Maximum Likelihood Neural Network Prediction Models

Neural networks have received much attention in recent years mostly by non-statisticians. The purpose of this paper is to incorporate neural networks in a non-linear regression model and obtain maximum likelihood estimates of the network parameters using a standard Newton-Raphson algorithm. We use maximum likelihood estimators instead of the usual back-propagation technique and compare the neural network predictions with predictions of quadratic regression models and with non-parametric nearest neighbor predictions. These comparisons are made using data generated from a variety of functions. Because of the number of parameters involved, neural network models can easily over-fit the data, hence validation of results is crucial.     

Refining predictions of climate change impacts on plant species distribution through the use of local statistics

Bioclimate envelope models are often used to predict changes in species distribution arising from changes in climate. These models are typically based on observed correlations between current species distribution and climate data. One limitation of this basic approach is that the relationship modelled is assumed to be constant in space; the analysis is global with the relationship assumed to be spatially stationary. Here, it is shown that by using a local regression analysis, which allows the relationship under study to vary in space, rather than conventional global regression analysis it is possible to increase the accuracy of bioclimate envelope modelling. This is demonstrated for the distribution of Spotted Meddick in Great Britain using data relating to three time periods, including predictions for the 2080s based on two climate change scenarios. Species distribution and climate data were available for two of the time periods studied and this allowed comparison of bioclimate envelope model outputs derived using the local and global regression analyses. For both time periods, the area under the receiver operating characteristics curve derived from the analysis based on local statistics was significantly higher than that from the conventional global analysis; the curve comparisons were also undertaken with an approach that recognised the dependent nature of the data sets compared. Marked differences in the future distribution of the species predicted from the local and global based analyses were evident and highlight a need for further consideration of local issues in modelling ecological variables.     

Estimation of the Spirometric Residual Volume (RV) by a Regression Built from Gower Distances

We estimate the Residual Volume, a spirometric parameter, by use of four continuous and four categorical variables. The estimation is done using distance-based regression, which allows to construct the predicting regression equation from mixed-type explanatory variables. The additionally introduced categorical variables improve essentially the goodness of fit of the regression equation.     

Experimental analysis of soft‐tissue fossilization: opening the black box

Taphonomic experiments provide important insights into fossils that preserve the remains of decay‐prone soft tissues, tissues that are usually degraded and lost prior to fossilization. These fossils are among the most scientifically valuable evidence of ancient life on Earth, giving us a view into the past that is much less biased and incomplete than the picture provided by skeletal remains alone. Although the value of taphonomic experiments is beyond doubt, a lack of clarity regarding their purpose and limitations, and ambiguity in the use of terminology, are hampering progress. Here we distinguish between processes that promote information retention and those that promote information loss, in order to clarify the distinction between fossilization and preservation. Recognizing distinct processes of decay, mineralization and maturation, the sequence in which they act, and the potential for interactions, has important consequences for analysis of fossils, and for the design of taphonomic experiments. The purpose of well‐designed taphonomic experiments is generally to understand decay, maturation and preservation individually, thus limiting the number of variables involved. Much work remains to be done, but these methodologically reductionist foundations will allow researchers to build towards more complex taphonomic experiments and a more holistic understanding and analysis of the interactions between decay, maturation and preservation in the fossilization of non‐biomineralized remains. Our focus must remain on the key issue of understanding what exceptionally preserved fossils reveal about the history of biodiversity and evolution, rather than on debating the scope and value of an experimental approach.     

Trait differentiation and adaptation of plants along elevation gradients

Studies of genetic adaptation in plant populations along elevation gradients in mountains have a long history, but there has until now been neither a synthesis of how frequently plant populations exhibit adaptation to elevation nor an evaluation of how consistent underlying trait differences across species are. We reviewed studies of adaptation along elevation gradients (i) from a meta‐analysis of phenotypic differentiation of three traits (height, biomass and phenology) from plants growing in 70 common garden experiments; (ii) by testing elevation adaptation using three fitness proxies (survival, reproductive output and biomass) from 14 reciprocal transplant experiments; (iii) by qualitatively assessing information at the molecular level, from 10 genomewide surveys and candidate gene approaches. We found that plants originating from high elevations were generally shorter and produced less biomass, but phenology did not vary consistently. We found significant evidence for elevation adaptation in terms of survival and biomass, but not for reproductive output. Variation in phenotypic and fitness responses to elevation across species was not related to life history traits or to environmental conditions. Molecular studies, which have focussed mainly on loci related to plant physiology and phenology, also provide evidence for adaptation along elevation gradients. Together, these studies indicate that genetically based trait differentiation and adaptation to elevation are widespread in plants. We conclude that a better understanding of the mechanisms underlying adaptation, not only to elevation but also to environmental change, will require more studies combining the ecological and molecular approaches.     

Fixed‐effect variance and the estimation of repeatabilities and heritabilities: issues and solutions

Linear mixed‐effects models are frequently used for estimating quantitative genetic parameters, including the heritability, as well as the repeatability, of traits. Heritability acts as a filter that determines how efficiently phenotypic selection translates into evolutionary change, whereas repeatability informs us about the individual consistency of phenotypic traits. As quantities of biological interest, it is important that the denominator, the phenotypic variance in both cases, reflects the amount of phenotypic variance in the relevant ecological setting. The current practice of quantifying heritabilities and repeatabilities from mixed‐effects models frequently deprives their denominator of variance explained by fixed effects (often leading to upward bias of heritabilities and repeatabilities), and it has been suggested to omit fixed effects when estimating heritabilities in particular. We advocate an alternative option of fitting models incorporating all relevant effects, while including the variance explained by fixed effects into the estimation of the phenotypic variance. The approach is easily implemented and allows optimizing the estimation of phenotypic variance, for example by the exclusion of variance arising from experimental design effects while still including all biologically relevant sources of variation. We address the estimation and interpretation of heritabilities in situations in which potential covariates are themselves heritable traits of the organism. Furthermore, we discuss complications that arise in generalized and nonlinear mixed models with fixed effects. In these cases, the variance parameters on the data scale depend on the location of the intercept and hence on the scaling of the fixed effects. Integration over the biologically relevant range of fixed effects offers a preferred solution in those situations.     

Light environment change induces differential expression of guppy opsins in a multi‐generational evolution experiment

Light environments critically impact species that rely on vision to survive and reproduce. Animal visual systems must accommodate changes in light that occur from minutes to years, yet the mechanistic basis of their response to spectral (color) changes is largely unknown. Here, we used a laboratory experiment where replicate guppy populations were kept under three different light environments for up to 8–12 generations to explore possible differences in the expression levels of nine guppy opsin genes. Previous evidence for opsin expression‐light environment “tuning” has been either correlative or focused exclusively on the relationship between the light environment and opsin expression over one or two generations. In our multigeneration experiment, the relative expression levels of nine different guppy opsin genes responded differently to light environment changes: some did not respond, while others differed due to phenotypic plasticity. Moreover, for the LWS‐1 opsin we found that, while we observed a wide range of plastic responses under different light conditions, common plastic responses (where the population replicates all followed the same trajectory) occurred only after multigenerational exposure to different light environments. Taken together this suggests that opsin expression plasticity plays an important role in light environment “tuning” in different light environments on different time scales, and, in turn, has important implications for both visual system function and evolution.     

基于地理加权模型的南设得兰群岛北部南极磷虾渔场空间分布影响分析

结合南极磷虾渔业科学观察员收集的渔业数据和海洋环境数据,本研究利用地理加权回归模型(GWR),分析了具有空间属性的虾群深度和离岸距离两个因子,以及海水表温对南设得兰群岛北侧水域南极磷虾渔场空间分布的影响.结果表明:各年南极磷虾渔业单位捕捞努力量渔获量(CPUE)在空间上的分布无显著的集聚性;2010和2013年,3个因子之间存在空间自相关性(正相关),而2012和2016年则无自相关性.GWR模型结果显示,3个因子对CPUE的空间分布具有不同程度的影响,影响程度大小依次为虾群深度>离岸距离>温度.拟合结果发现,南设得兰群岛东、西两侧水域中表温对CPUE空间分布的影响与其他两个因子具有相反的趋势.虾群深度和离岸距离对CPUE的空间效应主要表现为负相关,但存在着年际和区域性差异.本研究结果可为南极磷虾渔场形成机制研究提供方法上的参考.