Halogenated aromatic pollutants in routine animal-derived food of south China: Occurrence, sources, and dietary intake risks.

Halogenated aromatic pollutants (HAPs) including polychlorinated dibenzo-p-dioxins/furans (PCDD/Fs), polychlorinated biphenyls (PCBs), polybrominated dibenzo-p-dioxins/furans (PBDD/Fs), and polybrominated diphenyl ethers (PBDEs) exhibit diverse toxicities and bio-accumulation in animals, thereby imposing risks on human via animal-derived food (ADF) consumption. Here we examined these HAPs in routine ADFs from South China and observed that PBDEs and PCBs showed statistically higher concentrations than PCDD/Fs and PBDD/Fs. PCDD/Fs and PCBs in these ADFs were mainly from the polluted feed and habitat of animals, except PCDD/Fs in egg, which additionally underwent selective biotransformation/progeny transfer after the maternal intake of PCDD/F-polluted stuff. PBDEs and PBDD/Fs were mostly derived from the extensive use of deca-BDE and their polluted environments. Significant interspecific differences were mainly observed for DL-PCBs and partly for PBDD/Fs and PBDEs, which might be caused by their distinct transferability/biodegradability in animals and the different living habit and habitat of animals. The dietary intake doses (DIDs) of these HAPs via ADF consumption were all highest for toddlers, then teenagers and adults. Milk, egg, and fish contributed most to the DIDs and risks for toddlers and teenagers, which results of several cities exceeded the recommended thresholds and illustrated noteworthy risks. Pork, fish, and egg were the top three risk contributors for adults, which carcinogenic and non-carcinogenic risks were both acceptable. Notably, PBDD/Fs showed the lowest concentrations but highest contributions to the total risks of these HAPs, thereby meriting continuous attention.

APANet: Auto-Path Aggregation for Future Instance Segmentation Prediction.

Despite the remarkable progress achieved in conventional instance segmentation, the problem of predicting instance segmentation results for unobserved future frames remains challenging due to the unobservability of future data. Existing methods mainly address this challenge by forecasting features of future frames. However, these methods always treat features of multiple levels (e.g., coarse-to-fine pyramid features) independently and do not exploit them collaboratively, which results in inaccurate prediction for future frames; and moreover, such a weakness can partially hinder self-adaption of a future segmentation prediction model for different input samples. To solve this problem, we propose an adaptive aggregation approach called Auto-Path Aggregation Network (APANet), where the spatio-temporal contextual information obtained in the features of each individual level is selectively aggregated using the developed "auto-path". The "auto-path" connects each pair of features extracted at different pyramid levels for task-specific hierarchical contextual information aggregation, which enables selective and adaptive aggregation of pyramid features in accordance with different videos/frames. Our APANet can be further optimized jointly with the Mask R-CNN head as a feature decoder and a Feature Pyramid Network (FPN) feature encoder, forming a joint learning system for future instance segmentation prediction. We experimentally show that the proposed method can achieve state-of-the-art performance on three video-based instance segmentation benchmarks for future instance segmentation prediction.

Fast Collective Activity Recognition Under Weak Supervision.

Collective activity recognition, which tells what activity a group of people is performing, is a cutting-edge research topic in computer vision. Different from action performed by individuals, collective activity needs to consider the complex interactions among different people. However, most previous works require exhaustive annotations such as accurate label information of individual actions, pairwise interactions, and poses, which could not be easily available in practice. Moreover, most of them treat human detection as a decoupled task before collective activity recognition and leverage all detected persons. This not only ignores the mutual relation between the two tasks, which makes it hard for filtering out irrelevant people, but also probably increases the computation burden when reasoning the collective activities. In this paper, we propose a fast weakly supervised deep learning architecture for collective activity recognition. For fast inference, we propose to make the actor detection and weakly supervised collective activity reasoning collaborate in an end-to-end framework by sharing convolutional layers between them. The joint learning makes the two tasks united and reinforced each other, so that it is more effective to filter out the outliers who are not involved in the activity. For the weakly supervised learning, we propose a latent embedding scheme for mining person-group interactive relationship to get rid of the use of any pairwise relation between people and the individual action labels as well. The experimental results show that the proposed framework achieves comparable or even better performance as compared to the state-of-the-art on three datasets. Our joint modelling reasons collective activities at the speed of 22.65 fps, which is the fastest ever known and substantially makes collective activity recognition more towards real-time applications.

Early Action Prediction by Soft Regression.

We propose a novel approach for predicting on-going action with the assistance of a low-cost depth camera. Our approach introduces a soft regression-based early prediction framework. In this framework, we estimate soft labels for the subsequences at different progress levels, jointly learned with an action predictor. Our formulation of soft regression framework 1) overcomes a usual assumption in existing early action prediction systems that the progress level of on-going sequence is given in the testing stage; and 2) presents a theoretical framework to better resolve the ambiguity and uncertainty of subsequences at early performing stage. The proposed soft regression framework is further enhanced in order to take the relationships among subsequences and the discrepancy of soft labels over different classes into consideration, so that a Multiple Soft labels Recurrent Neural Network (MSRNN) is finally developed. For real-time performance, we also introduce a new RGB-D feature called "local accumulative frame feature (LAFF)", which can be computed efficiently by constructing an integral feature map. Our experiments on three RGB-D benchmark datasets and an unconstrained RGB action set demonstrate that the proposed regression-based early action prediction model outperforms existing models significantly and also show that the early action prediction on RGB-D sequence is more accurate than that on RGB channel.

Global-Local Temporal Saliency Action Prediction.

Action prediction on a partially observed action sequence is a very challenging task. To address this challenge, we first design a global-local distance model, where a global-temporal distance compares subsequences as a whole and local-temporal distance focuses on individual segment. Our distance model introduces temporal saliency for each segment to adapt its contribution. Finally, a global-local temporal action prediction model is formulated in order to jointly learn and fuse these two types of distances. Such a prediction model is capable of recognizing action of: 1) an on-going sequence and 2) a sequence with arbitrarily frames missing between the beginning and end (known as gap-filling). Our proposed model is tested and compared with related action prediction models on BIT, UCF11, and HMDB data sets. The results demonstrated the effectiveness of our proposal. In particular, we showed the benefit of our proposed model on predicting unseen action types and the advantage on addressing the gapfilling problem as compared with recently developed action prediction models.

Brominated dioxins/furans and hydroxylated polybrominated diphenyl ethers: Occurrences in commercial 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE) and 2,4,6-tribromophenol, and formation during synthesis of BTBPE.

Polybrominated dibenzo-p-dioxins (PBDDs) and hydroxylated polybrominated diphenyl ethers (OH-PBDEs) can be formed from bromophenols (BPs) by thermal degradation, biosynthesis or phototransformation. However, it is unknown whether PBDDs and OH-PBDEs can be formed during the chemical production processes that utilize BPs as raw materials. 2,4,6-tribromophenol (2,4,6-TBP) is an important raw material for the synthesis of 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE), a novel brominated flame retardant. In this study, PBDDs, polybrominated dibenzofurans (PBDFs) and OH-PBDEs have been identified and quantified in commercially available BTBPE and 2,4,6-TBP. Furthermore, their formation as unintentional by-products during the laboratory synthesis of BTBPE from 2,4,6-TBP and 1,2-dibromoethane in the presence of sodium carbonate has also been investigated. 2,3,7,8-substituted PBDDs and PBDFs (2,3,7,8-PBDD/Fs) were undetectable in commercial samples of BTBPE and present in low levels (nanogram per gram) in 2,4,6-TBP. Two tetrabrominated dibenzo-p-dioxins (TeBDDs), namely 1,3,6,8- and 1,3,7,9-TeBDD, and three hydroxylated pentabrominated diphenyl ethers (OH-pentaBDEs), namely 4'-OH-BDE121, 2'-OH-BDE121, and 6'-OH-BDE100, were identified or tentatively identified, and quantitatively estimated to be at concentrations in the range of undetectable to several thousands of nanograms per gram in commercial BTBPE and 2,4,6-TBP. TeBDDs and OH-pentaBDEs were formed as by-products from 2,4,6-TBP during BTBPE synthesis. Further studies need to be conducted in order to determine whether PBDD/Fs and OH-PBDEs are also formed during the industrial synthesis of other chemical compounds that utilize BPs as raw materials or intermediates.

Jointly Learning Heterogeneous Features for RGB-D Activity Recognition.

In this paper, we focus on heterogeneous features learning for RGB-D activity recognition. We find that features from different channels (RGB, depth) could share some similar hidden structures, and then propose a joint learning model to simultaneously explore the shared and feature-specific components as an instance of heterogeneous multi-task learning. The proposed model formed in a unified framework is capable of: 1) jointly mining a set of subspaces with the same dimensionality to exploit latent shared features across different feature channels, 2) meanwhile, quantifying the shared and feature-specific components of features in the subspaces, and 3) transferring feature-specific intermediate transforms (i-transforms) for learning fusion of heterogeneous features across datasets. To efficiently train the joint model, a three-step iterative optimization algorithm is proposed, followed by a simple inference model. Extensive experimental results on four activity datasets have demonstrated the efficacy of the proposed method. A new RGB-D activity dataset focusing on human-object interaction is further contributed, which presents more challenges for RGB-D activity benchmarking.

HPV E6/p53 mediated down-regulation of miR-34a inhibits Warburg effect through targeting LDHA in cervical cancer.

MicroRNAs (miRNA) play crucial roles in regulating cell proliferation, differentiation and developmental timing. Aberrantly expressed miRNAs have recently emerged as key regulators of metabolism. However, little is known about its role in tumor metabolism of cervical cancer. In this study, we determined the oncogenic effects of miRNAs on Warburg effect, a metabolic phenotype that allows cancer cells to utilize glucose even under aerobic conditions. A gain-of-function study was performed in 12 down-regulated miRNAs that frequently reported in cervical cancer. We found that miR-34a plays a suppressive role in Warburg effect as evidenced by decreased lactate production and glucose consumption. Knockdown of oncoprotein E6 expression of human papillomavirus in SiHa and HeLa cells by siRNAs lead to an increased protein level of p53, decreased level of miR-34a, as well as reduced Warburg effect. Subsequently, lactate dehydrogenase A (LDHA), which catalyzes the last key step in glycolysis, was identified as a direct target of miR-34a. Silencing of LDHA or introduction of miR-34a significantly attenuated colony formation ability and invasive capacity of SiHa and HeLa cells, and these effects were fully compromised by reintroduction of LDHA. In conclusion, our findings demonstrated that deregulated miR-34a/LDHA axis induced by HPV E6/p53 signaling facilitates tumor growth and invasion through regulating Warburg effect in cervical cancer, and provided new insights into the mechanism by which miR-34a contributes to the development and progression of cervical cancer.

[Effect of early intervention of liver-smoothing and blood-activating decoction combined with acupuncture on patients with post-stroke depression].

To study the effect of early intervention of liver-soothing and Blood-activating decoction combined with acupuncture in improving neurological functions, depressive symptom and life quality of patients with post-stroke depression, and compare with fluoxetine hydrochloride. Specifically, 63 patients with post-stroke depression were randomly divided into the traditional Chinese medicine (TCM) acupuncture group (31 cases) and the western medicine group (32 cases). On the basis of the conventional treatment of the primary disease, the TCM acupuncture group was treated with liver-soothing and blood-activating decoction and acupuncture, while the western medicine group was treated with fluoxetine hydrochloride for four weeks. In the follow-up visit six months later, scores of HAMD, NIHSS and SS-QOL were observed. The scores of HAMD and NIHSS of both groups were significantly decreased (P < 0.01), while the scores of SS-QOL increased significantly, with a notable difference compared with that before the treatment (P < 0.01). Specifically, the TCM acupuncture group's was superior to the western medicine group (P < 0.05). The study suggests that the early intervention of liver-soothing and blood-activating decoction combined with acupuncture on patients with post-stroke depression has the effect in relieving depression symptom and improving neurological functions, thereby improving their quality of life and prognosis.

[Eco-environmental evolution inferred from n-alkanes and delta13C records in the sediments of Shijiu Lake].

The study of global changes has focused on the reconstruction of paleovegetation and paleoclimate by n-alkanes and delta13C. 210Pb contents were measured for dating. The distribution characteristics of n-alkanes and delta13C were used to indicate the source of the organic matter in the sediments of Shijiu Lake. The relationship between modern eco-environmental evolution and human behaviors was discussed in this paper. The combination characteristics of n-alkanes showed a significant odd-even predominance in high-carbon number and main peak at C29, suggesting that the organic matter in the sediments were mainly derived from macrophytes and terrestrial higher plants. The delta13C contents of C27, C29 and C31 n-alkanes were analyzed. Results indicated that C3 plants are the dominant species. The distribution characteristics of n-alkanes and delta13C in different periods revealed the impact of human behaviors on Shijiu Lake. From 1862 to 1970, the low relative content of TOC, TN, C17-C25 and the light delta13C25-31 values showed that there were less human behaviors effects on Shijiu Lake and the eco-environment around the lake was stable. From 1970 to 1983, the relative content of TOC, TN and C17-C25 increased significantly, the delta13C25-31 values became weight. In this period, large areas of Shijiu Lake turned into farmland; pollution by fertilizers and pesticides was serious; large amounts of industrial and domestic wastewater were discharged into the lake. All these human behaviors resulted in the degradation of terrestrial higher plants around the lake. Meanwhile, the eutrophication levels were significantly increased. From 1983 to 2010, the relative contents of TOC, TN and C17-C25 were still in high-value ranges, the problem of eutrophication was not effectively controlled and the eco-environment of Shijiu Lake was relatively degradated.

A confocal technique applicable to studies of cellular pH-related signaling in plants.

pH may act as a crucial signal in both animal and plant cells. It is very difficult to monitor pH signals and this has largely hindered progress in the investigation of pH signaling, particularly systematic pH signaling. Here, we report the development of a confocal technique to monitor leaf apoplastic pH in intact plants, which is particularly suitable for the studies on root to shoot signaling. A variety of different pH indicators and plant species were tested. It was found that different pH indicators, for example, 2',7'-Bis-(2-carboxyethyl)-5-(and-6)-carboxyfluoresce (BCECF), SNARF-4F 5-(and-6)-carboxylic acid (SNARF) and DM-NERF (NERF), were of different properties, and to successfully monitor pH at a sub-cellular level, the comparability between the pH indicator and plant species must be involved according to their suitable pH range and loading characteristics. The loading characteristics of different pH indicators differ with different plant species, cell types and their developing stages. No matter what methods were adopted, BCECF and SNARF could not be loaded specifically in the leaf apoplast in sunflower, tomato, and Comelina communis L. In contrast, regardless of the methods adopted, NERF could be loaded efficiently and specifically in the leaf apoplast in C. communis, but not in other plants. In C. communis, the determination coefficient for in vitro and in situ calibration of NERF was very high, which was respectively 0.9951 and 0.9916, and therefore, the adoption of NERF together with C. communis could construct an ideal experimental system that is suitable for the investigation of pH systematic signaling. Ratio image analysis demonstrated that the leaf apoplastic pH was about 5.5 in non-stressed conditions, and water deficit could trigger an increase in pH by about half a pH unit, which is the first evidence to directly indicate that pH is able to act as a systematic signal under water deficit conditions.