Measurement of concurrent selection episodes

Current methods for measuring selection with longitudinal data have been developed with the assumption that episodes of selection are sequential. However, a number of empirical examinations have demonstrated that natural and sexual selection may act concurrently and in opposing directions. Other recent work has highlighted the difficulty of assigning fitness values for survival when reproduction and mortality within a population temporally overlap. I treat these as facets of a single problem; how to analyze selection where mortality and reproduction are concurrent. To address this problem, I formalize a method to estimate total fitness of individuals over a period of time utilizing longitudinal data. I then show how the fitness may be partitioned to provide two separate estimates of fitness for reproductive opportunity and reproductive success. In addition, another total fitness estimate for the period can be obtained from the two partitioned estimates. This procedure will allow calculation of total fitness where there are some missing datapoints for reproductive success of an individual. A simulation indicates that bias is generally low for the various fitness estimates. These methods should expand our ability to understand the interaction of different selection episodes.     

The file-drawer problem revisited: a general weighted method for calculating fail-safe numbers in meta-analysis

Quantitative literature reviews such as meta-analysis are becoming common in evolutionary biology but may be strongly affected by publication biases. Using fail-safe numbers is a quick way to estimate whether publication bias is likely to be a problem for a specific study. However, previously suggested fail-safe calculations are unweighted and are not based on the framework in which most meta-analyses are performed. A general, weighted fail-safe calculation, grounded in the meta-analysis framework, applicable to both fixed- and random-effects models, is proposed. Recent meta-analyses published in Evolution are used for illustration.     

Missing covariates in longitudinal data with informative dropouts: bias analysis and inference

We consider estimation in generalized linear mixed models (GLMM) for longitudinal data with informative dropouts. At the time a unit drops out, time-varying covariates are often unobserved in addition to the missing outcome. However, existing informative dropout models typically require covariates to be completely observed. This assumption is not realistic in the presence of time-varying covariates. In this article, we first study the asymptotic bias that would result from applying existing methods, where missing time-varying covariates are handled using naive approaches, which include: (1) using only baseline values; (2) carrying forward the last observation; and (3) assuming the missing data are ignorable. Our asymptotic bias analysis shows that these naive approaches yield inconsistent estimators of model parameters. We next propose a selection/transition model that allows covariates to be missing in addition to the outcome variable at the time of dropout. The EM algorithm is used for inference in the proposed model. Data from a longitudinal study of human immunodeficiency virus (HIV)-infected women are used to illustrate the methodology.     

Infections with varying contact rates: application to varicella

We develop methods for the analysis of infectious disease data when age-specific contact rates vary over time. Our methods are valid when contact rates vary slowly on the time scale of the infection process, and are applicable to a variety of data types including serial seroprevalence surveys and case reports. The methods exploit approximate endemic equilibria, and require numerical solution of an associated integral equation in age and time. We also estimate summary statistics such as time-dependent analogs of the basic reproduction number and critical immunization threshold. We illustrate the methods with data on varicella (chickenpox) in the United Kingdom.     

Effects of gene expression on molecular evolution in Arabidopsis thaliana and Arabidopsis lyrata

We analyzed the complete genome sequence of Arabidopsis thaliana and sequence data from 83 genes in the outcrossing A. lyrata, to better understand the role of gene expression on the strength of natural selection on synonymous and replacement sites in Arabidopsis. From data on tRNA gene abundance, we find a good concordance between codon preferences and the relative abundance of isoaccepting tRNAs in the complete A. thaliana genome, consistent with models of translational selection. Both EST-based and new quantitative measures of gene expression (MPSS) suggest that codon preferences derived from information on tRNA abundance are more strongly associated with gene expression than those obtained from multivariate analysis, which provides further support for the hypothesis that codon bias in Arabidopsis is under selection mediated by tRNA abundance. Consistent with previous results, analysis of protein evolution reveals a significant correlation between gene expression level and amino acid substitution rate. Analysis by MPSS estimates of gene expression suggests that this effect is primarily the result of a correlation between the number of tissues in which a gene is expressed and the rate of amino acid substitution, which indicates that the degree of tissue specialization may be an important determinant of the rate of protein evolution in Arabidopsis.     

A Test of the Higher-Taxon Approach in the Identification of Candidate Sites for Marine Reserves

Alternatives to species-level identification have been advocated as one solution to the problem of selecting marine reserves with limited information on the distribution of marine biodiversity.This study evaluated the effects on selection of candidate sites for marine reserves from using the higher-taxon approach as a surrogate for species-level identification of intertidal molluscs and rocky reef fishes. These effects were evaluated by determining the percentage of species included in candidate reserves identified from genus-, family- and order-level data by a complementarity-based reserve selection algorithm, and by testing for correlations between the irreplaceability values of locations. Candidate reserves identified from genus- and family-level data of intertidal molluscs included a similar percentage of all species as the reserves identified from species-level data. Candidate reserves selected from genus- and family-level data of rocky reef fishes included, respectively, 3–7% and 14–23% fewer species than reserves selected from species-level data. When the reserve identification process was constrained by a practical planning limit (a maximum of 20% locations able to be reserved) the reserves selected from genus- and family-level data of intertidal molluscs, and genus-level data of rocky reef fishes, included a similar percentage of species as the reserves identified from species-level data. Irreplaceability values of locations for species, genera and families of intertidal molluscs were highly correlated, and irreplaceability values of locations for species and genera of rocky reef fishes were highly correlated. This study suggests that genus- and family-level data for intertidal molluscs, and genus-level data for rocky reef fishes, are suitable surrogates for species in the identification of candidate sites for marine reserves.     

Comparing the likelihood ratios of two binary diagnostic tests in the presence of partial verification

The comparison of the efficiency of two binary diagnostic tests requires one to know the disease status for all patients in the sample, by applying a gold standard. In two-phase studies the gold standard is not applied to all patients in a sample, and the problem of partial verification of the disease arises. At present, one of the approaches most used for comparing two binary diagnostic tests are the likelihood ratios. In this study, the maximum likelihood estimators of likelihood ratios are obtained. The tests of hypothesis to compare the likelihood ratios of two binary diagnostic tests when both are applied to the same random sample in the presence of verification bias are deduced, and simulation experiments are performed in order to investigate the asymptotic behaviour of the tests of hypothesis. The results obtained have been applied to the study of Alzheimer's disease.