Search

Begonia ferox C.I Peng & Yan Liu (2013) was rated as endangered according to Red List of Chinese Plants. In this study, we report the complete chloroplast genome of B. ferox. The chloroplast genome is 169,114 bp in length as the circular, with the GC content of 35.5%, composed by a large single-copy (LSC) region of 75,887 bp, a small single-copy (SSC) region of 18,105 bp, and two inverted repeat regions (IRs) of 37,561 bp in each. The genome comprises 174 encoded genes in total, including 114 protein-coding genes, eight ribosomal RNA genes, and 52 transfer RNA genes. Phylogenetic analysis indicated that B. ferox is genetically closest to B. gulongshanensis.

Abstract With the intensification of global warming, uncertainty regarding ecosystem evolution trends in drylands has increased. The objective of this article was to test whether the desertification process of tamarisk shrubland was intensified by comparing fertile island effects and soil ecological stoichiometry characteristics at different time periods. The tamarisk shrubland in the “shrubland” and “shrub duneland” stages in the northern Tarim Basin was selected, and soil samples were collected from the tamarisk nabkhas and from the interspaces between tamarisk nebkhas at the 0‐ to 30‐cm soil layers in 2009 and 2018. Soil carbon (C), nitrogen (N), and phosphorous (P) were analyzed, and the C/N, C/P, and N/P ratios were calculated. In comparison with 2009, fertile island effects were completely collapsed in the shrub duneland stage, with no obvious changes in the shrubland stage in 2018. The average soil C/N ratio decreased from a value greater than 25 in both stages. The average value of soil C/P ratio was below 200 in the shrubland and shrub duneland stages, with almost no marked change after nine years. The soil N/P ratio in our study significantly increased in 2018 than in 2009, but the soil N/P ratio was still much less than 11.9. Significantly positive correlations were observed between the relative interaction index (RII) of soil organic carbon (SOC) and soil total nitrogen (TN) and the soil C/P and N/P ratios in the tamarisk nebkhas. These results suggested that (1) fertile island effects were positive in relation to soil stoichiometry in tamarisk nebkhas in most cases; (2) the desertification process was accelerated in the shrub duneland stage, with no obvious progress during the shrubland stage; and (3) the loss of soil C from the collapsed fertile islands could therefore accelerate global warming.

Agricultural land use is an important source of CO2 emissions. Therefore, taking the CO2 emissions of China’s agricultural land use during 1995–2020 as a case, we firstly calculated its composition and analyzed the spatiotemporal evolution characteristics. Then, the Tapio decoupling model and logarithmic mean Divisia index (LMDI) were, respectively, used to identify the decoupling relationship between the CO2 emission change and economic growth, and analyze the driving factors for CO2 emissions. (1) The CO2 emissions of China’s agricultural land use were composed of two main phases (fluctuating growth phase (1995–2015) and rapid decline phase (2016–2020)). The total CO2 emissions exhibited a non-equilibrium spatial distribution. The inter-provincial CO2 emissions differences first expanded and then shrank, but the inter-provincial differences of CO2 emissions intensity continuously decreased. (2) The total CO2 emissions of China’s agricultural land use increased from 50.443 Mt in 1995 to 79.187 Mt in 2020, with an average annual growth rate of 1.82%. Fertilizer, agricultural diesel and agricultural (plastic) film were the main sources of anthropogenic agricultural-land-use CO2 emissions. Controlling the use of fertilizer and agricultural diesel and improving the utilization efficiency of agricultural (plastic) film could be an effective way to reduce CO2 emissions. (3) The Tapio decoupling relationship between the CO2 emission change and economic growth was a weak decoupling state during 1995–2015 and a strong decoupling state during 2016–2020. This result indicates that China’s agricultural land use can be effectively controlled. (4) The agricultural economic level is the decisive factor in promoting CO2 emissions increase, and its cumulative contribution was 476.09%. Inversely, the CO2 emission intensity, agricultural structure and agricultural labor force were three key factors, with cumulative contributions of −189.51%, −16.86% and −169.72%, respectively. Collectively, based on the findings obtained from the present research, we have proposed some suggestions to promote the sustainable use of agriculture lands in China.

Low-dimensional metal oxides-based electronic noses have been applied in various fields, such as food quality, environmental assessment, coal mine risk prediction, and disease diagnosis. However, the applications of these electronic noses are limited for conditions such as precise safety monitoring because electronic nose systems have problems such as poor recognition ability of mixed gas signals and sensor drift caused by environmental factors. Advanced algorithms, including classical gas recognition algorithms and neural network-based algorithms, can be good solutions for the key problems. Classical gas recognition methods, such as support vector machines, have been widely applied in electronic nose systems in the past. These methods can provide satisfactory results if the features are selected properly and the types of mixed gas are under five. In many situations, this can be challenging due to the drift of sensor signals. In recent years, neural networks have undergone revolutionary changes in the field of electronic noses, especially convolutional neural networks and recurrent neural networks. This paper reviews the principles and performances of typical gas recognition methods of the electronic nose up to now and compares and analyzes the classical gas recognition methods and the neural network-based gas recognition methods. This work can provide guidance for research in related fields.

Abstract Background Electronic health records (EHRs) provide possibilities to improve patient care and facilitate clinical research. However, there are many challenges faced by the applications of EHRs, such as temporality, high dimensionality, sparseness, noise, random error and systematic bias. In particular, temporal information is difficult to effectively use by traditional machine learning methods while the sequential information of EHRs is very useful. Method In this paper, we propose a general-purpose patient representation learning approach to summarize sequential EHRs. Specifically, a recurrent neural network based denoising autoencoder (RNN-DAE) is employed to encode inhospital records of each patient into a low dimensional dense vector. Results Based on EHR data collected from Shuguang Hospital affiliated to Shanghai University of Traditional Chinese Medicine, we experimentally evaluate our proposed RNN-DAE method on both mortality prediction task and comorbidity prediction task. Extensive experimental results show that our proposed RNN-DAE method outperforms existing methods. In addition, we apply the “Deep Feature” represented by our proposed RNN-DAE method to track similar patients with t-SNE, which also achieves some interesting observations. Conclusion We propose an effective unsupervised RNN-DAE method to summarize patient sequential information in EHR data. Our proposed RNN-DAE method is useful on both mortality prediction task and comorbidity prediction task.

Named entities composed of multiple continuous words frequently occur in domain-specific knowledge graphs. In general, these named entities are composable and extensible, such as names of symptoms and diseases in the medical domain. Unlike the general entities, we address them as compound entities, and try to identify hypernymy relations between them. Hypernymy detection between compound entities plays a critical role in domain-specific knowledge graph construction. Traditional hypernymy detection approaches do not perform well on compound entities for two reasons. One is the lack of contextual information, and the other is the absence of compound entities, i.e. out-of-vocabulary (OOV) problem. In this paper, we propose a hybrid-attention-based method called Bi-GRU-CapsNet for the detection of hypernymy relations. The hybrid attention mechanism consists of heuristic attention and self-adaptive attention, which are used for the lack of contextual information. The attentions focus on the differences of two compound entities on the lexical and semantic level, respectively. For OOV problem, the English words or Chinese characters in compound entities are fed into bidirectional gated recurrent units (Bi-GRUs). Additionally, we use capsule network (CapsNet) to determine the existence of hypernymy relations under different cases. Experimental results show that our proposed method outperforms other baseline methods on both English and Chinese corpora of symptom and disease pairs.

The existing watermarking algorithms in H.265/high efficiency video coding (HEVC) compressed domain cannot achieve intra-frame distortion drift-free and resist the recompression and signal processing attacks simultaneously. Therefore, the application of HEVC watermarking in copyright protection is limited in practice. To solve the problem, this paper proposes an intra-drift-free robust watermarking algorithm, which uses a multi-coefficients modification method to embed the watermark into intra prediction residual pixels of 4 × 4 luminance transform blocks in the spatial domain. Based on the spatial-domain embedding method, watermark information can be extracted from the blocks that have been attacked and merged into 8 × 8 size, which improves the robustness. To further improve the performance of the watermarking algorithm, the proposed algorithm uses a combination of three thresholds according to the application requirements. Experimental results on a publicly available video database (JCT-VC) have justified the effectiveness of the proposed algorithm which has better imperceptibility and robustness than existing algorithms. In addition, the proposed algorithm has high real-time performance and is suitable for application scenarios where the video is distributed on a large scale.

Residential energy consumption (REC) is now the second largest energy sector in China. However, decoupling analysis of REC has not received enough attention. Here, we explore the decoupling relationship between REC and economic growth in urban and rural Guangdong from 2000 to 2017. First, we use the Tapio decoupling model to study the decoupling state. Then, key drivers affecting the decoupling were explored by combining the Tapio decoupling model with the Log-mean Divisa Index (LMDI) method. Finally, the decoupling efforts were evaluated by using the decoupling efforts model. The main results are as follows: (1) The decoupling state in urban regions was better than that in rural regions. Urban regions experienced weak decoupling (WD) in most years during 2000–2017, while expansive negative decoupling (END) mostly was seen in rural regions. (2) The per capita income effect was the most important driver inhibiting decoupling in Guangdong urban and rural regions. The energy intensity effect was the biggest driver promoting decoupling in urban Guangdong, followed by the family size effect. In contrast, the family size effect exerted the dominant influence on accelerating the decoupling in rural Guangdong. Overall, the household effect inhibited decoupling in urban regions, but promoted decoupling in rural regions. (3) The decoupling efforts results suggest that the energy efficiency improvement played a major role in the decoupling efforts of urban Guangdong, while the decrease of family size in rural Guangdong was the leading contributor to the decoupling efforts.