应用近红外光谱法估测小麦叶片糖氮比

糖氮比能够反映作物碳氮代谢的协调程度,及时、准确地监测糖氮比对于作物氮素营养诊断和调控具有重要意义.本研究以不同年份、品种、施氮水平的小麦大田试验为基础,获取鲜叶和粉末状干叶近红外(NIR)光谱及糖氮比信息,分别运用偏最小二乘法(partial least squares, PLS)、BP神经网络(back propagation neural network, BPNN)和小波神经网络(wavelet neural network, WNN)3种方法建立了小麦叶片糖氮比预测模型,并利用随机选择的样品集对所建模型进行测试和检验.结果表明： 小麦鲜叶光谱模型预测性能不佳;而干叶片预测模型表现了较好的准确性,在1655~2378 nm谱区范围内基于3种方法构建的干叶粉末糖氮比估算模型,其预测均方根误差均低于0.3%,决定系数均高于0.9.比较而言,WNN法表现最佳.总体显示,近红外光谱法可以准确预测小麦叶片糖氮比状况,为科学诊断糖氮比提供了理论基础和技术途径.     

基于BP神经网络的流域生态恢复度计算——以福建长汀朱溪小流域为例

以福建省长汀县朱溪小流域为研究对象,通过野外调查、室内分析以及遥感影像提取相结合的方法获取数据。利用Matlab7.0软件建立BP神经网络生态恢复模型,定量评价退化生态系统的恢复程度。选择土壤理化性质(有机质、全N、全P、全K、容重和p H)、植被结构(植被盖度)、物种多样性指数(Shannon-Wiener指数)和热环境(地表温度)等4个方面的9个指标建立退化生态系统评价体系,并作为生态恢复模型的输入层数据,生态恢复度作为输出层数据。使用Matlab7.0进行数据预处理、样本训练、样本检验并建立生态恢复模型。利用建立的生态恢复模型对整个朱溪小流域生态恢复度进行定量评价。结果表明,生态恢复模型预测结果与流域生态恢复的实际情况基本吻合,利用BP神经网络模型定量评价退化生态系统的恢复程度具有可行性。朱溪小流域内生态恢复程度极低的区域面积仅占0.94%,95.48%区域为中等恢复程度,说明生态保护措施已初见成效;生态恢复程度高的区域面积仅占3.62%,意味着未来仍需加强治理和保护工作。     

蛋白质相互作用数据库及其应用

对蛋白质相互作用及其网络的了解不仅有助于深入理解生命活动的本质和疾病发生的机制,而且可以为药物研发提供靶点.目前,通过高通量筛选、计算方法预测和文献挖掘等方法,获得了大批量的蛋白质相互作用数据,并由此构建了很多内容丰富并日益更新的蛋白质相互作用数据库.本文首先简要阐述了大规模蛋白质相互作用数据产生的3种方法,然后重点介绍了几个人类相关的蛋白质相互作用公共数据库,包括HPRD、BIND、 IntAct、MINT、 DIP 和MIPS,并概述了蛋白质相互作用数据库的整合情况以及这些数据库在蛋白质相互作用网络构建上的应用.     

Intercompany Energy Integration

Reusing heat through process integration in heat exchanger networks has long been a key measure for increasing energy efficiency in energy‐intensive industries. Thermal pinch analysis is commonly used for a systematic matching of process streams and thus planning of optimal process integration in large chemical plants. The possible savings increase with the amount of heat and the number of integrated process streams. Therefore co‐ siting of several companies in a symbiotic network opens new opportunities for process integration even in small and medium‐size enterprises (SMEs), but also introduces new challenges. Thermal pinch analysis is extended here to account for piping distances and fluctuations and limited availability of energy flows by adding additional costs for the piping system and a backup utility system in the optimization function. Cooperative game theory is proposed to derive a sharing of savings between the partners of the industrial symbiosis that is optimal for each partner and should prevent partners from leaving the network because of higher benefits in a subgroup or alone. It is argued that knowledge about the optimality of a network for each partner creates trust between the partners that is a necessary base for the long‐term commitment needed in industrial symbioses. An exemplary symbiotic network combining the production of pulp and woody biomass energy carriers is used to illustrate the proposed approaches.     

No Evidence‐Based Restoration Without a Sound Evidence Base: A Reply to Guldemond et al.

Evidence‐based practice is not possible without an evidence base. Guldemond et al. confuse our attempt at assessing the status of the evidence base of restoration programs in South Africa with attempting to assess whether restoration is evidence‐based. While we fully agree with them that there is a need to assess whether practitioners use evidence in their decision‐making, we assert that use of evidence is the last step in the evidence‐based approach. It is preceded by the generation (and documentation) of evidence through baseline condition assessment, proper goal setting, sound monitoring of the impacts of the chosen intervention as well as effective dissemination of resulting evidence. To answer the question whether restoration is evidence‐based would require the assessment of all stages from generation to use. We chose to start at the beginning, a logical place to start.     

Limited influence of local and landscape factors on finescale gene flow in two pond‐breeding amphibians

Dispersal and gene flow within animal populations are influenced by the composition and configuration of the landscape. In this study, we evaluated hypotheses about the impact of natural and anthropogenic factors on genetic differentiation in two amphibian species, the spotted salamander (Ambystoma maculatum) and the wood frog (Lithobates sylvaticus) in a commercial forest in central Maine. We conducted this analysis at two scales: a local level, focused on factors measured at each breeding pond, and a landscape level, focused on factors measured between ponds. We investigated the effects of a number of environmental factors in six categories including Productivity, Physical, Land Composition, Land Configuration, Isolation and Location. Embryos were sampled from 56 spotted salamander breeding ponds and 39 wood frog breeding ponds. We used a hierarchical Bayesian approach in the program GESTE at each breeding pond and a random forest algorithm in conjunction with a network analysis between the ponds. We found overall high genetic connectivity across distances up to 17 km for both species and a limited effect of natural and anthropogenic factors on gene flow. We found the null models best explained patterns of genetic differentiation at a local level and found several factors at the landscape level that weakly influenced gene flow. This research indicates multiscale investigations that incorporate local and landscape factors are valuable for understanding patterns of gene flow. Our findings suggest that dispersal rates in this system are high enough to minimize genetic structuring and that current forestry practices do not significantly impede dispersal.     

Ectomycorrhizal fungi are shared between seedlings and adults in a monodominant Gilbertiodendron dewevrei rain forest in Cameroon

Ectomycorrhizal networks may facilitate the establishment and survival of seedlings regenerating under the canopies of tropical forests and are often invoked as a potential contributor to monodominance. We identified ectomycorrhizal fungi in a monodominant Gilbertiodendron dewevrei (Fabaceae) rain forest in Cameroon, using sporocarps and ectomycorrhizae of three age categories (seedlings, intermediate trees, and large trees) and tentatively revealed nutrient transfer through ectomycorrhizal networks by measuring spontaneous isotopic (¹³C and ¹⁵N) abundances in seedlings. Sporocarp surveys revealed fewer ectomycorrhizal fungal taxa (59 species from 1030 sporocarps) than molecular barcoding of ectomycorrhizal roots (75 operational taxonomic units from 828 ectomycorrhizae). Our observations suggest that ectomycorrhizal fungal diversity is similar to that in other mixed tropical forests and provide the first report of the Tuber‐Helvella lineage in a tropical forest. Despite some differences, all age categories of G. dewevrei had overlapping ectomycorrhizal fungal communities, with families belonging to Thelephoraceae, Russulaceae, Sebacinaceae, Boletaceae, and Clavulinaceae. Of the 49 operational taxonomic units shared by the three age categories (65.3% of the ectomycorrhizal fungal community), 19 were the most abundant on root tips of all categories (38.7% of the shared taxa), supporting the likelihood of ectomycorrhizal networks. However, we obtained no evidence for nutrient transfer from trees to seedlings. We discuss the composition of the ectomycorrhizal fungal community among the G. dewevrei age categories and the possible role of common ectomycorrhizal networks in this rain forest.     

从酵母表达时间序列估计基因调控网络

基因调控网络是生命功能在基因表达层面上的展现。用组合线性调控模型、调控元件识别和基因聚类等方法 ,从基因组表达谱解读酵母在细胞周期与环境胁迫中的基因调控网络。结果表明 ,细胞在不同环境条件下会调整基因调控网络。在适应的环境下 ,起主要作用的是细胞生长和增殖有关的基因调控网络 ;而在响应环境胁迫时 ,细胞会再规划调控网络 ,抑制细胞生长和增殖相关的基因 ,诱导跟适应性糖类代谢与结构修复相关的基因 ,还可能启动减数分裂产生孢子。分别从细胞周期和环境胁迫响应相关基因中 ,搜索到转录因子Mcm1结合位点TT CC T GGAAA ,和Dal82在尿囊素代谢途径相关基因上的结合位点TGAAAAWTTT。从而 ,从酵母表达时间序列估计基因调控网络是可行的 ,与至今已知的实验观察相当吻合