Priors for the Bayesian star paradox

We show that the Bayesian star paradox, first proved mathematically by Steel and Matsen for a specific class of prior distributions, occurs in a wider context including less regular, possibly discontinuous, prior distributions.     

Pulsed-field gel electrophoresis for genotypic comparison of Rhizobium bacteria that nodulate leguminous trees

Abstract Pulsed-field gel electrophoresis (PFGE) was applied to characterize Rhizobium bacteria isolated from the root nodules of Acacia senegal and Prosopis chilensis trees growing in Sudan and Keya. For the electrophoresis, the total DNA of 42 isolates, embedded in agarose, was digested by a rare-cutting restriction endonuclease, Xba I. The PFGE run resulted in good resolution of the DNA fragments and gave the strains distinctive fingerprint patterns. The patterns were analysed visually and using automated clustering analysis, which divided the strains into groups resembling the results generated by numerical taxonomy. However, several strains had unique banding patterns, which indicates that these strains are genetically very diverse.     

Bayesian Quantile Regression for Longitudinal Studies with Nonignorable Missing Data

Summary .  We study quantile regression (QR) for longitudinal measurements with nonignorable intermittent missing data and dropout. Compared to conventional mean regression, quantile regression can characterize the entire conditional distribution of the outcome variable, and is more robust to outliers and misspecification of the error distribution. We account for the within-subject correlation by introducing a   ℓ₂   penalty in the usual QR check function to shrink the subject-specific intercepts and slopes toward the common population values. The informative missing data are assumed to be related to the longitudinal outcome process through the shared latent random effects. We assess the performance of the proposed method using simulation studies, and illustrate it with data from a pediatric AIDS clinical trial.     

ABC as a flexible framework to estimate demography over space and time: some cons,many pros

The analysis of genetic variation to estimate demographic and historical parameters and to quantitatively compare alternative scenarios recently gained a powerful and flexible approach: the Approximate Bayesian Computation (ABC). The likelihood functions does not need to be theoretically specified, but posterior distributions can be approximated by simulation even assuming very complex population models including both natural and human‐induced processes. Prior information can be easily incorporated and the quality of the results can be analysed with rather limited additional effort. ABC is not a statistical analysis per se, but rather a statistical framework and any specific application is a sort of hybrid between a simulation and a data‐analysis study. Complete software packages performing the necessary steps under a set of models and for specific genetic markers are already available, but the flexibility of the method is better exploited combining different programs. Many questions relevant in ecology can be addressed using ABC, but adequate amount of time should be dedicated to decide among alternative options and to evaluate the results. In this paper we will describe and critically comment on the different steps of an ABC analysis, analyse some of the published applications of ABC and provide user guidelines.     

How intraspecific variation in seed‐dispersing animals matters for plants

Seed dispersal by animals is a complex phenomenon, characterized by multiple mechanisms and variable outcomes. Most researchers approach this complexity by analysing context‐dependency in seed dispersal and investigating extrinsic factors that might influence interactions between plants and seed dispersers. Intrinsic traits of seed dispersers provide an alternative way of making sense of the enormous variation in seed fates. I review causes of intraspecific variability in frugivorous and granivorous animals, discuss their effects on seed dispersal, and outline likely consequences for plant populations and communities. Sources of individual variation in seed‐dispersing animals include sexual dimorphism, changes associated with growth and ageing, individual specialization, and animal personalities. Sexual dimorphism of seed‐dispersing animals influences seed fate through diverse mechanisms that range from effects caused by sex‐specific differences in body size, to influences of male versus female cognitive functions. These differences affect the type of seed treatment (e.g. dispersal versus predation), the number of dispersed seeds, distance of seed dispersal, and likelihood that seeds are left in favourable sites for seeds or seedlings. The best‐documented consequences of individual differences associated with growth and ageing involve quantity of dispersed seeds and the quality of seed treatment in the mouth and gut. Individual specialization on different resources affects the number of dispersed plant species, and therefore the connectivity and architecture of seed‐dispersal networks. Animal personalities might play an important role in shaping interactions between plants and dispersers of their seeds, yet their potential in this regard remains overlooked. In general, intraspecific variation in seed‐dispersing animals often influences plants through effects of these individual differences on the movement ecology of the dispersers. Two conditions are necessary for individual variation to exert a strong influence on seed dispersal. First, the individual differences in traits should translate into differences in crucial characteristics of seed dispersal. Second, individual variation is more likely to be important when the proportions of particular types of individuals fluctuate strongly in a population or vary across space; when proportions are static, it is less likely that intraspecific differences will be responsible for changes in the dynamics and outcomes of plant–animal interactions. In conclusion, focusing on variation among foraging animals rather than on species averages might bring new, mechanistic insights to the phenomenon of seed dispersal. While this shift in perspective is unlikely to replace the traditional approach (based on the assumption that all important variation occurs among species), it provides a complementary alternative to decipher the enormous variation observed in animal‐mediated seed dispersal.     

Fine‐scale Population Genetic Structure of Two Dioecious Indian Keystone Species,Ficus hispida and Ficus exasperata (Moraceae)

Although Ficus (Moraceae) is a keystone plant genus in the tropics, providing resources to many frugivorous vertebrates, its population genetic structure, which is an important determinant of its long‐term survival, has rarely been investigated. We examined the population genetic structure of two dioecious fig species (Ficus hispida and Ficus exasperata) in the Indian Western Ghats using co‐dominant nuclear microsatellite markers. We found high levels of microsatellite genetic diversity in both species. The regression slopes between genetic relationship coefficients (f_ij) and spatial distances were significantly negative in both species indicating that, on average, individuals in close spatial proximity were more likely to be related than individuals further apart. Mean parent–offspring distance (σ) calculated using these slopes was about 200 m in both species. This should be contrasted with the very long pollen dispersal distances documented for monoecious Ficus species. Nevertheless, overall population genetic diversity remained large suggesting immigrant gene flow. Further studies will be required to analyze broader scale patterns.     

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)     

Juvenile bimodal length distribution and sea-run migration of the lower modal group in the Pacific Ocean form of three-spined stickleback

Juvenile three-spined stickleback Gasterosteus aculeatus in their nursery pond had a bimodal length distribution after October in their first year of life. A sea-run migration of all individuals belonging to the lower modal group was observed, with a peak in early November. These data suggest that partial migration based on the early status of individual juveniles occurs in the Pacific Ocean form of three-spined stickleback.