The Treatment of Missing Values in Logistic Regression

The efficiencies of the estimators in the linear logistic regression model are examined using simulations under six missing value treatments. These treatments use either the maximum likelihood or the discriminant function approach in the estimation of the regression coefficients. Missing values are assumed to occur at random. The cases of multivariate normal and dichotomous independent variables are both considered. We found that in general, there is no uniformly best method. However, mean substitution and discriminant function estimation using existing pairs of values for correlations turn out to be favourable for the cases considered.     

The effects of X-rays on the proliferation dynamics of cells in the imaginal wing disc ofDrosophila melanogaster

Summary We report on the size distribution of clones marked by mitotic recombination induced by several different doses of X-rays applied to 72 h oldDrosophila larvae. The results indicate that the radiation significantly reduces the number of cells which undergo normal proliferation in the imaginal wing disc. We estimate that 1000 r reduces by 40–60% the number of cells capable of making a normal contribution to the development of the adult wing. Part of this reduction is due to severe curtailment in the proliferative ability of cells which nevertheless remain capable of adult differentiation; this effect is possibly due to radiation-induced aneuploidy. Cytological evidence suggests that immediate cell death also occurs as a result of radiation doses as low as 100 r. The surviving cells are stimulated to undergo additional proliferation in response to the X-ray damage so that the result is the differentiation of a normal wing.     

Mass Coral Reef Bleaching: A Recent Outcome of Increased El Niño Activity?

Coral reefs are generally considered to be the most biologically productive of all marine ecosystems, but in recent times these vulnerable aquatic resources have been subject to unusual degradation. The general decline in reefs has been greatly accelerated by mass bleaching in which corals whiten en masse and often fail to recover. Empirical evidence indicates a coral reef bleaching cycle in which major bleaching episodes are synchronized with El Niño events that occur every 3–4 years on average. By heating vast areas of the Pacific Ocean, and affecting the Indian and Atlantic Oceans as well, El Niño causes widespread damage to reefs largely because corals are very sensitive to temperature changes. However, mass bleaching events were rarely observed before the 1970s and their abrupt appearance two decades ago remains an enigma. Here we propose a new explanation for the sudden occurrence of mass bleaching and show that it may be a response to the relative increase in El Niño experienced over the last two decades.     

Assessment of Optimal Selected Prognostic Factors

The identification and assessment of prognostic factors is one of the major tasks in clinical research. The assessment of one single prognostic factor can be done by recently established methods for using optimal cutpoints. Here, we suggest a method to consider an optimal selected prognostic factor from a set of prognostic factors of interest. This can be viewed as a variable selection method and is the underlying decision problem at each node of various tree building algorithms. We propose to use maximally selected statistics where the selection is defined over the set of prognostic factors and over all cutpoints in each prognostic factor. We demonstrate that it is feasible to compute the approximate null distribution. We illustrate the new variable selection test with data of the German Breast Cancer Study Group and of a small study on patients with diffuse large B‐cell lymphoma. Using the null distribution for a p‐value adjusted regression trees algorithm, we adjust for the number of variables analysed at each node as well. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Phylogeography of the leaf beetle Chrysolina virgata in wetlands of Japan inferred from the distribution of mitochondrial haplotypes

The genetic differentiation among populations of the leaf beetle Chrysolina virgata living in wetlands of Japan was studied based on the sequence data of the mitochondrial cytochrome oxidase subunit I gene region (750 bp). Two distinct lineages of mitochondrial haplotypes were found: one (clade A) consisted of 26 haplotypes distributed over the distribution range of C. virgata between north‐east Honshu and Kyushu, whereas the other (clade B) was monotypic and confined to a small region in north‐east Honshu where it coexisted with clade A. Nested clade analysis for these haplotypes suggested that range expansion and following differentiation due to isolation by distance might have resulted in the present distribution pattern of the haplotypes in clade A. We discuss the evolutionary process leading to the occurrence of two distinct haplotype clades in Japan in terms of repeated colonization from the continent and range expansion and contraction during climatic changes.     

Conditional Maximum Likelihood Estimation Following a Group Sequential Test

We consider estimation after a group sequential test. An estimator that is unbiased or has small bias may have substantial conditional bias (Troendle and Yu, 1999, Coburger and Wassmer, 2001). In this paper we derive the conditional maximum likelihood estimators of both the primary parameter and a secondary parameter, and investigate their properties within a conditional inference framework. The method applies to both the usual and adaptive group sequential test designs. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Metazoan parasites of Engraulis ringens as tools for stock discrimination along the Chilean coast

The value of the metazoan parasites of the anchoveta Engraulis ringens as biological tags for stock discrimination is assessed. Three hundred and fifty-nine specimens, obtained from six landing ports in northern and central Chile (Arica, Antofagasta, Caldera, Coquimbo, Valparaiso and Talcahuano) covering a latitudinal gradient of c . 18°, were analysed. Six metazoan parasite species (three ectoparasites and three endoparasites) were obtained. A correspondence analysis suggested that anchoveta from northern and central Chile are comprised of two stocks. Identification of stocks was largely based on qualitative differences (absence in southern localities of the monogenean Pseudanthocotyloides heterocotyle , a parasite proper of engraulids, and the copepod Caligus sp.) and quantitative differences in the prevalence of infection for the remaining species. Inertia and mass value for the correspondence analysis supported the hypothesis of two stocks.     

Stable isotopic niche predicts fitness of prey in a wolf–deer system

Interindividual variation in niche presents a potentially central object on which natural selection can act. This may have important evolutionary implications because habitat use governs a suite of selective forces encountered by foragers. In a free-living native black-tailed deer, Odocoileus hemionus , population from coastal British Columbia, we used stable isotope analysis to identify individual variation in foraging niche and investigated its relationship to fitness. Using an intragenerational comparison of surviving and nonsurviving O. hemionus over 2 years of predation by wolves, Canis lupus, we detected resource-specific fitness. Individuals with isotopic signatures that suggested they foraged primarily in cedar ( Thuja plicata )-dominated and low-elevation hemlock ( Tsuga heterophylla )-dominated forest stands were more likely to be killed by C. lupus . High-quality forage in T. plicata stands, as indexed by protein content, may be involved in maintaining this foraging phenotype. Moreover, nonsurvivors diverged more than survivors from median isotopic signatures, suggesting selection against foraging specialization. Stable isotope analysis provides a novel opportunity to integrate ecological and selective landscapes in order to identify underlying ecological mechanisms of selection and provide insight into the maintenance of variability.  © 2007 The Linnean Society of London, Biological Journal of the Linnean Society , 2007, 90 , 125–137.