Pattern of cooling benefits from ecospaces during urbanization: A case study of the Yangtze River Economic Belt.

Urban ecological spaces are effective thermoregulators under global warming. However, the cooling efficiency of urban ecological spaces during the urbanization has not been studied comprehensively. Here, we investigate the spatio-temporal dynamics of Urban Cold Island (UCI) intensity in 11 typical cities of the Yangtze River Economic Belt (YREB). We determined the impact of ecological landscape trends on these dynamics by using GlobalLand and MODIS 8 d mean land surface temperature (LST) data for three periods (2000, 2010, and 2020), and the landscape pattern index and diversity index. We found that in the past 20 years, the built-up area has increased by sixfold; 62.53 % and 37.47 % of YREB were warming or cooling, with 71.22 % of the daytime cooling and 93 % of the nighttime warming. The average UCI intensity of YREB has increased from 0.518 to 0.847 and is negatively correlated with LST with a decreasing slope. As the UCI intensity of green spaces increased, that of blue spaces decreased. Surface area and landscape pattern are the key determinants of UCI intensity in blue and green spaces, respectively, especially the landscape shape index (LSI). Therefore, maintaining ecological spaces, enriching the structural integrity of green spaces, and improving blue space connectivity can help cities at different development levels cope with heat stress during regional urbanization.

Deep asymmetric extraction and aggregation for infrared small target detection.

Infrared small target detection is widely applied in military and civilian fields. Due to the small size of infrared targets, textural detail is missing. Common target detection methods extract semantic feature by narrowing down the feature map several times, which may lead to the small targets lost in deep layers and are not effective for infrared small target detection. To solve this problem, we propose a novel network called deep asymmetric extraction and aggregation. The network mainly consists of two processes - the vertical feature extraction and the horizontal feature aggregation, both of which are enhanced by an asymmetric attention mechanism. In the vertical process, the use of asymmetric attention mechanism combined with the reduction of down-sampling makes the small target better retained in the deep layers. Then through the horizontal process, shallow spatial feature and deep semantic feature are aggregated to further highlight the small targets while suppressing background noise. Experiments on the public datasets NUAA-SISRT, NUDT-SISRT and MDvsFA-cGan show that our proposed network outperforms the state-of-the-art methods in terms of detection accuracy and parameter efficiency.

Widespread colonization of the lung by Tropheryma whipplei in HIV infection.

RATIONALE: Lung infections caused by opportunistic or virulent pathogens are a principal cause of morbidity and mortality in HIV infection. It is unknown whether HIV infection leads to changes in basal lung microflora, which may contribute to chronic pulmonary complications that increasingly are being recognized in individuals infected with HIV. OBJECTIVES: To determine whether the immunodeficiency associated with HIV infection resulted in alteration of the lung microbiota. METHODS: We used 16S ribosomal RNA targeted pyrosequencing and shotgun metagenomic sequencing to analyze bacterial gene sequences in bronchoalveolar lavage (BAL) and mouths of 82 HIV-positive and 77 HIV-negative subjects. MEASUREMENTS AND MAIN RESULTS: Sequences representing Tropheryma whipplei, the etiologic agent of Whipple's disease, were significantly more frequent in BAL of HIV-positive compared with HIV-negative individuals. T. whipplei dominated the community (>50% of sequence reads) in 11 HIV-positive subjects, but only 1 HIV-negative individual (13.4 versus 1.3%; P = 0.0018). In 30 HIV-positive individuals sampled longitudinally, antiretroviral therapy resulted in a significantly reduced relative abundance of T. whipplei in the lung. Shotgun metagenomic sequencing was performed on eight BAL samples dominated by T. whipplei 16S ribosomal RNA. Whole genome assembly of pooled reads showed that uncultured lung-derived T. whipplei had similar gene content to two isolates obtained from subjects with Whipple's disease. CONCLUSIONS: Asymptomatic subjects with HIV infection have unexpected colonization of the lung by T. whipplei, which is reduced by effective antiretroviral therapy and merits further study for a potential pathogenic role in chronic pulmonary complications of HIV infection.

Graph accordance of next-generation sequence assemblies.

MOTIVATION: No individual assembly algorithm addresses all the known limitations of assembling short-length sequences. Overall reduced sequence contig length is the major problem that challenges the usage of these assemblies. We describe an algorithm to take advantages of different assembly algorithms or sequencing platforms to improve the quality of next-generation sequence (NGS) assemblies. RESULTS: The algorithm is implemented as a graph accordance assembly (GAA) program. The algorithm constructs an accordance graph to capture the mapping information between the target and query assemblies. Based on the accordance graph, the contigs or scaffolds of the target assembly can be extended, merged or bridged together. Extra constraints, including gap sizes, mate pairs, scaffold order and orientation, are explored to enforce those accordance operations in the correct context. We applied GAA to various chicken NGS assemblies and the results demonstrate improved contiguity statistics and higher genome and gene coverage. AVAILABILITY: GAA is implemented in OO perl and is available here: http://sourceforge.net/projects/gaa-wugi/. CONTACT: lye@genome.wustl.edu