Search

It is essential to use fault tolerance techniques on exascale high-performance computing systems, but this faces many challenges such as higher probability of failure, more complex types of faults, and greater difficulty in failure detection. In this paper, we designed the Fail-Lagging model to describe HPC process-level failure. The failure model does not distinguish whether the failed process is crashed or slow, but is compatible with the possible behavior of the process due to various failures, such as crash, slow, recovery. The failure detection in Fail-Lagging model is implemented by local detection and global decision among processes, which depend on a robust and efficient communication topology. Robust means that failed processes do not easily corrupt the connectivity of the topology, and efficient means that the time complexity of the topology used for collective communication is as low as possible. For this purpose, we designed a torus-tree topology for failure detection, which is scalable even at the scale of an extremely large number of processes. The Fail-Lagging model supports common fault tolerance methods such as rollback, replication, redundancy, algorithm-based fault tolerance, etc. and is especially able to better enable the efficient forward recovery mode. We demonstrate with large-scale experiments that the torus-tree failure detection algorithm is robust and efficient, and we apply fault tolerance based on the Fail-Lagging model to iterative computation, enabling applications to react to faults in a timely manner. [ABSTRACT FROM AUTHOR]

[Display omitted] • Sterol geochemical features depend on lake natural processes and anthropogenic perturbations. • Cholesterol and total C 27 sterols are enriched in the sediments of deep lakes. • Sitosterol and total C 29 sterols are enriched in the sediments of shallow lakes. • Alpine lakes are characterized by an extreme dominance of sitosterol. • Cholesterol/brassicasterol ratio can be used to reconstruct historical lake nutrient changes. Sterol-based geochemical proxies are used to reconstruct the historical environmental evolution of lakes and their catchment areas. However, the reliability of these proxies remains challenging because the link between environmental drivers and sterol geochemical characteristics has not been fully assessed. In this study, the abundance and composition of sterols were determined in sediment cores from nine lakes with contrasting natural conditions and anthropogenic perturbations to establish the response pattern of sterol geochemical characteristics to environmental drivers, and the controlling mechanisms behind this phenomenon were also tentatively speculated. Our results revealed that cholesterol and total C 27 sterols were consistently enriched in the sediment cores from deep lakes (Fuxian, Lugu, and Erhai). In contrast, sitosterol and total C 29 sterols were consistently enriched in the sediment cores from shallow (Changdang, Taibai, Chaohu, and Liangzi) and alpine lakes (Heihai and Jiren). The variable sterol distribution can be attributed to different natural processes in these lakes. In deep lakes, the productive zooplankton, owing to the long water column, contributed to high cholesterol levels; in shallow lakes, the delivery of massive terrestrial organic matter through abundant inflowing rivers caused a large sitosterol influx. In alpine lakes, the rapid input of alpine meadows due to the dense growth of these plants on lake shores and adjacent areas resulted in the dominance of sitosterol. The significant influence of natural processes on sterol geochemical characteristics stated above implies that the lake natural background should be considered when sterol proxies were used in paleoenvironmental studies. Comparison of sterol ratios in sediments deposited in the oligotrophic and eutrophic phases of each lake revealed a marked influence of anthropogenic perturbations on sterol distribution. In particular, the cholesterol/brassicasterol ratio exhibited consistent and significant variation with nutrient enrichment in shallow or alpine lakes, but not in deep lakes, revealing the possibility of using this ratio as a nutrient indicator in the former two types of lakes. [ABSTRACT FROM AUTHOR]

Many pioneering approaches have verified the effectiveness of utilizing the global temporal and local object information for video understanding tasks and have achieved significant progress. However, existing methods utilize object detectors to extract all objects overall video frames. This may bring performance degradation due to the information redundancy both spatially and temporally. To address this problem, we propose an adaptive spatial location module for the video captioning task which dynamically predicts an important position of each video frame in the procedure of generating the description sentence. The proposed adaptive spatial location method not only makes our model focus on local object information, but also reduces time and memory consumption brought by the temporal redundancy in extensive video frames and improves the accuracy of generated description. Besides, we propose a balanced loss function to address the class imbalance problem existing in training data. The proposed balanced loss assigns different weight to each word of ground-truth sentence in the training process which can generate more diversified description sentences. Extensive experimental results on the MSVD and MSR-VTT dataset show that the proposed method achieves competitive performance compared to state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The rapid development and exploitation of the Yangtze River basin in order to ensure human food security and increase living space in recent decades has resulted in significant potential for degradation of water quality in the river and in hundreds of lakes. Understanding how lake environments have evolved to their present state under a variety of external influences is crucial for evaluating their current status and anticipating future scenarios of environmental changes. However, the lakes along the middle reaches of the Yangtze River (MRY) are as yet little studied. Here, we described the long-term anthropogenic environmental transformations of a small lake (Lake Taibai) in the MRY area, based on a detailed quantitative geochemical analysis of the aliphatic hydrocarbons, nutrients (N and P), biogenic silica (BSi), and major and trace elements present in a dated sediment core retrieved from the lake. Our data revealed that levels of short-chain n -alkanes, αβ-hopanes and the trace elements arsenic (As) and cadmium (Cd) were all low for the entire record in sediments prior to ca. 1970, reflecting unpolluted natural state of the lake. Pronounced anthropogenic effects began to appear in sediments deposited in the subsequent years ca. 1970–1990, during which the levels of all these components were elevated, most likely driven by input of nitrogen (N) and phosphorus (P) containing chemical fertilizers, pesticides and diesel oil respectively. Since ca. 1990, changes of short-chain n -alkane levels in the sediment suggested the lake had undergone dramatic eutrophication in which existing anthropogenic stressors were exacerbated by technological advances that extended the use of chemical fertilizer into aquaculture. This pattern contrasted with an otherwise comparable lake in the lower Yangtze River basin, Lake Changdang, in which trace element and petroleum pollution were much more prominent due to dramatic urbanization and industrialization of the catchment. • Lipid biomarkers and trace elements in sediment core were used to reflect lake pollution. • Geochemical variables remained low in sediments prior to 1970s, reflecting an unpolluted state. • Pollution of the lake in ca. 1970–1990 was driven by use of chemical fertilizers, pesticides and diesel oil. • Direct application of chemical fertilizers in aquaculture resulted in a massive pollution. [ABSTRACT FROM AUTHOR]

• Hypolimnetic deoxygenation intensified in Lake Fuxian since 1994. • Eutrophication since 1986 did not promote hypolimnetic deoxygenation. • Greater catchment disturbance in the 1950s and 1960s did not alter hypolimnetic deoxygenation. • Intensification of hypolimnetic deoxygenation was driven mainly by climate warming. In monomictic lakes, the potential intensification of hypolimnetic deoxygenation due to human activities and climate change threatens lake ecosystems and water quality. Understanding the trends and mechanisms of hypolimnetic deoxygenation would allow for the adaptive management of monomictic lakes. This study investigated the trends and drivers of hypolimnetic deoxygenation in Lake Fuxian (a monomictic lake in south-western China) over the past 150 years using paleolimnological evidence and historical records. Analysis of enrichment factors for molybdenum (Mo), uranium (U), and vanadium (V), and ratios of U/thorium (Th), pristane (Pr)/phytane (Ph), and total organic carbon (TOC)/total phosphorus (TP) in a dated sediment core indicated that the extent of hypolimnetic deoxygenation was relatively low before 1994 and increased substantially thereafter. Nutrient monitoring records and sediment n -alkane proxies indicated notable eutrophication of the lake since 1986 and a greater degree of human disturbance in the catchment between 1951 and 1967. Due to their asynchronous occurrence, eutrophication and catchment disturbance could not be the major drivers of hypolimnetic deoxygenation. Interestingly, climate records indicated that the local ambient temperature exhibited similar trends to that of hypolimnetic deoxygenation, including markedly higher temperatures after 1994. This implies that prolonged water column stratification due to climate warming could have promoted hypolimnetic deoxygenation by limiting oxygen exchange between the upper and lower water columns. In comparing the results with those of Lake Erhai, another monomictic lake with contrasting morphology and hydrology characteristics, we found that the low surface area/depth ratio and long water residence time in Lake Fuxian enable dissolved oxygen of the lake hypolimnion vulnerable to temperature over the lake. [ABSTRACT FROM AUTHOR]

• Long-term changes of OM input in Lake Chaohu sediments were assessed. • Anthropogenic activities influenced OM input via shaping local environment and ecosystems. • Agricultural activities enhanced terrestrial allochthonous OM input. • Dam installation inhabited OM input from submerged macrophytes. • Industrialization and urbanization promoted OM input from phytoplankton. Characterizing how and to what extent human activities have influenced organic matter (OM) input in lake sediments is essential for a thorough study of the carbon cycle and carbon burial in lake systems, but this issue has not yet been addressed. Here, the source and composition of OM in sediments of Lake Chaohu, Eastern China, over the past 166 years, were determined by analyzing stratigraphic variations in aliphatic hydrocarbons and OM pyrolysis parameters in a dated sediment core. These data, coupled with documental records, were further used to elucidate the pattern and controlling factors of OM input in the sediments. The lowest hydrogen index (HI), lowest long-chain n -alkane concentrations, and highest Paq (proxy of aquatic macrophyte input) in sediments prior to ca. 1956 indicate a minimal input of OM from phytoplankton and terrestrial plants and a larger input from submerged macrophytes. This may represent the features of OM input in primitive conditions given the mild human activities around the lake before the 1950s. In sediments from ca. 1962 to 1985, elevated HI and C 17 /C 16 n -alkane ratios indicate a greater input from phytoplankton, while the high abundances of long-chain n -alkanes and αβ-hopanes reflect an anomalously high input from terrestrial plant and fossil OM. These variations might have been caused by the rapid development of agriculture during this period, which accelerated catchment erosion and petroleum contamination and promoted the transport of allochthonous OM and nutrients to the lake. The lowest Paq in sediments during this period indicates reduced input from submerged macrophytes, probably resulting from a dam installation that modified the water to an unfavorable plant depth. In sediments after ca. 1989, the highest values of HI, short-chain n -alkane abundances, and C 17 /C 16 n -alkane ratios indicate a further increase in OM input from phytoplankton, a response to domestic and industrial discharge and subsequent lake eutrophication. [ABSTRACT FROM AUTHOR]

The carcinoembryonic antigen, glypican-3 (GPC3), is a putative therapeutic target and diagnostic marker of hepatoma. In the present study, a monoclonal antibody (mAb) specifically against GPC3 was obtained via cloning the sequence of GPC3 via polymerase chain reaction and inserting it into a pET16b vector prior to transfection into Escherichia coli (E. coli) BL21. BALB/c mice were immunized with 20 µg purified antigen by intrasplenic embedding. Splenocytes and mouse myeloma cells SP2/0 were fused; then, the hybridoma cells were screened by an indirect ELISA. The properties of the mAb were examined by western blotting and immunofluorescence analysis against the purified protein. The results revealed that the prokaryotic expression vector of GPC3 had been successfully generated and GPC3 was stably expressed in E. coli BL21. A stable hybridoma cell line, 2F3, was generated in the present study, which produced mAbs against GPC3. The mAb 2F3 had a high antibody titer and the isotype was identified as IgG1/κ; 2F3 hybridomas had a median chromosome number of 98. Western blot and immunofluorescence analyses revealed that 2F3 specifically recognized recombinant and native GPC3. The 2F3 clone was proposed as a stable secretor of this mAb against GPC3. The results of present study indicated that the successful preparation of recombinant GPC3 protein and an anti-human GPC3 mouse mAb may be provide a basis for developments in the diagnosis and treatment of liver cancer. [ABSTRACT FROM AUTHOR]

Highlights • Fatty acids and alcohols were analyzed in surface sediments of lakes in Eastern China. • Abundances of fatty acids were dependent on the type of organisms within the lakes. • Variation of trophic status in lakes exerted influence on some alcohols in sediments. • (C26 + C28)/(C22 + C24) n -alkanoic acids or n -alkanols can indicate transition of lake ecosystems. Abstract During the past hundred years, the majority of lakes along the middle and lower reaches of the Yangtze River (MLRYZ) have been greatly impacted by human intervention, resulting in a significant spatial heterogeneity in water and ecosystem quality. We quantified the fatty acid (FA), n -alkanol, sterol and long-chain alkyl diol content of surface sediments of MLRYZ lakes in order to determine how ecological differences (algae versus macrophyte-dominated state) and trophic status influence the composition, origin and abundance of lipid biomarkers in the sediments. Based on the results, variation in FAs appears to be mainly controlled by the type of organisms within the lakes. Short-chain n -alkanoic acids, monounsaturated FAs (MUFAs) and polyunsaturated FAs (PUFAs) were generally more abundant in sediments of algal lakes (lacking macrophyte growth) than in lakes with macrophyte growth. The opposite trend was apparent for long-chain n -alkanoic acids. Ratios of (C26 + C28)/(C22 + C24) n -alkanoic acids differed significantly between macrophyte and algal lakes. In the alcohol fraction, lipid biomarker abundance appeared to be controlled by trophic status of the lake, with the exception of cholesterol and long-chain n -alkanols whose abundances were closely related to the present organisms and exhibiting greater abundances in algal lakes. Enhanced levels of total phosphorus (TP) and total nitrogen (TN) in a lake create greater concentrations of C32 1,15 alkyl diol and brassicasterol respectively, while the abundance of dinosterol and the ratio of dinosterol/brassicasterol decline in line with increase in TP and TN, respectively. Ratios of (C26 + C28)/(C22 + C24) n -alkanol were also greatly influenced by the type of organisms present, exhibiting higher values in algal lakes. The sensitivity of these sediment lipid biomarker responses points to a potentially useful role as proxies for biological community types and trophic status in paleolimnology studies. [ABSTRACT FROM AUTHOR]

Highlights • The transmittance was determined by the atmosphere and the emitted energy. • The method of coupling analysis was established in the paper. • Materials with a greater distribution of energy should be priority considerations. Abstract This paper established a mathematical model to explore the characteristics of infrared energy of 8–14 μm emitted from an urban surface and analyzed the atmospheric transmittance under actual weather conditions. Then, the amount of infrared energy that penetrated the atmospheric infrared window was determined, based on the coupling analysis of urban radiation energy and atmospheric transmittance. Six typical urban surface materials were researched in this case study. The results showed that the transmittance was changed under different weather conditions and different urban surface materials had different infrared energy penetration of the atmospheric infrared window, which indicated different urban surface materials may have potentially different effects on the urban thermal environment. The results also indicated that the transmittance for different materials was affected not only by atmospheric transmittance but also by the distribution of emitted infrared energy along wavelengths, which provides a new perspective for related researchers to understand the phenomenon. [ABSTRACT FROM AUTHOR]

Aliphatic hydrocarbon biomarkers from the surface sediments of lakes along the lower reaches of the Yangtze River (Eastern China) were analyzed in order to determine the origin and composition of organic matter (OM) in the sediment. This is necessary to understand how anthropogenic environmental change in the region is impacting on OM dynamics and carbon cycles. The results indicate that OM in lake sediments is derived from biogenic sources, including phytoplankton, aquatic macrophytes, bacteria and terrestrial plants and abiotic sources of petroleum hydrocarbon contamination. The significance of each source varies greatly between lakes. Inputs of autochthonous OM to sediments were closely dependent on nutrient levels and the organisms present in the lake water column. Cyanobacterial OM input, signified by the distribution of short-chain n -alkanes, was most abundant in sediments from the most severely polluted lake, West Lake Chaohu. OM derived from diatoms, indicated by C 25 highly branched isoprenoids (HBIs), was relatively abundant in sediments from lakes Gucheng and Shijiu, the two studied lakes with the lowest nutrient levels. Growth of macrophytes in lakes Gucheng, Shijiu, Gehu and Yangcheng resulted in a significant accumulation of OM from these organisms in the sediments. In contrast, the input of allochthonous OM was controlled mainly by natural and anthropogenic activities around the lakes. For example, terrestrial plant OM input, deduced from the abundance and proportion of long-chain n -alkanes, was exceptionally high in the sediment of Lake Nanyi, perhaps stemmed from frequent severe flooding in the catchment. Input of fossil fuel OM reflected direct petroleum contamination of the water or anthropogenic petroleum burning in the catchment. In addition, the allochthonous refractory OM tended to be enriched in sediments of lakes with strong hydrodynamic process. The results of this study provided a lot of detail in explaining the variation of terrestrial OM dynamics and carbon cycle in response to anthropogenic activities. [ABSTRACT FROM AUTHOR]

During the past hundred years, the Lake Taihu basin has been greatly impacted by human interventions. The undesirable changes in water quality of lakes, presumably caused by the human activities, remain relatively undescribed in this area. In order to investigate these anthropogenic effects, a 210 Pb dated sediment core from a relatively small lake in the upper reaches of Lake Taihu known as Lake Changdang was subject to a detailed lipid biomarker study and other geochemical analyses including quantification of biogenic silica (BSi), nutrients and heavy metals. Based on the results, the recent environmental history of Lake Changdang can be divided into three periods. The first period from approximately 1906 to 1950, represents a natural state, with minimal anthropogenic impact on the lake. Human induced environmental change is recorded in the following stage, ca. 1950–1982, during which the trophic status of the lake increased slightly in response to inputs of agricultural waste and nitrogen (N) and phosphorus (P) based fertilizers. In particular, the signs of eutrophication during this phase accelerated from ca. 1973, perhaps owing to large-scale using phosphate based chemical fertilizer around the lake at the time. A second phase of nutrient input in the most recent stage, ca. from 1982 to 2016, initiated by wastewater discharge from rapid urbanization and industrialization of the catchment, greatly enhanced the nutrient level in the lake. However, the central zone of the lake has yet to reach a phytoplankton-dominated stable state, with both algae and aquatic macrophytes tracking an increased trend in productivity driven by intermediate nutrient levels in the water of this zone. [ABSTRACT FROM AUTHOR]

Salient object detection is a fundamental problem in computer vision. Existing methods using only low-level features failed to uniformly highlight the salient object regions. In order to combine high-level saliency priors and low-level appearance cues, we propose a novel Background Prior based Salient detection method (BPS) for high-quality salient object detection. Different from other background prior based methods, a background estimation is added before performing saliency detection. We utilize the distribution of bounding boxes generated by a generic object proposal method to obtain background information. Three background priors are mainly considered to model the saliency, namely background connectivity prior , background contrast prior and spatial distribution prior , allowing the proposed method to highlight the salient object as a whole and suppress background clutters. Experiments conducted on two benchmark datasets validate that our method outperforms 11 state-of-the-art methods, while being more efficient than most leading methods. [ABSTRACT FROM AUTHOR]

A concise, straightforward and efficient method has been developed for the synthesis of α-aminonitriles by an one-pot three-component condensation of aldehydes or ketones, amines and trimethylsilyl cyanide catalyzed by MgI 2 etherate under solvent-free conditions. This protocol has some advantages such as mild reaction condition, simple work-up, short reaction time and high product yields. [ABSTRACT FROM AUTHOR]

• We evaluated the reliability of n -alkane proxies for recording nutrient changes in lakes. • Common proxies such as short-chain n -alkane abundances were useful in relatively eutrophic lakes. • Most proxies failed in relatively oligotrophic lake due to influence from submerged macrophytes. • CPI 15–20 and n -C 17 / n -C 16 alkane can record nutrient changes in all kinds of lakes. • CPI 15–20 and n -C 17 / n -C 16 alkane had a potential to keep constant during OM diagenesis. Profiles of n -alkane proxies in sediment cores have been frequently used to detect historical trends in the nutrient status of lakes. However, the results obtained from these proxies are occasionally confusing because the situations in which a given n -alkane proxy might make a valid indication have tended thus far to be elusive. The current study compared geochemical characteristics of n -alkanes in sediments from the oligotrophic (before 1950) and relatively eutrophic phases (after 1980) of five representative lakes in order to assess their suitability and reliability as proxies for lake trophic state. The results from Lakes Liangzi, Changdang, Taibai and Erhai demonstrated that short-chain odd carbon n -alkane abundances were significantly higher (P < 0.05) in sediments deposited after 1980 than in those dating to before 1950, whereas TAR (terrigenous: aquatic ratios) and ACL 15–33 (average chain length of C 15 –C 33 n -alkanes) were significantly lower (P < 0.05) in the more recent sediments. These results are compatible with generally accepted variation seen in lakes that have undergone eutrophication. In Lake Fuxian, however, short-chain odd carbon n -alkane abundances, TAR and ACL 15–33 were all higher in sediments after 1980 than in sediments before 1950, although the variation was only significant (P < 0.05) for ACL 15–33. These results, especially the unexpected trends in TAR and ACL 15–33 , mean these cannot be relied upon as proxies for eutrophication in this lake. The anomaly is likely to be related to the predominantly low phytoplankton productivity and OM input in Lake Fuxian, coupled with the availability of OM from submerged macrophytes in the relatively eutrophic phase. Interestingly and importantly, two new proxies CPI 15–20 ((n -C 15 + n -C 17 + n -C 19)/(n -C 16 + n -C 18 + n -C 20) alkanes) and n -C 17 / n -C 16 alkane, which are indicative of OM input from phytoplankton over non-photosynthetic bacteria, were significantly higher (P < 0.05) in post-1980 sediments than in those dating to before 1950 in all five studied lakes. These two are thus shown to be widely applicable for recording lake eutrophication. Their presence in Lake Fuxian can be explicitly attributed to the absence of influence from submerged macrophytes. Moreover, as ratio results of n -alkanes with similar carbon number, these two proxies are less influenced by processes of OM diagenesis due to comparable degradation of n -alkanes in sediments. The combined results suggest that these previously overlooked proxies may be of considerable value to studies of lake trophic history. [ABSTRACT FROM AUTHOR]

In order to elucidate the effect of recent warming, we studied lipid biomarkers and trace elements in a dated sediment core from Lake Heihai, a small, deep, and ultraoligotrophic alpine lake in Yunnan Province (SW China), being only marginally affected by anthropogenic activities. The variation in lipid biomarkers (such as 10-methyl-C16:0 fatty acid (FA), iso-branched C15 (i-C15) and anteiso-branched C15 (ai-C15) FA, and tetrahymanol) suggests a rapid productivity increase in sulfate reducing bacteria and ciliates since 1980, likely reflecting expansion of the hypolimnion anoxia and a prolonged duration of an oxic–anoxic chemocline in the water column. The concentrations of element molybdenum (Mo) in pre-1980 sediments approach the values in average crust. After 1980, the concentration increased, reaching levels approximately sixfold higher than the initial abundances. This likely reflects a high authigenic Mo deposition when the bottom water was more anoxic and enrichment in H2S. The suggested spatial and temporal expansion of the anoxic bottom water since 1980 was probably a response to the regional climate warming, resulting in stronger water column stratification and terrestrial grass inputs to the lake, and thus higher dissolved oxygen (DO) loss in hypolimnion. [ABSTRACT FROM AUTHOR]

• Geochemical behaviors of n -alkanes were studied during hydrocarbon expulsion. • Higher hydrocarbon production potential can result in earlier expulsion of hydrocarbons. • The < C 17 n -alkanes were expelled preferentially than > C 18 n -alkanes. • The < C 17 n -alkanes might have been expelled earlier by molecule diffusion. • The > C 18 n -alkanes were likely expelled in a bulk oil flow mode. Lipid biomarker parameters and molecular carbon isotope compositions of n -alkanes were used as tracers to determine geochemical behavior of aliphatic hydrocarbons during their migration from two sequences of source rock-type mudstones (4050.8–4062.2 and 4067.4–4073 m depth, respectively) to the interbedded reservoir sandstones (4062.6–4066.3 m depth) in the Qikou Sag, Bohai Bay Basin, eastern China. Abnormally high values of the production index (0.69–0.90) and total extract/total organic carbon (TOC) ratios (0.53–1.21) in the sandstone samples indicated that the hydrocarbons in the sandstones were not produced in situ but migrated from the adjoining mudstones. A comparison of values in lipid biomarker parameters (oleanane index, gammacerane index, C 23 tricyclic terpane/C 30 αβ-hopane, and C 29 /C 27 ααα20R sterane ratios) in the sandstone with the mudstone samples indicated steroids and hopanoids in the upper section of the sandstone layer (4062.6–4065.2 m depth) were transported from the overlying mudstones, while those in the lower sandstone layer (4066.3 m depth) were transported from the underlying mudstones. Molecular carbon isotope data suggest that all n -alkanes in the upper section of the sandstone layer were transported from the overlying mudstones. However, in the lower sandstone layer, most of the C 15 –C 17 n -alkanes were transported from the overlying mudstones, while the C 18 –C 31 n -alkanes were transported mainly from the underlying mudstones. These combined results suggest the short-chain n -alkanes (C 18) and steroids/hopanoids may have been expelled in a separate oil phase. [ABSTRACT FROM AUTHOR]

The prevailing of Web 2.0 techniques has led to the boom of web video content as well as its social network. To overcome the information overload problem, effective web video topic discovery and structuring techniques are highly demanded. To this end, existing works go to two respective directions: video topic discovery based on content or community detection in social network, with limited interplay between topics and network structures. In this paper, we construct the video social network based on web user interactions over videos. By comparing the topics and communities discovered on this network, we unveil the loose correspondence relationship between content and social network, and correspondingly propose a novel community-driven web video topic discovery model, which regularizes the topic model in relaxed community-level. Quantitatively analysis on real-world YouTube data shows that our model has achieved a significant improvement over the purely content-based or network-based baselines. Meanwhile, we propose a community-based topic structuralization framework, which decomposes a topic in social network space, and tracks the spreading trajectory of this topic among different communities on the time line. This structuralization can help users to catch the important facets of topics, such as “Who is interested with this topic” and “How does it propagate among the communities”, which provide valuable insights in related applications such as web monitoring and market development. [ABSTRACT FROM AUTHOR]

Reactive oxygen species (ROSs) are generated in organisms in response to infections caused by invading microbes. However, excessive ROSs will inflict oxidative damage on the host. Peroxiredoxins (Prxs) are antioxidative enzymes that may eliminate ROSs efficiently. In this study, ApPrx1 from the pea aphid Acyrthosiphon pisum was cloned, and its function was investigated in vitro and in vivo . In the presence of DTT, recombinant ApPrx1 protein from Escherichia coli showed antioxidative activity by eliminating H 2 O 2 effectively. The H 2 O 2 levels were significantly higher in Micrococcus luteus -infected aphids than in uninfected aphids, and ApPrx1 expression was remarkably up-regulated when the aphids were infected with M. luteus or injected with H 2 O 2 . When ApPrx1 expression was reduced by dsRNA injection, the survival of the aphids decreased significantly after M. luteus infection. Knockdown of ApPrx1 decreased M. luteus loads inside the aphids 48 h post-infection. While under infection conditions, the H 2 O 2 levels were much higher in ApPrx1 knockdown aphids than in dsGFP -injected aphids, indicating that the decreased survival of the aphids was caused by increased oxidative stress. Taken together, our results reveal that ApPrx1 plays a protective role in oxidative stress caused by bacterial infection. [ABSTRACT FROM AUTHOR]

Image reranking is an effective way for improving the retrieval performance of keyword-based image search engines. A fundamental issue underlying the success of existing image reranking approaches is the ability in identifying potentially useful recurrent patterns from the initial search results. Ideally, these patterns can be leveraged to upgrade the ranks of visually similar images, which are also likely to be relevant. The challenge, nevertheless, originates from the fact that keyword-based queries are used to be ambiguous, resulting in difficulty in predicting the search intention. Mining useful patterns without understanding query is risky, and may lead to incorrect judgment in reranking. This paper explores the use of click-through data, which can be viewed as the footprints of user searching behavior, as an effective means of understanding query, for providing the basis on identifying the recurrent patterns that are potentially helpful for reranking. A new reranking algorithm, named click-boosting multi-modality graph-based reranking, is proposed. The algorithm leverages clicked images to locate similar images that are not clicked, and reranks them in a multi-modality graph-based learning scheme. Encouraging results are reported for image reranking on a real-world image dataset collected from a commercial search engine with click-through data. [ABSTRACT FROM AUTHOR]

We describe a convenient and useful procedure for the synthesis of various 2,5-dimethyl-N-substituted pyrrole derivatives by the addition of 2,5-hexadione with aromatic amines, heteroaromatic amines and aliphatic amines catalyzed by MgI2 etherate (MgI2·(OEt2) n) in good to excellent yields. [ABSTRACT FROM AUTHOR]

High Efficiency Video Coding (HEVC) provides superior coding efficiency than previous video coding standards at the cost of increasing encoding complexity. The complexity increase of motion estimation (ME) procedure is rather significant, especially when considering the complicated partitioning structure of HEVC. To fully exploit the coding efficiency brought by HEVC requires a huge amount of computations. In this paper, we analyze the ME structure in HEVC and propose a parallel framework to decouple ME for different partitions on many-core processors. Based on local parallel method (LPM), we first use the directed acyclic graph (DAG)-based order to parallelize coding tree units (CTUs) and adopt improved LPM (ILPM) within each CTU (DAGILPM), which exploits the CTU-level and prediction unit (PU)-level parallelism. Then, we find that there exist completely independent PUs (CIPUs) and partially independent PUs (PIPUs). When the degree of parallelism (DP) is smaller than the maximum DP of DAGILPM, we process the CIPUs and PIPUs, which further increases the DP. The data dependencies and coding efficiency stay the same as LPM. Experiments show that on a 64-core system, compared with serial execution, our proposed scheme achieves more than 30 and 40 times speedup for $1920\times 1080$ and $2560\times 1600$ video sequences, respectively. [ABSTRACT FROM AUTHOR]

In this article, we propose a unified geolocation framework to automatically determine where on the earth a web video was shot. We analyze different social, visual, and textual relationships from a real-world dataset and find four relationships with apparent geography clues that can be used for web video geolocation. Then, the geolocation process is formulated as an optimization problem that simultaneously takes the social, visual, and textual relationships into consideration. The optimization problem is solved by an iterative procedure, which can be interpreted as a propagation of the geography information among the web video social network. Extensive experiments on a real-world dataset clearly demonstrate the effectiveness of our proposed framework, with the geolocation accuracy higher than state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Lacustrine hydrocarbon source rocks with high quantity and quality of organic matter (OM) are developed within the third member of the Eocene Shahejie formation (Es 3) in the Qikou Sag of the Huanghua Depression, Bohai Bay Basin, Eastern China. However, the paleolake environments associated with deposition of these rocks remain relatively undescribed, restricting our understanding of the biogeochemical processes which contributed to the development of excellent source rocks and inhibiting efforts for exploration of lacustrine hydrocarbon deposits. In this study, 18 core samples from the Es 3 interval of Well Gangshen 4 in Qikou Sag were subject to a detailed analysis of bulk OM, biomarkers, molecular carbon isotopes, trace elements, and phosphorus (P) to investigate the environment and ecosystem of the ancient lake in which the rocks were deposited. High values of gammacerane index (av. 0.35) recorded in the samples reveals the oxic-anoxic stratification in the water column that persisted for the majority of the year due to the hot climate and great depth of the lake during this period. Meanwhile 13C-depleted hopanes (lower than −63.2‰) indicate an input of OM from chemoautotrophic and methanotrophic bacteria, supported by the production of CO 2 and CH 4 due to the anaerobic respiration of OM in the hypolimnion or sediments. The abundances of total organic carbon (TOC) relative to total P (TOC/P, mole ratio, av. 289.9) indicates the preferential release of P from sediments in an anoxic environment. However, the low (<2.0) Ni/Co and V/Cr ratios and moderate (av. 1.48) pristane/phytane (Pr/Ph) ratios suggest periodic oxidation of the hypolimnion, which may be a result of replenishment of oxygen during the short-term mixing of the water column driven by seasonal cooling and wind disturbance. Excess P was introduced into the epilimnion via mixing, which initiated high algal productivity after further stratification of the water column. Algal OM produced in this early stratification phase was crucial for the formation of excellent source rocks. The large variation of δ13C in TOC, n -alkanes, and isoprenoids reflects changes in CO 2 concentrations or carbon isotopes in the photic zone, which may be attributable to increased inputs of terrestrial OM in response to a changing climate. • Algae provided the crucial OM for developing source rocks in the Es 3 member of the Qikou Sag, Bohai Bay Basin. • The development of anoxic bottom water promoted P release from sediments to the overlying water. • Massive input of internal P to the surface water fueled algal productivity. • Excessive input of terrestrial materials diluted OM abundance and deteriorated OM quality. [ABSTRACT FROM AUTHOR]

Hashing-based similarity search techniques is becoming increasingly popular in large data sets. To capture meaningful neighbors, the topology of a data set, which represents the neighborhood relationships between its subregions and the relative proximities between the neighbors of each subregion, e.g., the relative neighborhood ranking of each subregion, should be exploited. However, most existing hashing methods are developed to preserve neighborhood relationships while ignoring the relative neighborhood proximities. Moreover, most hashing methods lack in providing a good result ranking, since there are often lots of results sharing the same Hamming distance to a query. In this paper, we propose a novel hashing method to solve these two issues jointly. The proposed method is referred to as topology preserving hashing (TPH). TPH is distinct from prior works by also preserving the neighborhood ranking. Based on this framework, we present three different TPH methods, including linear unsupervised TPH, semisupervised TPH, and kernelized TPH. Particularly, our unsupervised TPH is capable of mining semantic relationship between unlabeled data without supervised information. Extensive experiments on four large data sets demonstrate the superior performances of the proposed methods over several state-of-the-art unsupervised and semisupervised hashing techniques. [ABSTRACT FROM AUTHOR]

Lake Fuxian is a large, deep, and oligotrophic freshwater lake. In recent sediment of this lake, a series of branched and cyclic alkanes, including branched alkanes with one or two quaternary carbon atoms (C compounds), alkylcycloheptane, and alkylcyclohexane, were detected in high abundance. For each homolog of long-chain branched/cyclic alkanes, molecules with only an odd or even carbon number were present in the sediment. The abundance variation of each compound homolog corresponded to those of other compound homolog in the section, indicating the same origin of these compounds. The concentrations of long-chain branched/cyclic alkanes in the sediment profile exhibited variations similar to those of phytoplankton biomarkers, such as 3-methyl-alkanes, short-chain odd carbon n-alkanes, hopene, polyunsaturated fatty acids, phytol, sterols, and long-chain alkyl diols, but slightly different from those of long-chain odd carbon n-alkanes, which are ubiquitous biomarkers for higher plants. Therefore, the C compounds and long-chain cyclic alkanes in the lake sediment were most likely produced by algae and bacteria in the water, primarily, by the heterotrophic bacteria in the moderately oxygenated waters, which flourishing might be the results of specific water environments in Lake Fuxian. [ABSTRACT FROM AUTHOR]

Abstract: Mobile visual search is a new class of applications that use images taken by camera phone to initiate search queries. It is a very challenging task mainly because of image affine transformations caused by viewpoints changes, and motion blur due to hand tremble. These problems are unavoidable in mobile visual search and often result in low recall. Query expansion is an effective strategy for recall improvement, but existing methods are highly memory and time consuming, and often involve lots of redundant features. Integrating robust local patch mining and geometric parameter coding, this paper proposes an accurate offline query expansion method for large-scale mobile visual search. Concretely, a novel criterion is presented for robust patch evaluation and mining. Then multiple representative features are extracted from these selected local patches to deal with viewpoint changes. Moreover, the geometric parameter of each representative viewpoint is also recorded, to support fast and accurate feature matching. Experimental results on several well-known datasets and a large image set (1M) have demonstrated the effectiveness and efficiency of our method, especially its high robustness to viewpoint changes. The proposed approach can also be well generalized to other multimedia content analysis tasks. [Copyright &y& Elsevier]

Abstract: In recent years, Locality sensitive hashing (LSH) has been popularly used as an effective and efficient index structure of multimedia signals. LSH is originally proposed for resolving the high-dimensional approximate similarity search problem. Until now, many kinds of variations of LSH have been proposed for large-scale indexing. Much of the interest is focused on improving the query accuracy for skewed data distribution and reducing the storage space. However, when using LSH, a final filtering process based on exact similarity measure is needed. When the dataset is large-scale, the number of points to be filtered becomes large. As a result, the filtering speed becomes the bottleneck of improving the query speed when the scale of data becomes larger and larger. Furthermore, we observe a “Non-Uniform” phenomenon in the most popular Euclidean LSH which can degrade the filtering speed dramatically. In this paper, a pivot-based algorithm is proposed to improve the filtering speed by using triangle inequality to prune the search process. Furthermore, a novel method to select an optimal pivot for even larger improvement is provided. The experimental results on two open large-scale datasets show that our method can significantly improve the query speed of Euclidean LSH. [Copyright &y& Elsevier]

• Short-chain n -alkanol profiles in lake sediments were influenced by eutrophication and aquaculture. • Sources of n -alkanols (>C 20) in lakes were dependent on transparency and water residence time. • (C 26 + C 28)/(C 22 + C 24) n -alkanols showed a marked variation between shallow and deep lakes. • (C 26 + C 28)/(C 22 + C 24) n -alkanol ratios in shallow lakes were influenced by eutrophication and aquaculture. In order to examine how natural backgrounds and anthropogenic activities influenced the n -alkanol profiles of lake sediments, abundances and compositions of n -alkanols were determined in sediment cores taken from four lakes with notable differences in depth, nutrient level, transparency, water residence time and human impact. The results indicate the short-chain n -alkanols (C 14 , C 16 and C 18) are generally more abundant in sediments from eutrophic lakes than in oligotrophic bodies, due to greater inputs from algae and zooplankton. In individual lakes, however, high levels of short-chain n -alkanols were not always recorded in sediments of relatively eutrophic phases because contributions from zooplankton can be significant even if the lake remains relatively oligotrophic. The >C 20 n -alkanol profiles varied markedly between shallow and deep lakes. A prevalence of long-chain compounds (C 26 , C 28 , C 30 and C 32) was observed in shallow lakes and is attributable to terrestrial organic matter (OM) carried by inflowing rivers, whose relatively huge annual input is evidenced by the short water residence time in these lakes. Sediments from deep lakes were dominated by middle-chain n -alkanols (C 20 , C 22 and C 24) as a result of significant input from submersed macrophytes, which achieve high productivity in the clear, sunlit water. In shallow lakes, (C 26 + C 28)/(C 22 + C 24) n -alkanol ratios in sediments were greatly dependent on anthropogenic eutrophication and aquaculture via their influence on the productivity of submersed macrophytes. [ABSTRACT FROM AUTHOR]

Highlights: [•] A new common visual patterns (CVPs) discovery framework is proposed. [•] It contains three successive steps: initialization, expansion and combination. [•] It can effectively deal with image photometric and geometric transformations. [•] The CVPs can be used for object recognition and near-duplicate image retrieval. [ABSTRACT FROM AUTHOR]

This paper studies the data redundancy of the coefficient matrix of the corresponding discrete system which forms a basis for fast algorithms of solving the integral equation whose kernel includes a convolution function factor. We develop lossless matrix compression strategies, which reduce the cost of integral evaluations and the storage to linear complexity, i.e., the same order of the approximation space dimensions. We establish that this algorithm preserves the convergence order of the approximate solution. We also propose a hardware-aware parallel algorithm for these strategies. [ABSTRACT FROM AUTHOR]

Although the distribution of mono-, di- and trimethylated 2-methyl-2-(4,8,12-Trimethyltrideeyl) chromans (MTTCs) is well understood as an indicator of water salinity, their origin and formation mechanism are still ambiguous and under debate. In this paper, abnormally high levels of MTTCs were detected in Cenozoic saline lacustrine source rocks from the Western Qaidam Basin. Using a two-step column chromatography method, the MTTCs and naphthalenes were separated from other aromatic compounds and concentrated in one fraction, so that the stable carbon isotope compositions of these compounds could be accurately measured. Similar carbon isotope ratios for the mono-, di- and trimethylated 2-methyl-2-(4,8,12-Trimethyltrideeyl) chromans in a given sample suggest the MTTCs may share the same biological source(s). The MTTCs from the Western Qaidam Basin have similar carbon isotope compositions to primary producer-derived pristane and phytane. However, the δC values for the MTTCs showed significant differences with other primary producers-derived biomarkers (e.g. C and C steranes), probably indicating a distinct microalgae source from specific niches for the MTTC compounds. We speculate that the MTTCs distribution patterns may be controlled by the water chemistry dynamics in niches, which, in turn, is affected by hypersaline bottom water. [ABSTRACT FROM AUTHOR]

Invariant features extraction is important for object detection. Affine-SIFT (ASIFT) [J.M. Morel and G. Yu, ASIFT: A new framework for fully affine invariant image comparison, SIAM J. Imaging Sci. 2(2) (2009)] has been proved to be fully affine-invariant. However, the high cost of memory and query time hampers its application in large-scale object detection tasks. In this paper, we present a novel algorithm for mining concise and distinctive invariant features called affine-stable characteristics (ASC). Two new notions, global stability and local stability, are introduced to calculate the robustness of each feature from two mutually complementary aspects. Furthermore, to make these stable characteristics more distinctive, spatial information taken from several representative scales is encoded in a concise method. Experiments show that the robustness of our ASC is comparable with ASIFT, while the cost of memory can be reduced significantly to only 5%. Moreover, compared with the traditional SIFT method [D. Lowe, Distinctive image features from scale invariant keypoints, Int. J. Comput. Vis. 60(2) (2004), pp. 91–110], the accuracy of object detection can be improved 38.6% by our ASC using similar amount of features. [ABSTRACT FROM AUTHOR]

The saline lacustrine deposit of the Oligocene Lower Ganchaigou Formation is the main source rock for the Western Qaidam Basin, NW China. In this study, abundant highly branched isoprenoids with 25 carbon atoms (CHBI) were detected in the upper section of the Lower Ganchaigou Formation. CHBI is a biomarker for diatoms, and can provide information regarding biogeochemical processes during production and preservation of sedimentary organic matter. The carbon isotopic values of CHBI in these source rocks were in the range of −18‰ to −20‰. The relative enrichment in C of CHBI suggests that these isoprenoids were derived from diatom blooms. In this case, the diatoms used HCO as an additional carbon source to dissolved CO in water column due to their high biological productivity. Conversely, the diatom blooms indicated an abundant nutrient supply to the environment and high primary productivity. Thus, the occurrence of C-enriched CHBI could be an ideal marker for good source rocks. The decreased concentration of CO (aq) in the water column induced by diatom blooms can result in an enrichment in C of organic matter synthesized by primary producers. This may be a possible reason for the occurrence of C-enriched organic matter in the Cenozoic source rocks of the Western Qaidam Basin. Previous studies have shown that the high carbon number n-alkanes sourced from diatoms have no carbon preference. Hence, the sources of n-alkanes in regional source rocks are complex, and should be considered when discussing the sources of organic matter in future studies. [ABSTRACT FROM AUTHOR]

Abstract: Anemonin (the dilactone of cyclobutane-1, 2-diol-1, 2-diacrylic acid) was isolated from the root of Pulsatilla chinensis Regel. Pulsatilla chinensis Regel has been used in the treatment of enteritis in China for years. However, only little was known about the mechanism underlying its anti-inflammatory effects. We investigated the effect of anemonin on the release of nitric oxide (NO), endothelin-1 (ET-1) and soluble intercellular adhesion molecule-1 (sICAM-1) induced by lipopolysaccharide (LPS) in primary cultures of rat intestinal microvascular endothelial cells (RIMECs). RIMECs were challenged with 1μg/ml LPS with or without the presence of various concentrations of anemonin (1, 5 and 10μg/ml). Anemonin significantly inhibited the production of NO and ET-1 induced by LPS at a concentration of 5μg/ml and at 10μg/ml anemonin down-regulated LPS-induced sICAM-1 expression. Anemonin itself had no effect on either factor. These findings suggest that anemonin may exert some beneficial therapeutic action in intestinal inflammation, at least in part by inhibiting the production of NO, ET-1 and ICAM-1 in RIMECs and thus preventing intestinal microvascular dysfunction. [Copyright &y& Elsevier]

• FAs in surface sediments of Lake Taihu showed a spatial heterogeneity. • The abundance of phytoplankton FAs is markedly affected by degradation. • Terrestrial FAs were abundant in regions with adjoining inflowing rivers. • All FAs were low in the region subject to greatest sediment resuspension. Fatty acids (FAs) present in the surface sediments of Lake Taihu were analyzed for spatial heterogeneity in abundance and composition in order to test the suitability of these lipids to indicate variation in sources and preservation status of organic matter (OM) in sediments. The results indicate that terrestrial plant-derived FAs were relatively enriched in Meiliang and Zhushan Bays, the most eutrophic regions of Lake Taihu, while FAs of phytoplankton origin were relatively low. This unexpected result might reflect a preferential degradation of phytoplankton OM and phytoplankton-derived FAs. Furthermore, massive inputs of refractory terrestrial OM could significantly dilute concentrations of phytoplankton FAs. In sediments from East Bay, phytoplankton-derived FAs were more abundant than in Zhushan Bay, although this region had the lowest levels of nutrients and the lowest phytoplankton productivity. This result was attributed to a greatly enhanced preservation of labile FAs, probably caused by reduced sediment resuspension in this region with dense growth of submerged macrophytes. Phytoplankton productivity and terrestrial OM inputs were moderate in Central Region, but abundances of all FAs and ratios of C16:1/ n -C16, C18:1/ n -C18 and C18:1ω7/C18:1ω9 both reached their lowest in sediments from this region, possibly as a result of significant degradation of FAs driven by strong sediment resuspension. Sediments from Gonghu Bay and Southern Region were relatively enriched in phytoplankton FAs, and this was ascribed to low inputs of terrestrial OM, which limited the dilution effect. In contrast to FAs, short-chain n -alkane abundances provided an accurate indication of variation in phytoplankton input. [ABSTRACT FROM AUTHOR]

Cross-Domain Few-Shot Learning (CD-FSL) aims at recognizing unseen classes from target domains that vastly differ from training classes from source domains, utilizing only a few labeled samples. However, the substantial domain disparities between target and source domains pose huge challenges to few-shot generalization. To resolve domain disparities, we propose HybridPrompt, a novel architecture for Domain-Aware Prompting that integrates a variety of cross-domain learned prompts as knowledge experts for CD-FSL. The proposed method enjoys several merits. First, to encode knowledge from diverse source domains, several Domain Prompts are introduced to capture domain-specific knowledge. Subsequently, to facilitate the cross-domain transfer of valuable knowledge, a Transferred Prompt is specifically tailored for each target task by retrieving highly relevant Domain Prompts based on domain properties. Finally, to complement insufficient transferred information, an Adaptive Prompt is learned to incorporate additional target characteristics for model adaptation. Consequently, the collaboration of these three types of prompts contributes to a hybridly prompted model that achieves domain-aware encoding, transfer, and adaptation, thereby enhancing adaptability on unseen domains. Extensive experimental results on the Meta-Dataset benchmark demonstrate that our method achieves superior performance against state-of-the-art methods. The source code is available at https://github.com/Jamine-W/HybridPrompt. [ABSTRACT FROM AUTHOR]

Transcatheter arterial embolization has been used clinically for the treatment of intermediate to advanced unresectable tumors, which is a complex procedure and may pose other risks. In order to achieve an efficient and simple therapeutic diagnostic method for tumors, this paper is based on the fact that the fluorescence properties of AIE fluorescent molecules are similar to those of pH-responsive polyamino acid materials undergoing a "sol–gel" phase transition in a tumor micro-acid environment, and two polyamino acid materials, PGTTC and PGCLTGCL, were linked with TPA derivatives of AIE fluorescent molecules to obtain catheter-free embolic agents, PGTTC-DCDPP-2TPA and PGCLTGCL-DCDPP-2TPA. In tumor fluorescence imaging experiments in tumor-bearing mice, both were enriched at the tumor 72 h after injection and that of the PGCLTGCL-DCDPP-2TPA was stronger, moreover, the tumor cells were suppressed and necrotic after 25 days. Therefore, this pH-responsive noninterventional embolization combination therapeutic diagnostic system has good prospects for application in the field of combined therapeutic diagnostic multifunctional platforms. [ABSTRACT FROM AUTHOR]

• A predominance of pregnane was found in lacustrine rocks. • Pregnane/C 27 ααα20R sterane ratios correlated with bacterial activity. • The δ13C of pregnane reflects an origin from primary producers. • Unusual enrichment of pregnane indicates stable water column stratification. The distribution and carbon isotope composition of pregnane was determined in eight samples of carbonate-evaporitic rock collected from depths of 2938–3012 m in Well Luojia 14, Bonan Sag of Bohai Bay Basin, Eastern China, as a means of identification of the sources and to draw conclusions about possible formation processes and associated lake environments. The results in four samples from the 2989–3006 m depth range indicated abnormally high abundances of pregnane relative to long-chain (≥C 27) homologues, with ratios of pregnane to C 27 ααα20R sterane between 5.24 and 21.01 and averaging 13.91. Meanwhile, values of the gammacerane index (gammacerane/C 30 αβ hopane) and n -C 21− / n -C 22+ ratios were as high as 3.41–9.29 (mean 5.01) and 21.68–60.76 (mean 33.42) respectively, indicating the presence of stable oxic-anoxic stratification in the water column and blooms of heterotrophic bacteria. The δ13C values for pregnane were similar to those of primary producer-derived lipids, such as pristane and phytane, but lower by ∼8‰ than that of 4-methyl pregnane. This is consistent with the reports of 13C depletion in steranes (≥C 27) relative to their 4-methyl analogues. Taken together, these results suggest the abnormal enrichment of pregnane might result from severe bacterial alteration of algal-derived lipids in response to the stabilization of water column stratification forced by a salinity gradient. In four samples from within the 2938–2948 m band and at 3012 m, similar δ13C values for pregnane, pristane and phytane suggest an algal origin for pregnane. However, pregnane/C 27 ααα20R sterane ratios, gammacerane index values and n -C 21− / n -C 22+ ratios were all relatively low, with averages at 3.20, 0.43 and 4.60, respectively, suggesting much less bacterial influence on algal lipids and perhaps indicating relatively unstable stratification of the water column, possibly caused by a reduced input of freshwater. [ABSTRACT FROM AUTHOR]

• We found that zooplankton were significant contributors of sediment FAs. • C18:1 ω9 and short-chain n -alkanoic acids might be largely derived from zooplankton. • Zooplankton might be significant contributors of middle-chain n -alkanoic acids in some lakes. • Input of zooplankton FAs was greatly reduced due to the development of aquaculture. • Contribution of algal and bacterial FAs increased with lake eutrophication and aquaculture. During the past few decades, the majority of shallow lakes in the Yangtze River basin (Eastern China) have been impacted by heavy discharges of sewage and the intensification of aquaculture. These activities have significant potential to alter aquatic ecosystems and these changes may be reflected in the composition of lipid biomarkers in sediments at the time. In order to evaluate this effect, the abundances and distribution of fatty acids (FAs) were measured in 210Pb dated sediment cores from two anthropogenically impacted shallow lakes in the Yangtze River basin, Changdang and Taibai. Results from Lake Changdang revealed an extreme enrichment of zooplankton-derived C22:1 ω11 in sediments, with C22:1 ω11/ n -C16 ratios reaching ~1.6 on average. This unusual distribution may have been caused by zooplankton blooms in the lake in the years before aquaculture was developed. Elevated abundances of C18:1 ω9 and short-chain n -alkanoic acids observed in pre-1977 sediments might be also derived from zooplankton. After 1977, FAs of zooplankton origin in Lake Changdang greatly declined in abundance, coincident with the introduction of planktivorous fish. Meanwhile, inputs of FAs from submerged/floating macrophytes, bacteria and algae increased as a result of eutrophication and reduced grazing pressure from zooplankton. In Lake Taibai, abundances of C22:1 ω11 were greatly reduced in all sediments and accounted for 38% of n -C16 on average, reflecting comparatively smaller inputs of zooplankton FAs owing to the earlier establishment of aquaculture in the lake. Relative enrichment of polyunsaturated FAs (PUFAs) and branched FAs (Br-FAs) in the sediments was attributed to high inputs of algal and bacterial organic matter (OM) in response to reduced predation and grazing by zooplankton. Rapid increases in PUFAs and Br-FAs in post-1970 sediments were driven by eutrophication in this period. [ABSTRACT FROM AUTHOR]

• A predominance of C 28 steranes was found in lacustrine source rocks. • δ 13C of the C 28 ααα 20R sterane was as low as −37‰, similar to that of a C 27 20R triaromatic steroid. • C 28 steranes were produced by algae blooming in a mixed water column. • Other steranes derived from algae growing in a stratified water column. C 27 –C 29 steranes are common constituents of crude oils and hydrocarbon source rocks and are usually dominated by either C 27 or C 29 compounds. In the current study, however, steranes dominated by C 28 components were detected in eight lacustrine source rock samples in the first member of the Shahejie Formation in the Qikou Sag, Bohai Bay Basin of Eastern China. Ratios of C 28 /(C 27 + C 28 + C 29) ααα 20R steranes in four samples can exceeded 50%. Furthermore, C 27 triaromatic steroids, presumably originating from C 28 steranes, were also observed at high levels, relative to C 26 and C 28 homologues. Stable carbon isotope compositions of a C 28 ααα 20R sterane and C 27 20R triaromatic steroid were found to be similar, with δ13C at −37‰ and −36‰ respectively, much lower than values for pristane (by ∼7‰) and total organic carbon (by ∼10‰). The abnormally low δ13C values of the steroids suggest that C 28 ααα 20R sterane was biosynthesized by algae blooming in the water column during phases of overturn and mixing, such as in autumn, during which recycled CO 2 and nutrients welling up from hypolimnion to the epilimnion are assimilated, contributing to a high algal productivity and low δ13C in algal organic matter. [ABSTRACT FROM AUTHOR]

• Profiles of n -alkanes in sediments were dependent on water column structure, nutrient inputs and catchment disturbance. • In deep lakes, non-photosynthetic bacterial and aquatic macrophytes were major contributors of n -alkanes. • Terrestrial plant sourced n -alkanes were exceptionally abundant in alpine lakes. • Eutrophication driven enrichment of short-chain odd carbon n -alkanes was recorded in shallow lakes. In paleolimnology studies, the trajectories of n -alkane abundances and distributions in dated sediments are widely used indicators of changing environmental and ecological status in the overlying water over time. However, the factors contributing to the variable n -alkane profiles of lake systems are not yet fully understood. Here, a comparative study was conducted to examine natural and anthropogenic influences on the sediment n -alkane profiles of five representative lakes in China. Core samples from the deep oligo- mesotrophic lakes Fuxian and Lugu (FX-1 and LG-1 cores), the shallow eutrophic lakes Taibai and Changdang (TB-1 and CD-1 cores) and the alpine Lake Heihai (HH-1 core) were used. While there were striking differences in n -alkane abundances and distributions between the cores, the results suggested that the n -alkanes in all five studied lakes were derived mainly from biogenic sources, rather than from petroleum contamination. The short-chain (C 21 n -alkane profiles of different cores varied significantly. Abundances of long-chain n -alkanes (C 29 , C 31 and C 33) in HH-1 were approximately 3–7 times greater than in other cores. Coupled with the lowest Paq (average 0.29) values recorded in the study, this indicates a predominantly terrestrial plant origin for the >C 21 n -alkanes, most likely the dense alpine meadow which surrounds the lake. In other cores, however, aquatic macrophytes appear to be more important contributors of the n -alkanes, a result that corresponds with the relatively low density of terrestrial plants in those catchments. [ABSTRACT FROM AUTHOR]

Understanding the exact origin of sedimentary organic matter (OM) was crucial for assessing carbon cycle processes and environmental contamination in continental margin. However, such studies are less performed on shelves far away from the mainland. Here, the abundance and composition of lipid biomarkers were determined in surface sediments from 58 sites at the eastern Sunda Shelf, South China Sea, to identify the sedimentary OM sources and characterize the spatial variation of OM input from specific sources. The results indicated that the sedimentary OM were derived mainly from algae, bacteria, seagrass, terrestrial plants, mangroves, petroleum, and rock erosion products, and the relative significance of these sources varied between different sites. Algal OM input, as indicated by the abundances of short-chain n -alkanes, cholest-2-ene, dinosterol, brassicasterol, and C 30 1,15 alkyl diol, was relatively high at five sites (ZW 47, 51, 55, 82, and 86). Conversely, long-chain n -alkane and long-chain n -alkanol proxies suggested terrestrial plant input was most predominant at three other sites (ZW68, 103, and 106), and high abundances of taraxerol and β-amyrin in these sites imply the dominance of mangrove vegetation in the terrestrial OM. UCM (unresolved complex mixture) and αβ-hopanes indicated a relatively high input of petroleum-derived OM at another five sites (ZW3, 8, 25, 51, and 97). The analysis of abundance variations in the biomarkers suggested a substantial spatial heterogeneity in terrestrial plant OM input and a much lower heterogeneity in algal OM input, probably resulting from great variation in riverine transport intensity and relatively homogeneous water environment across the studied region. Correlation analysis indicated that both algal and terrestrial plant OM significantly contributed to the OM buried in the sediments. These results can form a basis for understanding the carbon cycle processes in Sunda Shelf. The finding of petroleum contamination in several sites can benefit for environmental protection and management of the shelf. • Sedimentary OM source and spatial variation in OM input were studied in eastern Sunda Shelf. • Both algae and terrestrial plants contributed significantly to the sediment OM. • Mangroves may be the dominant component of terrestrial OM in some sites. • Input of petroleum-derived OM was prominent at a few sites of the shelf. • Input of terrestrial plant OM had a much greater spatial heterogeneity than algal OM. [ABSTRACT FROM AUTHOR]

The development of social networks provides a broad platform for the dissemination of information and also leads to the proliferation of fake news and false information, which we collectively refer to as rumors. The spread of rumors causes unnecessary panic and loss to individuals and society. To reduce the negative impacts of rumors, an appropriate rumor control strategy is necessary. To come up with some reasonable strategies, we need to have a clearer understanding of the spread of rumors. In this paper, we analyze crowd attitudes during the spreading of rumors by setting the misinformation prevalent progress on the social network as a dynamic system. Considering that most people do not have a clear supportive or opposing attitude when exposed to rumor information, we introduce a new group, stiflers who remain neutral, based on the infectious disease model scheme. By deriving the mean-field equation describing the rumor propagation process, we judge the stability of the constructed model. Finally, we use the model to fit the real-world data related to COVID-19, and based on this, we discuss the properties of the model and propose related strategies. [ABSTRACT FROM AUTHOR]

Pediatric bone age assessment (BAA) is a common clinical practice to investigate endocrinology, genetic and growth disorders of children. Different specific bone parts are extracted as anatomical Regions of Interest (RoIs) during this task, since their morphological characters have important biological identification in skeletal maturity. Following this clinical prior knowledge, recently developed deep learning methods address BAA with an RoI-based attention mechanism, which segments or detects the discriminative RoIs for meticulous analysis. Great strides have been made, however, these methods strictly require large and precise RoIs annotations, which limits the real-world clinical value. To overcome the severe requirements on RoIs annotations, in this paper, we propose a novel self-supervised learning mechanism to effectively discover the informative RoIs without the need of extra knowledge and precise annotation—only image-level weak annotation is all we take. Our model, termed PEAR-Net for Part Extracting and Age Recognition Network, consists of one Part Extracting (PE) agent for discriminative RoIs discovering and one Age Recognition (AR) agent for age assessment. Without precise supervision, the PE agent is designed to discover and extract RoIs fully automatically. Then the proposed RoIs are fed into AR agent for feature learning and age recognition. Furthermore, we utilize the self-consistency of RoIs to optimize PE agent to understand the part relation and select the most useful RoIs. With this self-supervised design, the PE agent and AR agent can reinforce each other mutually. To the best of our knowledge, this is the first end-to-end bone age assessment method which can discover RoIs automatically with only image-level annotation. We conduct extensive experiments on the public RSNA 2017 dataset and achieve state-of-the-art performance with MAE 3.99 months. Project is available at http://imcc.ustc.edu.cn/project/ssambaa/. [ABSTRACT FROM AUTHOR]

Weakly supervised temporal sentence grounding has better scalability and practicability than fully supervised methods in real-world application scenarios. However, most of existing methods cannot model the fine-grained video-text local correspondences well and do not have effective supervision information for correspondence learning, thus yielding unsatisfying performance. To address the above issues, we propose an end-to-end Local Correspondence Network (LCNet) for weakly supervised temporal sentence grounding. The proposed LCNet enjoys several merits. First, we represent video and text features in a hierarchical manner to model the fine-grained video-text correspondences. Second, we design a self-supervised cycle-consistent loss as a learning guidance for video and text matching. To the best of our knowledge, this is the first work to fully explore the fine-grained correspondences between video and text for temporal sentence grounding by using self-supervised learning. Extensive experimental results on two benchmark datasets demonstrate that the proposed LCNet significantly outperforms existing weakly supervised methods. [ABSTRACT FROM AUTHOR]

AbstractTranscatheter arterial embolization (TAE) has been used to treat advanced and unresectable tumors, whose treatment is complex and may involve other risks. Here, inspired by the acid and reductive microenvironment of the solid tumors, a new approach for noninvasive and targeted intravenous embolization and chemotherapy is developed. The method was based on a series of acid microenvironment-responsive polypeptides poly[(L-Glutamic acid -ran-L-Tyrosine) -b-L-Threonine-b-L-Cysteines]s, PGTTCs.) to construct the reduction-responsive chemoembolic agent by grafting modified paclitaxel (HS-PTX) to sulfhydryl side groups on PGTTCs for noncatheter targeted-embolization, which effectively improved the solubility of paclitaxel. HS-PTX could release in the presence of glutathione (GSH) due to disulfide bond disruption in vitro, and the results of tumor-bearing mice experiments showed that these polypeptides could effectively target to the tumor, embolize the tumor supply vessels rapidly and release paclitaxel in tumor sites. By comparing the therapeutic effect of embolization therapy only with that of embolization combined with chemotherapy, the chemoembolization agent showed excellent effect of embolization chemotherapy in subcutaneous tumor models, and cured tumor in mice within 20 days. Therefore, this polypeptide is expected to be a favorable candidate for effective catheterless inteavenous embolization combined with chemotherapy. [ABSTRACT FROM AUTHOR]

Unsupervised person re-identification (Re-ID) has better scalability and practicability than supervised Re-ID in the actual deployment. However, it is difficult to learn a discriminative Re-ID model without annotations. To address the above issue, we propose an end-to-end Self-supervised Agent Learning (SAL) algorithm by exploiting a set of agents as a bridge to reduce domain gaps for unsupervised cross-domain person Re-ID. The proposed SAL model enjoys several merits. First, to the best of our knowledge, this is the first work to exploit self-supervised learning for unsupervised person Re-ID. Second, our model has designed three effective learning mechanisms including supervised label learning in source domain, similarity consistency learning in target domain, and self-supervised learning in cross domain, which can learn domain-invariant yet discriminative representations through the principled lens of agent learning by reducing domain discrepancy adaptively. Extensive experimental results on three standard benchmarks demonstrate that the proposed SAL performs favorably against state-of-the-art unsupervised person Re-ID methods. [ABSTRACT FROM AUTHOR]

Many existing scene parsing methods adopt Convolutional Neural Networks with receptive fields of fixed sizes and shapes, which frequently results in inconsistent predictions of large objects and invisibility of small objects. To tackle this issue, we propose perspective-adaptive convolutions to acquire receptive fields of flexible sizes and shapes during scene parsing. Through adding a new perspective regression layer, we can dynamically infer the position-adaptive perspective coefficient vectors utilized to reshape the convolutional patches. Consequently, the receptive fields can be adjusted automatically according to the various sizes and perspective deformations of the objects in scene images. Our proposed convolutions are differentiable to learn the convolutional parameters and perspective coefficients in an end-to-end way without any extra training supervision of object sizes. Furthermore, considering that the standard convolutions lack contextual information and spatial dependencies, we propose a context adaptive bias to capture both local and global contextual information through average pooling on the local feature patches and global feature maps, followed by flexible attentive summing to the convolutional results. The attentive weights are position-adaptive and context-aware, and can be learned through adding an additional context regression layer. Experiments on Cityscapes and ADE20K datasets well demonstrate the effectiveness of the proposed methods. [ABSTRACT FROM AUTHOR]

Retweeting is a powerful driving force in information propagation on microblogging sites. However, identifying the most effective retweeters of a message (called the ”key retweeter prediction” problem) has become a significant research topic. Conventional approaches have addressed this topic from two main aspects: by analyzing either the personal attributes of microblogging users or the structures of user graph networks. However, according to sociological findings, author-retweeter dependencies also play a crucial role in influencing message propagation. In this paper, we propose a novel model to solve the key retweeter prediction problem by incorporating the auxiliary relations between a tweet author and potential retweeters. Without loss of generality, we formulate the relations from four relational factors: status relation, temporal relation, locational relation, and interactive relation. In addition, we propose a novel method, called “Relation-based Learning to Rank (RL2R),” to determine the key retweeters for a given tweet by ranking the potential retweeters in terms of their spreadability. The experimental results show that our method outperforms the state-of-the-art algorithms at top-k retweeter prediction, achieving a significant relative average improvement of 19.7–29.4 percent. These findings provide new insights for understanding user behaviors on social media for key retweeter prediction purposes. [ABSTRACT FROM AUTHOR]