat stress measured by wet-bulb globe temperature (WBGT) in those urban regions based on machine learning and mathematical derivation, while also examining the impacts of global warming and urbanization on prospective heat risks. To generate fine-scale climate projections targeted at the PRD and YRD, two global projections forced by Representative Concentration Pathway (RCP) 8.5 scenario are dynamically downscaled using non-hydrostatic Regional Climate Model version 4.7 (RegCM4), with the urban density and extent updated every year based on Shared Socioeconomic Pathways 5-8.5 (SSP5) scenario, thereby incorporating the transient urban growth into future projections. The bias-corrected downscaled simulations effectively capture the distinct interdependencies between TAS and RH on WBGT across different regions, similar to the observed patterns during the historical period. While the absolute contribution of TAS to WBGT is larger than RH regardless of warming levels and regions, the relative increase in RH becomes more pronounced with warming. Under RCP8.5 scenario, unprecedentedly extreme WBGT is projected to emerge in the far future (2080–2099). In contrast, the effect of urbanization appears to be more dominant in the near future (2030–2049) as urban density under SSP5 scenario is projected to peak around the 2040s and gradually decrease afterwards. The reduction of RH is found in the intensely urbanized areas locally, but it does not significantly lower WBGT because the positive contribution of increased TAS is more dominant. As a result, highly urbanized regions still exhibit higher WBGT compared to other areas. In addition, urban heat island effect is more pronounced for compact areas with high urban density (i.e., PRD) and at night. Despite the smaller temperature increase from urban heat island effect compared to global warming, it can play a critical role in exacerbating heat stress, adding to the already dangerous humid and hot conditions.

Abstract Background DNA damage and DNA damage repair (DDR) are important therapeutic targets for triple-negative breast cancer (TNBC), a subtype with limited chemotherapy efficiency and poor outcome. However, the role of microRNAs in the therapy is emerging. In this study, we explored whether miR-26a-5p could act as BRCAness and enhance chemotherapy sensitivity in TNBC. Methods Quantitative reverse transcription polymerase chain reaction (RT-qPCR) was used to detect the expression of miR-26a-5p in breast cancer tissues and cell lines. CCK-8 was used to measure drug sensitivity in concentration gradient and time gradient. Comet assay was used to detect DNA damage. Flow cytometry was performed to examine apoptosis. Moreover, we used western blot and immunofluorescence to detect biomarkers. Luciferase reporter assay was performed to verify the combination of miR-26a-5p and 3’UTR of target gene. Hormone deprivation and stimulation assay were used to validate the effect of hormone receptors on the expression of miR-26a-5p. Chromatin immunoprecipitation (ChIP) assays were used to verify the binding sites of ER-a or PR with the promoter of miR-26a-5p. Animal experiments were performed to the effect of miR-26a-5p on Cisplatin treatment. Results The expression of miR-26a-5p was significantly downregulated in TNBC. Overexpressing miR-26a-5p enhanced the Cisplatin-induced DNA damage and following apoptosis. Interestingly, miR-26a-5p promoted the expression of Fas without Cisplatin stimulating. It suggested that miR-26a-5p provided a hypersensitivity state of death receptor apoptosis and promoted the Cisplatin sensitivity of TNBC cells in vitro and in vivo. Besides, miR-26a-5p negatively regulated the expression of BARD1 and NABP1 and resulted in homologous recombination repair defect (HRD). Notably, overexpressing miR-26a-5p not only facilitated the Olaparib sensitivity of TNBC cells but also the combination of Cisplatin and Olaparib. Furthermore, hormone receptors functioned as transcription factors in the expression of miR-26a-5p, which explained the reasons that miR-26a-5p expressed lowest in TNBC. Conclusions Taken together, we reveal the important role of miR-26a-5p in Cisplatin sensitivity and highlight its new mechanism in DNA damage and synthetic lethal.

We experimentally demonstrate a low-loss and low-crosstalk 16-channel 200 GHz-channel-spacing arrayed waveguide grating router (AWGR) with a size of 0.67 × 0.37 mm2, targeting for C-band wavelength division multiplexing (WDM) interconnect routing applications. As a result, the AWGR features a channel spacing of 200±10 GHz, a free spectral range of 25.62 nm and a 3 dB channel bandwidth of 0.9 nm. Based on comprehensive optimal design, the measured characterization revealed 5.55 dB maximum channel loss non-uniformity with 1.13 dB best-case channel insertion loss. Besides, the fabricated device exhibits good cyclic-frequency operation for all 16 × 16 port combinations with channel crosstalk of –15.11 dB.

Wepropose and experimentally demonstrate the ultra-compact mode (de)multiplexers and polarization beam splitter (PBS) based on tapered bent asymmetric directional couplers. The strong coupling strength introduced by tapered bent waveguides significantly shortens the coupling length. As a result, high coupling and conversion efficiencies over a broad bandwidth can be obtained with small footprint. The coupling length of the fabricated PBS and mode (de)multiplexers are less than 4.9 μm and 8.7 μm, respectively. From 1534-1600 nm, the measured insertion loss (IL) is less than 1.4 and 1.55 dB for the TE1-TE0 and TE2-TE0 mode (de)multiplexers, respectively, and the crosstalk (CT) is less than −10 dB. The fabricated PBS shows low insertion loss of 0.55 dB and decent crosstalk of < −14 dB for both polarizations over an 85 nm wavelength range from 1505-1590 nm. Meanwhile, the larger fabrication tolerance ensures that the proposed devices can be readily fabricated with the standard complementary metal-oxide-semiconductor (CMOS) process.

A TM0 mode coupled travelling wave SNSPD based on SOI platform is designed. The light propagating in this device can achieve high-efficiency mode conversion from TE0 mode to TM0 mode and then absorbed by NbN nanowires on top of the waveguide. Thanks to the TM0 mode coupled detection, the length of NbN nanowire that is needed to achieve a close-to-unity photon absorption is cut down to an extremely short level of ∼5 μm. Meanwhile, this TM0 mode coupled travelling wave SNSPD is very robust against fabrication errors compared with TE0 coupled SNSPD.

Summary: Inhibition of TRPML1, which is encoded by MCOLN1, is known to deter cell proliferation in various malignancies. Here, we report that the tumor suppressor, p53, represses MCOLN1 in the urothelium such that either the constitutive loss or ectopic knockdown of TP53—in both healthy and bladder cancer cells—increased MCOLN1 expression. Conversely, nutlin-mediated activation of p53 led to the repression of MCOLN1. Elevated MCOLN1 expression in p53-deficient cancer cells, though not sufficient for bolstering proliferation, augmented the effects of oncogenic HRAS on proliferation, cytokine production, and invasion. Our data suggest that owing to derepression of MCOLN1, urothelial cells lacking p53 are poised for tumorigenesis driven by oncogenic HRAS. Given our prior findings that HRAS mutations predict addiction to TRPML1, this study points to the utility of TRPML1 inhibitors for mitigating the growth of a subset of urothelial tumors that lack p53.

Triple negative breast cancer (TNBC) is characterized by lack of expression of the estrogen and progesterone receptors and HER2, which are common therapeutic targets. CDK4/6 inhibitor Palbociclib has been approved as an anti-cancer agent for breast cancer. However, identifying biomarkers that predict the response to Palbociclib has always been a challenge for molecular targeted therapy. In this study, we identify microRNA as a hallmark in TNBC patients and explore if miR-3613-3p might serve as a tumor suppressor biomarker for triple negative breast cancer patients and if overexpression of miR-3613-3p could enhance the sensitivity of TNBC cells to Palbociclib. We show that the expression of miR3613-3p was down-regulated in TNBC tumors and cells, and the overexpression of miR-3613-3p in patients’ tumor tissues was clinically and pathologically correlated with favorable prognosis, such as smaller tumor size and the lower Ki-67. In vitro, overexpression of miR-3613-3p inhibited cell proliferation, induced G1 cell-cycle arrest, and enhanced the sensitivity of TNBC cells to Palbociclib treatment. In vivo study revealed that overexpression of miR-3613-3p inhibited TNBC tumorigenesis and exerted a significant inhibitory effect of Palbociclib on MDA-MB-231 cells. Mechanically, SMAD2 and EZH2 were found to be two direct targets of miR-3613-3p and mediate the proliferation of TNBC cells and the sensitivity of the cells to Palbociclib through inducing cellular senescence. Our findings suggested that miR-3613-3p acts as a cancer-suppressor miRNA in TNBC. Moreover, our study showed that miR-3613-3p might be used as a predictive biomarker for the response of TNBC to Palbociclib.

Scientific information extraction is a crucial step for understanding scientific publications. In this paper, we focus on scientific keyphrase extraction, which aims to identify keyphrases from scientific articles and classify them into predefined categories. We present a neural network based approach for this task, which employs the bidirectional long short-memory (LSTM) to represent the sentences in the article. On top of the bidirectional LSTM layer in our neural model, conditional random field (CRF) is used to predict the label sequence for the whole sentence. Considering the expensive annotated data for supervised learning methods, we introduce self-training method into our neural model to leverage the unlabeled articles. Experimental results on the ScienceIE corpus and ACL keyphrase corpus show that our neural model achieves promising performance without any hand-designed features and external knowledge resources. Furthermore, it efficiently incorporates the unlabeled data and achieve competitive performance compared with previous state-of-the-art systems.

Optimising the feed composition, especially the forage choice, and reducing costs are essential for improving dairy production efficiency. Different forage sources and proportions were used to formulate rations containing equal energy and protein, and their effects on rumen microbiota and milk production performance of dairy cows were evaluated in two experiments. In experiment 1, thirty lactating cows (235 ±13 d postpartum; milk production 29.1 ± 1.0 kg/day) were divided into Pangola and Bermuda groups. In experiment 2, twenty lactating cows (79.9 ± 8.1 d postpartum; milk production 34.7 ± 0.6 kg/day) were divided into Pangola and oat groups. In experiment 1, the Simpson index for rumen microbiota of the Pangola group was significantly higher than in the Bermuda group (P < 0.05). Analysis of the weighted unique fraction (UniFrac) distances indicated significant differences in the beta diversity of the community composition of rumen microbiota between Pangola, Bermuda and oat groups in both experiments (P < 0.001). The relative abundance of Prevotella brevis was significantly higher in the Pangola group than in the oat group in experiment 2 (P < 0.05). The somatic cell counts (SCCs), C18:0, and C18:1 in milk were significantly higher in the Bermuda group than in the Pangola group (P < 0.05) in experiment 1. On the other hand, milk crude protein (CP) and solids-not-fat (SNF) were significantly higher in the Oat group than in the Pangola group (P < 0.05) in experiment 2; however, milk urea nitrogen (MUN) was significantly higher in the Pangola group (P < 0.05). In conclusion, a switch of forage (Pangola vs Bermuda) at a lower proportion of the diet under the high forage level condition (experiment 1) caused only minor changes in rumen microbiota diversity (Simpson index, beta diversity) and milk production performance (milk SCCs, C18:0 and C18:1). On the other hand, a switch of forage (Pangola vs oat) at a higher proportion of the diet under the low forage level condition (experiment 2) resulted in greater changes in rumen microbiota diversity (beta diversity, relative abundances of bacterial taxa, P. brevis relative abundance) and milk production performance (milk CP, SNF, and MUN). [ABSTRACT FROM AUTHOR]

With increasing globalisation, organisations are confronted with intensifying market competition. Consequently, organisations continually implement novel technologies to address the challenges posed by market competition. In business process management (BPM), robotic process automation (RPA) is regarded as a tool or methodology that can automate specific tasks or activities within a business process. RPA is widely employed as a standard solution in various industries to assist enterprises in automating their processes, enhancing their operational efficiency, and reducing operational costs. Nevertheless, it is prudent to consider which processes can be automated by applying RPA and the most appropriate deployment methods for RPA. A literature review reveals that process identification is the optimal method for identifying processes suitable for automation, and three approaches - process mining, process quality-based and interview-based can be considered. There are three principal models for deploying RPA: local, cloud and hybrid. The RPA managers in an organisation are thus able to select the most appropriate deployment option based on the organisation's policies, requirements, and cost considerations. Furthermore, this study identifies the critical success factors (CSFs) and analyses their relationships using a 'why-why' diagram. This study could assist organisations in a more comprehensive understanding of RPA technology, which enables more effective utilisation of RPA for enhanced process management. [ABSTRACT FROM AUTHOR]