科研成果

Floristic regions reflect the geographic organization of floras and provide essential tools for biological studies. Previous global floristic regions are generally based on floristic endemism, lacking a phylogenetic consideration that captures floristic evolution. Moreover, the contribution of tectonic dynamics and historical and current climate to the division of floristic regions remains unknown. Here, by integrating global distributions and a phylogeny of 12,664 angiosperm genera, we update global floristic regions and explore their temporal changes. Eight floristic realms and 16 nested sub-realms are identified. The previously-defined Holarctic, Neotropical and Australian realms are recognized, but Paleotropical, Antarctic and Cape realms are not. Most realms have formed since Paleogene. Geographic isolation induced by plate tectonics dominates the formation of floristic realms, while current/historical climate has little contribution. Our study demonstrates the necessity of integrating distributions and phylogenies in regionalizing floristic realms and the interplay of macroevolutionary and paleogeographic processes in shaping regional floras.

The wastewater treatment plant (WWTP) effluent discharge affects the microorganisms in the receiving water bodies. Despite the ecological significance of microbial communities in pollutant degradation and element cycling, how the community diversity is affected by effluent remains obscure. Here, we compared the sediment bacterial communities exposed to different intensities of WWTP effluent discharge in Hangzhou Bay, China: i) a severely polluted area that receives effluent from an industrial WWTP, ii) a moderately polluted area that receives effluent from a municipal WWTP, and iii) less affected area that inner the bay. We found that the sediment bacterial diversity decreased dramatically with pollution levels of inorganic nutrients, heavy metals, and organic halogens. Microbial community assembly model analysis revealed increased environmental selection and decreased species migration rate in the severely polluted area, resulting in high phylogenetic clustering of the bacterial communities. The ecological networks were less complex in the two WWTP effluent receiving areas than in the inner bay area, as suggested by the smaller network size and lower modularity. Fewer negative network associations were detected in the severely (6.7%) and moderately (8.3%) polluted areas than in the less affected area (16.7%), indicating more collaborative inter-species behaviors are required under stressful environmental conditions. Overall, our results reveal the fundamental impacts of WWTP effluents on the ecological processes shaping coastal microbial communities and point to the potential adverse effects of diversity loss on ecosystem functions.

Sign language recognition is of great significance to connect the hearing/speech impaired and non-sign language communities. Compared to isolated word recognition, sentence recognition is more practical in real-world scenarios, but is also more complicated because continuous, high-quality sign data with distinct features must be collected and isolated signs must be identified with high accuracy. Here, we propose a wearable sign language recognition system enabled by a convolutional neural network (CNN) that integrates stretchable strain sensors and inertial measurement units attached to the body to perceive hand postures and movement trajectories. Forty-eight Chinese sign language words commonly used in daily life were collected and used to train the CNN model, and an isolated sign language word recognition accuracy of 95.85% was achieved. For sentence-level sign language recognition, we proposed a method that combines multiple sliding windows and uses correlation analysis to improve the CNN recognition performance, achieving a correct rate of 84% for 50 sign language sentence samples, showing good extendibility.

Purpose The study aims to investigate the relationship between chief executive officers' (CEOs') prenatal testosterone exposure, absorptive capacity and e-commerce adoption by small and medium-sized enterprises (SMEs). Design/methodology/approach Based on a study of 1,519 SMEs led by a male CEO in China, the impact of entrepreneur's digital ratio on e-commerce adoption is empirically analyzed through a multivariate logistic model. Findings The results show that the second-to-fourth digit ratio (2D:4D), a reverse measurement of prenatal testosterone exposure, is negatively correlated with the adoption of e-commerce by SMEs. This evidence suggests that CEOs with high prenatal testosterone exposure have a higher probability of adopting e-commerce. In addition, this research indicates a positive correlation between absorptive capacity, which is defined by market innovation, process innovation and marketing innovation, and the adoption of e-commerce by SMEs. Originality/value This research can contribute to the discussion by providing new insights into the antecedents of the adoption of e-commerce in SMEs.

In China, rangeland fragmentation leads to the problems of anticommons, in terms of livestock production and ecological conditions. As the countermeasure, rangeland use right transfer has been encouraged by the governments recently, which aims to integrate the fragmented rangelands by lease. Can transfer overcome the problems of anticommons? We addressed this question through a case study in Inner Mongolia, by comparing livelihoods and ecological conditions between the households with lease-in pastures and those without practicing transfer. We found that though transfer could make the livelihoods of lease-in households with larger rangeland better-off in weather good years, but worse-off in drought years; and the over grazing was intensified on the transferred pastures. We concluded that the transfer may not be able to fundamentally overcome the problems of anticommons. We argued that spatial anticommons and right anticommons are interrelated to each other, rather than two juxtaposed types as defined by anticommons scholars.