General description of Lake Peipsi-Pihkva

Neural networks and multiple linear regression models of the abundance of brown trout (Salmo trutta L.) on the mesohabitat scale were developed from combinations of physical habitat variables in 220 channel morphodynamic units (pools, riffles, runs, etc.) of 11 different streams in the central Pyrenean mountains. For all the 220 morphodynamic units, the determination coefficients obtained between the estimated and observed values of density or biomass were significantly higher for the neural network (r ² adjusted= 0.93 and r ² adjusted=0.92 (p<0.01) for biomass and density respectively with the neural network, against r ² adjusted=0.69 (p<0.01) and r ² adjusted = 0.54 (p<0.01) with multiple linear regression). Validation of the multivariate models and learning of the neural network developed from 165 randomly chosen channel morphodynamic units, was tested on the 55 other channel morphodynamic units. This showed that the biomass and density estimated by both methods were significantly related to the observed biomass and density. Determination coefficients were significantly higher for the neural network (r ² adjusted =0.72 (p<0.01) and 0.81 (p<0.01) for biomass and density respectively) than for the multiple regression model (r ² adjusted=0.59 and r ² adjusted=0.37 for biomass and density respectively). The present study shows the advantages of the backpropagation procedure with neural networks over multiple linear regression analysis, at least in the field of stochastic salmonid ecology.     

Cortical pattern formation during visual hallucinations

We theoretically investigate pattern formation during simple visual hallucinations caused by epileptic activity. To this end we analyze the activator-inhibitor model of Ermentrout and Cowan [1]. In contrast to these authors we focus on a different disease mechanism: According to experimental findings (cf. [2]) we decrease the influence of the inhibitor on the activator. This causes spontaneous pattern formation due to a bifurcation. The model parameters determine whether one or two or four modes become unstable. By means of the center manifold theorem, in all cases the order parameter equation is derived, the stability of the solution is proofed, and the bifurcating activity pattern is calculated explicitely in lowest order. Taking into account terms up to third order in all cases the order parameter equation has a potential. For the two-modes and the four-modes instability this potential causes a winner-takes all dynamics. We integrate the order parameter equation numerically and plot the visual hallucinations which result from the bifurcating cortical activity. The theoretically derived hallucinations correspond to clinically observed visual hallucinations (cf. [3, 4]), which are, for instance, well-known from petit mal epilepsy [5].Finally we investigate the influence of noise on the activity patterns as well as the visual hallucinations.     

基于视觉编码的图象处理研究──Ⅰ．原理、算法及实现

提出Ｇａｂｏｒ小波表达的概念及其在图象处理中实现的数学模型和算法框架,主要解决了运算中非正交性问题、收敛性问题,使得该算法及模型可以实现图象的分解和重建,并且可以应用于对视觉编码理论的解释。     

鄱阳湖生态经济区生态网络构建与评价

大型生态斑块是区域生物多样性保护的重要基础.然而,随着我国城镇化的快速推进,人类活动对环境的干扰越来越强烈,不仅大型生态斑块数量不断减少、质量不断下降,而且斑块之间连接出现弱化, “孤岛化”现象严重,严重影响区域生物多样性保护.本文以鄱阳湖生态经济区为例,通过GIS技术手段,充分考虑景观类型、坡度和人类活动干扰的影响,采用最小费用模型构建了该区域的潜在生态廊道.基于重力模型定量分析了生态斑块间的相互作用强度,并以此将潜在生态廊道分成2级以便区别保护,识别了研究区重要生态节点和生态断裂点,分析了该潜在生态网络结构的连通性.结果表明: 森林和农田是构成研究区生态廊道的主要景观类型;生态斑块之间的相互作用强度差异显著;区域生态网络结构较复杂,网络连接度和闭合度较高.研究结果可为鄱阳湖生态经济区生物多样性保护和生态网络优化提供科学依据.     

Modelling marine community responses to climate‐driven species redistribution to guide monitoring and adaptive ecosystem‐based management

As a consequence of global climate‐driven changes, marine ecosystems are experiencing polewards redistributions of species – or range shifts – across taxa and throughout latitudes worldwide. Research on these range shifts largely focuses on understanding and predicting changes in the distribution of individual species. The ecological effects of marine range shifts on ecosystem structure and functioning, as well as human coastal communities, can be large, yet remain difficult to anticipate and manage. Here, we use qualitative modelling of system feedback to understand the cumulative impacts of multiple species shifts in south‐eastern Australia, a global hotspot for ocean warming. We identify range‐shifting species that can induce trophic cascades and affect ecosystem dynamics and productivity, and evaluate the potential effectiveness of alternative management interventions to mitigate these impacts. Our results suggest that the negative ecological impacts of multiple simultaneous range shifts generally add up. Thus, implementing whole‐of‐ecosystem management strategies and regular monitoring of range‐shifting species of ecological concern are necessary to effectively intervene against undesirable consequences of marine range shifts at the regional scale. Our study illustrates how modelling system feedback with only limited qualitative information about ecosystem structure and range‐shifting species can predict ecological consequences of multiple co‐occurring range shifts, guide ecosystem‐based adaptation to climate change and help prioritise future research and monitoring.     

Genetic structure of native ant supercolonies varies in space and time

Ant supercolonies are the largest cooperative units known in nature. They consist of networks of interconnected nests with hundreds of reproductive queens, where individuals move freely between nests, cooperate across nest boundaries and show little aggression towards non‐nestmates. The combination of high queen numbers and free mixing of workers, queens and brood between nests results in extremely low nestmate relatedness. In such low‐relatedness societies, cooperative worker behaviour appears maladaptive because it may aid random individuals instead of relatives. Here, we provide a comprehensive picture of genetic substructure in supercolonies of the native wood ant Formica aquilonia using traditional population genetic as well as network analysis methods. Specifically, we test for spatial and temporal variation in genetic structure of different classes of individuals within supercolonies and analyse the role of worker movement in determining supercolony genetic networks. We find that relatedness within supercolonies is low but positive when viewed on a population level, which may be due to limited dispersal of individuals and/or ecological factors such as nest site limitation and competition against conspecifics. Genetic structure of supercolonies varied with both sample class and sampling time point, which indicates that mobility of individuals varies according to both caste and season and suggests that generalizing has to be carried out with caution in studies of supercolonial species. Overall, our analysis provides novel evidence that native wood ant supercolonies exhibit fine‐scale genetic substructure, which may explain the maintenance of cooperation in these low‐relatedness societies.     

An essential step of kinetochore formation controlled by the SNARE protein Snap29

The kinetochore is an essential structure that mediates accurate chromosome segregation in mitosis and meiosis. While many of the kinetochore components have been identified, the mechanisms of kinetochore assembly remain elusive. Here, we identify a novel role for Snap29, an unconventional SNARE, in promoting kinetochore assembly during mitosis in Drosophila and human cells. Snap29 localizes to the outer kinetochore and prevents chromosome mis‐segregation and the formation of cells with fragmented nuclei. Snap29 promotes accurate chromosome segregation by mediating the recruitment of Knl1 at the kinetochore and ensuring stable microtubule attachments. Correct Knl1 localization to kinetochore requires human or Drosophila Snap29, and is prevented by a Snap29 point mutant that blocks Snap29 release from SNARE fusion complexes. Such mutant causes ectopic Knl1 recruitment to trafficking compartments. We propose that part of the outer kinetochore is functionally similar to membrane fusion interfaces.