Books

Environmental Educational Advocacy: changing perspectives of Ecology and Education. Editors E. Johnson and M. Mappin, Cambridge 2005, hardback 333 pages. ISBN 0 531 824109 Price £55. Reviewed by Sue Dale Tunnicliffe

Minibeasts: An Identification Guide Peter Smithers Illustrated by John Walters. Published by John Walters Number of pages 36. £6.50. ISBN 0 9540256 2 8. Reviewed by Myles Ripley     

Journal of Cellular Biochemistry: Volume 113, Number 9, September, 2012

Cover: The nuclear envelope. Please see article by Walters et al., pages 2813–2821, this issue.     

Farmers and their fires,fishers and their fish: Production and productivity in pre-European South-East Queensland

Ian Walters is a Professor of Anthropology at Northern Territory University, Casuerina, Northern Territory.     

Some inequalities for the expectation of a product of functions of a random variable and for the multivariate distribution function at a random point

Under certain assumptions the expectation of a product of functions of a random variable is greater (smaller) than the product of expectations. The multivariate distribution function of m independent random variables at a random point is greater than the product of the distribution functions of the m variables.     

Booknote

Book reviewed:
Albert R. Jonsen, Robert M. Veatch and LeRoy Walters (eds), Sourcebook in Bioethics     

Selection Problems in Balanced Block Designs

If we have independent random variables the solution of BECHHOFER's selection problem in special cases is easily applied because we can use tables of a standardized multivariate normal distribution. The aim of this note is to show how such tables are also applicable in a case of correlated random variables like the estimates of treatment effects in BIB and BCB. Formulae for a choice of minimal sample sizes are also given.     

Book Reviews

Svante R. Orell, Gregory F. Sterrett, Max N‐I Walters, Darrel Whitaker. Churchill Livingstone. 446 pp     

Multiple imputation and posterior simulation for multivariate missing data in longitudinal studies

This paper outlines a multiple imputation method for handling missing data in designed longitudinal studies. A random coefficients model is developed to accommodate incomplete multivariate continuous longitudinal data. Multivariate repeated measures are jointly modeled; specifically, an i.i.d. normal model is assumed for time-independent variables and a hierarchical random coefficients model is assumed for time-dependent variables in a regression model conditional on the time-independent variables and time, with heterogeneous error variances across variables and time points. Gibbs sampling is used to draw model parameters and for imputations of missing observations. An application to data from a study of startle reactions illustrates the model. A simulation study compares the multiple imputation procedure to the weighting approach of Robins, Rotnitzky, and Zhao (1995, Journal of the American Statistical Association 90, 106-121) that can be used to address similar data structures.     

Accelerated failure time (AFT) modeling for the development and survival of Russian wheat aphid, <Emphasis Type="Italic">Diuraphis noxia</Emphasis> (Mordvilko)

Accelerated failure time model (AFT) and Cox’s proportional hazards model (PHM) are considered the two most significant models in survival analysis, which has become a de facto standard for biomedical data analysis and modeling. AFT not only plays an extremely significant role in survival analysis but also finds extensive applications in engineering reliability. Survival analysis studies a special type of random variables: time-to-event (also known as failure time, lifetime or survival time) random variables. Examples of time-to-event random variables include survival times of patients in a clinical trial and failure times of machine components. Since molting and death times of insect individuals are also perfect examples of time-to-event random variables, we argue that survival analysis including AFT modeling is ideal for analyzing insect development and survival data, and further for building dynamic models of insect development and survival. Here we demonstrate such an application with data collected by observing stage-to-stage development and survival of 1,800 Russian wheat aphids (RWA), Diuraphis noxia, reared in laboratory growth chambers arranged in 25 treatments (each with 72 individuals). The main advantages of survival analysis, including the unified modeling of survival and development as well as handling of information censoring, are also discussed.     

Bias in random forest variable importance measures: Illustrations,sources and a solution

Background  

Variable importance measures for random forests have been receiving increased attention as a means of variable selection in many classification tasks in bioinformatics and related scientific fields, for instance to select a subset of genetic markers relevant for the prediction of a certain disease. We show that random forest variable importance measures are a sensible means for variable selection in many applications, but are not reliable in situations where potential predictor variables vary in their scale of measurement or their number of categories. This is particularly important in genomics and computational biology, where predictors often include variables of different types, for example when predictors include both sequence data and continuous variables such as folding energy, or when amino acid sequence data show different numbers of categories.     

Predictors for Physical Activity in Adolescent Girls Using Statistical Shrinkage Techniques for Hierarchical Longitudinal Mixed Effects Models

We examined associations among longitudinal, multilevel variables and girls’ physical activity to determine the important predictors for physical activity change at different adolescent ages. The Trial of Activity for Adolescent Girls 2 study (Maryland) contributed participants from 8th (2009) to 11th grade (2011) (n=561). Questionnaires were used to obtain demographic, and psychosocial information (individual- and social-level variables); height, weight, and triceps skinfold to assess body composition; interviews and surveys for school-level data; and self-report for neighborhood-level variables. Moderate to vigorous physical activity minutes were assessed from accelerometers. A doubly regularized linear mixed effects model was used for the longitudinal multilevel data to identify the most important covariates for physical activity. Three fixed effects at the individual level and one random effect at the school level were chosen from an initial total of 66 variables, consisting of 47 fixed effects and 19 random effects variables, in additional to the time effect. Self-management strategies, perceived barriers, and social support from friends were the three selected fixed effects, and whether intramural or interscholastic programs were offered in middle school was the selected random effect. Psychosocial factors and friend support, plus a school’s physical activity environment, affect adolescent girl’s moderate to vigorous physical activity longitudinally.     

Immunoenzyme analysis. The isolation and characteristics of Soviet reagents for detecting the surface antigen of the hepatitis B virus in blood sera

A method for the preparation of highly specific conjugates on the basis of antibodies obtained as a result of dissociation of the immune complex HBsAg--anti-HBsAg, formed in vitro, is proposed. ELISA was performed according to Walters with some modifications in the concentration of the assay ingredients and in the time of incubation. The technique of the assay of blood sera for the presence of HBsAg with the use of reagents manufactured in the USSR has been worked out.     

Restricted BLUP for Mixed Linear Models

A new estimation procedure for mixed regression models is introduced. It is a development of Henderson's best linear unbiased prediction procedure which uses the joint distribution of the observed dependent random variables and the unknown realisations of the random components of the model. It is proposed to replace the likelihood of the observations given the random components by the asymptotic likelihood of the maximum likelihood estimators and the prior distribution of the random components by a restricted prior distribution which is consistent with the usual restrictions placed on the random components when they are considered conditionally fixed.     

Dependence Functions and Frailties

This note considers association between nonnegative random variables in which the two observed survival times depend on an unobservable random variable via the proportional hazard model. When the random variables are subject to censoring, the conditional hazard functions provides a reasonable means of describing the association between the two variables. A numerical example demonstrating association in disease incidence in ordered pairs of individuals is analysed. Also, examples of distributions satisfying the notions of dependence considered are provided.     

Spatio-temporal Foraging Dynamics in Two Coexisting Harvester Ants (Hymenoptera: Formicidae)

Different aspects of the foraging strategies of two harvester ant species, Messor wasmanni and M. minor, were investigated in a Mediterranean dry grassland area. Baits were used to evaluate the existence of a trade-off between resource discovery and domination as well as the effect of three variables (air temperature, relative humidity and distance) on the trade-off. Baits were also utilized to explore random vs non random use of time by colonies. Random vs non random utilization of space was instead evaluated by mapping the daily foraging area of colonies in a grid of 900 plots of 1 m² each. Results revealed that species coexistence is not preferentially supported by a trade-off in resource utilization with no overall effect of the examined variables. The foraging activity of the two species widely overlapped whilst a clear competition for space occurred. The observed space partitioning could represent an advantageous strategy for the coexistence of the two ant species.     

Quantification of sensory information transmission using timeseries decorrelation techniques

To estimate the information transmitted across a neuronal sensory system one has to deal with serial dependence among consecutive samples of the stimulus and the response signal. Common methods usually require a huge amount of data, or are restricted to Gaussian stimuli. Here, we describe stimulus and response as stochastic processes, i.e. as sequences of random variables, in the same coordinate system. Stimulus-response pairs of these random variables must not be considered independently because otherwise the transinformation is overestimated. To account for the linear fraction of the serial dependence, we present two decorrelation techniques based on coordinate transformation. They provide a representation of the processes with uncorrelated random variables and yield a more precise estimate of the transinformation.     

Gene selection and classification of microarray data using random forest

Background  

Selection of relevant genes for sample classification is a common task in most gene expression studies, where researchers try to identify the smallest possible set of genes that can still achieve good predictive performance (for instance, for future use with diagnostic purposes in clinical practice). Many gene selection approaches use univariate (gene-by-gene) rankings of gene relevance and arbitrary thresholds to select the number of genes, can only be applied to two-class problems, and use gene selection ranking criteria unrelated to the classification algorithm. In contrast, random forest is a classification algorithm well suited for microarray data: it shows excellent performance even when most predictive variables are noise, can be used when the number of variables is much larger than the number of observations and in problems involving more than two classes, and returns measures of variable importance. Thus, it is important to understand the performance of random forest with microarray data and its possible use for gene selection.     

Richard Meinertzhagen—a case of fraud examined

Museum collections are rich repositories of information. The specimens and the data they bear continue to provide new insights into ornithology and biological processes decades or even centuries after they were collected. The benefits to be gained from museum collections depend implicitly upon the accuracy of the information associated with the specimens and the correct interpretation of those data (Parkes 1989, Knox & Walters 1992). Collectors and dealers have often been suspected of fabricating data for a variety of reasons, but proven cases have been documented only rarely (e.g. Nicholson & Ferguson-Lees 1962). This paper examines one such case.     

Notes on the flora and vegetation of Canna and Sanday

Summary

During a visit to Canna and Sanday we recorded 26 species of vascular plant and 64 species of bryophyte that do not appear to have been recorded from these islands before. We list these species, plus nine additional vascular plants recently recorded by E. Anderson and P. Castro and seven species seen by us, recorded by Heslop Harrison (1939) but not included for the grid square containing Canna and Sanday by Perring and Walters (1976) or Jermy et al. (1978). Notes on some of the more interesting vegetation types are also included.