Search

Co-resident attacks are serious security threats in multi-tenant public cloud platforms. They are often implemented by building side channels between virtual machines (VMs) hosted on the same cloud server. Traditional defense methods are troubled by the deployment cost. The existing tenant classification methods can hardly cope with the real dataset that is quite large and extremely unevenly distributed, and may have problems in the processing speed considering the computation complexity of the DBSCAN algorithm. In this paper, we propose a novel co-resident attack defense strategy which solve these problems through an improved and efficient multi-level clustering algorithm and semi-supervised classification method. We propose a novel multi-level clustering algorithm which can efficiently reduce the complexity, since only a few parameter adjustments are required. Built on the proposed clustering algorithm, a semi-supervised classification model is designed. The experimental results of the classification effect and training speed show that our model achieves F-scores of over 85% and is significantly faster than traditional SVM classification methods. Based on the classification of unlabeled tenants into different security groups, the cloud service provider may modify the VM placement policy to achieve physical isolation among different groups, reducing the co-residency probability between attackers and target tenants. Experiments are conducted on a large-scale dataset collected from Azure Cloud Platform. The results show that the proposed model achieves 97.86% accuracy and an average 96.06% F-score, proving the effectiveness and feasibility of the proposed defense strategy. [ABSTRACT FROM AUTHOR]

Stripe rust fungus Puccinia striiformis f. sp. tritici (Pst) is a destructive pathogen of wheat worldwide. Pst has a macrocyclic-heteroecious lifecycle, in which one-celled urediniospores are dikaryotic, each nucleus containing one haploid genome. We successfully generated the first fully haplotype-resolved and nearly gap-free chromosome-scale genome assembly of Pst by combining PacBio HiFi sequencing and trio-binning strategy. The genome size of the two haploid assemblies was 75.59 Mb and 75.91 Mb with contig N50 of 4.17 Mb and 4.60 Mb, and both had 18 pseudochromosomes. The high consensus quality values of 55.57 and 59.02 for both haplotypes confirmed the correctness of the assembly. Of the total 18 chromosomes, 15 and 16 were gapless while there were only five and two gaps for the remaining chromosomes of the two haplotypes, respectively. In total, 15,046 and 15,050 protein-coding genes were predicted for the two haplotypes, and the complete BUSCO scores achieved 97.7% and 97.9%, respectively. The genome will lay the foundation for further research on genetic variations and the evolution of rust fungi. [ABSTRACT FROM AUTHOR]

Simple Summary: The Lower Tumen River basin habitat at the Sino-North Korean border is crucial for reestablishing Amur leopards in the Korean Peninsula, where they once thrived. However, except for the Jingxin–Dapanling (JD) and Mijiang (MJ) corridors, most areas have become impassable due to human activities and urbanization. In this study, we evaluated the effectiveness of the MJ corridor by analyzing the species abundance, forest structure, landscape features, and disturbance factors. Our findings indicate that leopard activity is predominantly concentrated in the northern part of the corridor, with little to no presence in the middle and southern regions near the North Korean border. Human disturbances, forest structure, and infrastructural obstacles seem to impede the movement of leopards. To ensure the resurgence of the leopard population in the Korean Peninsula, it is imperative to mitigate or eliminate the impacts of these hindrances. This entails reducing human disturbances, enhancing forest structure, and removing infrastructural barriers. Such efforts are vital to facilitate the revival of the Amur leopards in their former range in the Korean Peninsula. The interconnected forest regions along the lower Tumen River, at the Sino-North Korean border, provide critical habitats and corridors for the critically endangered Amur Leopard (Panthera pardus orientalis). In this region, there are two promising corridors for leopard movement between China and North Korea: the Jingxin–Dapanling (JD) and Mijiang (MJ) corridors. Past studies have confirmed the functionality of the JD corridor, but leopards' utilization of the MJ corridor has not yet been established or confirmed. In this study, we assessed the functionality of the MJ corridor. The study area was monitored using camera traps between May 2019 and July 2021. We also analyzed 33 environmental and vegetation factors affecting leopard survival and analyzed leopard movement. In the Mijiang area, the Amur leopard was mainly active in the region adjacent to the Northeast China Tiger and Leopard National Park and did not venture into area near the North Korean border. The complex forest structure allowed leopards to move into the Mijiang area. However, the high intensity of human disturbance and manufactured physical barriers restricted further southward movement. Therefore, human-induced disturbances such as grazing, mining, farming, logging, and infrastructure development must be halted and reversed to make the Mijiang region a functional corridor for the Amur leopard to reach the North Korean forest. This necessitates inter-governmental and international cooperation and is essential for the long-term survival of the Amur leopard. [ABSTRACT FROM AUTHOR]

Summary: Regulation of host gene expression to promote disease is a common strategy for plant pathogens. However, it remains unclear whether or not fungal pathogens manipulate host gene expression directly through secreted effectors with transcriptional activity.Here, we identified a fungal effector PstGTA1 from Puccinia striiformis f. sp. tritici (Pst), which has partial homology to the subunit of global transcriptional activator SNF2 from oyster. The transcriptional activating activity of PstGTA1 was validated in yeast, and the potential role of PstGTA1 in pathogenicity was assessed using gene silenced and overexpression transgenic wheat plants. Candidate targets regulated by PstGTA1 were screened by transcriptomic analysis, and the specific promoter region binding to PstGTA1 was further determined.PstGTA1 can be delivered to the wheat cell nucleus and contributes to the full virulence of Pst by targeting the promoter of TaSIG, a gene negatively regulating wheat immunity, and possibly activates its transcription by affecting the histone H3K4 acetylation level.Our study provides the first direct evidence for a fungal effector with transactivation activity modulating the transcription of a host specific susceptibility gene through promoter binding and histone acetylation. [ABSTRACT FROM AUTHOR]

Simple Summary: The water deer Hydropotes inermis, one of the most primitive members of Cervidae, is classified as vulnerable by the IUCN Red List in their native range, having declined drastically in recent years. In this study, we illustrated a northern extension of water deer using the current evidence, and assessed the status, phylogeny, and genetic ancestry of the newly recorded population. Our results showed that water deer had expanded to the Northeast China and the Russian Far East, where there had previously been no record of this species; thus, this could be a genuine range expansion rather than simply an expansion of the known range. A genetic investigation indicated that the expanding population had a close phylogenetic affinity with Korean water deer. The likely migration route and causes of the species' distribution range expansion are discussed. Given current anthropogenic pressures and climate change, wildlife range expansion offers a second chance for species conservation. The water deer Hydropotes inermis is a native to China and the Korean peninsula, but populations in North Korea and mainland China have declined drastically in recent years. However, the range of this species appears to be rapidly expanding northward. In this study, we employed camera traps and molecular technology to assess the status, phylogeny, and genetic ancestry of the newly recorded population. Our results showed an ongoing northward expansion of water deer, reaching at least 500 km from their historical distribution limit. We provided updated information on this species' geographical distribution in Northeast China and the Russian Far East. Based on historical survey data before the 1990s, there had previously been no record of this species in these two regions; thus, this could be a genuine range expansion rather than simply an expansion of the known range. A genetic investigation based on mitochondrial DNA indicated that the expanding population had a close phylogenetic affinity with Korean water deer. The likely migration route and causes of the species' distribution range expansion are discussed. We recommend revising the range of water deer in the IUCN Red List to facilitate the effective conservation and management of this threatened species, especially in new locations. [ABSTRACT FROM AUTHOR]

Global changes may direct species expansion away from their current range. When such an expansion occurs, and the species colonizes a new region, it is important to monitor the habitat used by the species and utilize the information to updated management strategies. Water deer (Hydropotes inermis) is listed as Vulnerable species in IUCN Red List and is restricted to east central China and the Korean Peninsula. Since 2017, water deer has expanded its range towards northeast China and the Russian Far East. The objective of our study is to provide support for a better understanding of habitat use and provide suggestions for developing conservation strategy. We collected occurrence data in northeast China and the Russian Far East during 2017–2021. We used MaxEnt to predict habitat suitability for water deer and applied Circuitscape to determine possible dispersal routes for the species. We used seven environmental variables, viz., altitude, slope, aspect, distance to built-up area, distance to water source, distance to cropland and distance to roads for habitat suitability prediction. We chose the MaxEnt model (AICc = 2572.86) suitable for our data with the AUC value result of 0.935±0.014. There is good quality habitat for water deer in the boundary area of the Yalu and Tumen River estuaries between China, North Korea, and the Russian Far East, as well as the east and west regions of the Korean Peninsula. We identified three main suitable habitat patches, two of them located in east (NK2) and west (NK3) North Korea, and one in the newly colonized area downstream of the Tumen River along the border of China, Russia, and North Korea (TM1). Elevation, distance to cropland and water sources, and presence of wetlands were the variables that positively contributed to modelling the suitable habitats. Two possible dispersal routes were determined using the circuit theory, one was across the area from North Korea to the downstream Tumen transboundary region (Route B), and the other was across North Korea to the boundary region in China and along the tiger national park in northern China (Route A). A series of protected areas in North Korea, China, and Russia may support the dispersal of water deer. From the study on water deer dispersal, we can understand the existing ecological network in northeast Asia, which will benefit the whole landscape and biodiversity conservation. However, there are many threats present, and there is need for continued monitoring inside and outside the protected areas. Information sharing with stakeholders and carrying out local communities awareness activities are important. The establishment of a Northeast Asia landscape conservation network would help establish monitoring and conservation planning at a broad scale, and this study provides an example of the need for such a network. [ABSTRACT FROM AUTHOR]

Face recognition methods for low resolution are often developed and tested on down‐sampled images instead of on real low‐resolution images. Although there is a growing awareness that down‐sampled and real low‐resolution images are different, few efforts have been made to analyse the differences in recognition performance. Here, the authors explore the differences and demonstrate that alignment is a major cause, especially in the absence of pose and illumination variations. The authors found that the recognition performances on down‐sampled images are flattered mostly due to the fact that the images are perfectly aligned before down‐sampling using high‐resolution landmarks, while the real low‐resolution images have much poorer alignment. To obtain better alignment for real low‐resolution images, the authors apply matching score‐based registration which does not rely on accurate landmarks. The authors propose to divide low resolution into three ranges to harmonise the terminology: upper low resolution (ULR), moderately low resolution (MLR), and very low resolution (VLR). Most face recognition methods perform well on ULR. MLR is a challenge for commercial systems, but a low‐resolution deep‐learning method can handle it very well. The performance of most methods degrades significantly for VLR, except for simple holistic methods which perform the best. [ABSTRACT FROM AUTHOR]

A very common case for law enforcement is recognition of suspects from a long distance or in a crowd. This is an important application for low‐resolution face recognition (in the authors' case, face region below 40 × 40 pixels in size). Normally, high‐resolution images of the suspects are used as references, which will lead to a resolution mismatch of the target and reference images since the target images are usually taken at a long distance and are of low resolution. Most existing methods that are designed to match high‐resolution images cannot handle low‐resolution probes well. In this study, they propose a novel method especially designed to compare low‐resolution images with high‐resolution ones, which is based on the log‐likelihood ratio (LLR). In addition, they demonstrate the difference in recognition performance between real low‐resolution images and images down‐sampled from high‐resolution ones. Misalignment is one of the most important issues in low‐resolution face recognition. Two approaches – matching‐score‐based registration and extended training of images with various alignments – are introduced to handle the alignment problem. Their experiments on real low‐resolution face databases show that their methods outperform the state‐of‐the‐art. [ABSTRACT FROM AUTHOR]