Confidence intervals for the mean of lognormal data with excess zeros

In this paper, we consider an approach based on the adjusted signed log-likelihood ratio statistic for constructing a confidence interval for the mean of lognormal data with excess zeros. An extensive simulation study suggests that the proposed approach outperforms all the existing methods in terms of coverage probabilities and symmetry of upper and lower tail error probabilities. Finally, we analyzed two real-life datasets using the proposed approach.     

The normal distribution as appropriate model of developmental instability in Opuntia cacti flowers

Developmental instability (DI) as measured by fluctuating asymmetry (FA) has been proposed to reflect fitness and stress. Furthermore, the associated developmental buffering may reduce morphological variation, conceal the expression of genetic variation and as such play an important role in evolutionary biology. However, observed associations between FA and various forms of stress and quality appear very heterogeneous. Presently it is difficult to interpret the biological relevance of this heterogeneity because little is known about the link between FA and the underlying process of DI, casting doubt whether DI can be viewed as an individual property and how closely FA reflects the underlying process of DI. Therefore, studies that explicitly test the validity of assumptions of the proposed theoretical models and estimate between‐individual variations in DI are needed. We present data on Opuntia cacti floral traits confirming that the normal distribution can be viewed as an appropriate approximation of the distribution of DI and that the concept of hypothetical repeatability can provide useful insights into the interpretation of patterns in FA as a measure of DI. Furthermore, we detected significant between‐individual variation in DI. Measuring petals from several flowers within individual plants allowed making inference of individual DI.     

ProMetheusDB: An In-Depth Analysis of the High-Quality Human Methyl-proteome

Protein arginine (R) methylation is a post-translational modification involved in various biological processes, such as RNA splicing, DNA repair, immune response, signal transduction, and tumor development. Although several advancements were made in the study of this modification by mass spectrometry, researchers still face the problem of a high false discovery rate. We present a dataset of high-quality methylations obtained from several different heavy methyl stable isotope labeling with amino acids in cell culture experiments analyzed with a machine learning–based tool and show that this model allows for improved high-confidence identification of real methyl-peptides. Overall, our results are consistent with the notion that protein R methylation modulates protein–RNA interactions and suggest a role in rewiring protein–protein interactions, for which we provide experimental evidence for a representative case (i.e., NONO [non-POU domain–containing octamer-binding protein]–paraspeckle component 1 [PSPC1]). Upon intersecting our R-methyl-sites dataset with the PhosphoSitePlus phosphorylation dataset, we observed that R methylation correlates differently with S/T-Y phosphorylation in response to various stimuli. Finally, we explored the application of heavy methyl stable isotope labeling with amino acids in cell culture to identify unconventional methylated residues and successfully identified novel histone methylation marks on serine 28 and threonine 32 of H3. The database generated, named ProMetheusDB, is freely accessible at https://bioserver.ieo.it/shiny/app/prometheusdb.     

Understanding ‘it depends’ in ecology: a guide to hypothesising,visualising and interpreting statistical interactions

Ecologists routinely use statistical models to detect and explain interactions among ecological drivers, with a goal to evaluate whether an effect of interest changes in sign or magnitude in different contexts. Two fundamental properties of interactions are often overlooked during the process of hypothesising, visualising and interpreting interactions between drivers: the measurement scale – whether a response is analysed on an additive or multiplicative scale, such as a ratio or logarithmic scale; and the symmetry – whether dependencies are considered in both directions. Overlooking these properties can lead to one or more of three inferential errors: misinterpretation of (i) the detection and magnitude (Type-D error), and (ii) the sign of effect modification (Type-S error); and (iii) misidentification of the underlying processes (Type-A error). We illustrate each of these errors with a broad range of ecological questions applied to empirical and simulated data sets. We demonstrate how meta-analysis, a widely used approach that seeks explicitly to characterise context dependence, is especially prone to all three errors. Based on these insights, we propose guidelines to improve hypothesis generation, testing, visualisation and interpretation of interactions in ecology.     

Mathematical Formulations for the Height/Weight Tables of U.K. Children after Tanner,Whitehouse and Takaishi

Models of human height and weight growth and their interrelations are applied to the height/weight tables for British children given by Tanner, Whitehouse and Takaishi (1966) in two elaborate papers. Approximations using nonlinear regressions yield very satisfying results for height/age and log weight/height relations as well as the velocities of height and weight. Graphs with the basic data and the approximated growth curves are added.     

Simultaneous confidence bands for log‐logistic regression with applications in risk assessment

In risk assessment, it is often desired to make inferences on the low dose levels at which a specific benchmark risk is attained. Applications of simultaneous hyperbolic confidence bands for low‐dose risk estimation with quantal data under different dose‐response models (multistage, Abbott‐adjusted Weibull, and Abbott‐adjusted log‐logistic models) have appeared in the literature. The use of simultaneous three‐segment bands under the multistage model has also been proposed recently. In this article, we present explicit formulas for constructing asymptotic one‐sided simultaneous hyperbolic and three‐segment bands for the simple log‐logistic regression model. We use the simultaneous construction to estimate upper hyperbolic and three‐segment confidence bands on extra risk and to obtain lower limits on the benchmark dose by inverting the upper bands on risk under the Abbott‐adjusted log‐logistic model. Monte Carlo simulations evaluate the characteristics of the simultaneous limits. An example is given to illustrate the use of the proposed methods and to compare the two types of simultaneous limits at very low dose levels.     

The effect of habitat stability on benthic invertebrate communities: the utility of species abundance distributions

Spatial and temporal patterns in the species abundance distribution of benthic invertebrate communities of 11 freshwater habitats (10 streams and a wind-swept lake shore) were examined with respect to habitat stability. Abundance patterns varied markedly between seasons at most sites. However, mean abundance distributions at 4 of the 5 unstable sites and the 2 most stable sites were dominated by one or two taxa with a large number of rare species, whereas sites of intermediate stability had more equitable distributions. Both the log series and log normal distributions were statistically indistinguishable, at the 5% level, from all the observed mean abundance patterns. In contrast, graphical comparisons of the observed and fitted distributions suggested the log series may be the better fit at most of the unstable sites and the two most stable sites, whereas the more equitable distribution at sites of intermediate stability suggested the log normal distribution was the better fit. If conditions at a site favoured one or two species, either through severe physical conditions, or through competitive superiority in the absence of disturbance then the log series distribution may result. However, if no species in the community was strongly advantaged over others, a log normal distribution should result. Given the discriminating power of the appropriate statistical test it may not, however, be possible to pick up these differences without graphical comparisons as well.     

Understanding regional species diversity through the log series distribution of occurrences

The distribution of benthic foraminifera around the continental margins of North America is extensively documented. Data from 2673 localities consists of a synonomized list of 2329 species (S) and 61 369 occurrences (n). Here, the margins are divided into five geographical regions: Pacific (PA), S=965, n=19 014; Arctic (AR), S=458, n=7342; Atlantic (AT), S=878, n=10 034; Gulf of Mexico (GM), S=849, n=18 011; Caribbean (CR), S=1188, N=6968. As for many other organisms, species richness is lowest in the Arctic and highest in the Caribbean. In each region, the distribution of species richness and occurrences is a log series. Consequently, the entire series of species occurrences is predicted by the single proportionality constant, α. After log series rarefaction, differences in species richness among areas are nearly all accounted for by species occurring ≤10 times. Most of the differences are accounted for by species occurring once, less by twice, and so on. For example, species occurring once account for 81% of the difference in species richness between the Atlantic and Caribbean, and those occurring once and twice account for 87% of the difference. Most rare species have no fossil record and most endemic species are rare. Probably most of these species evolved recently indicating more origination in species-rich areas. High origination might also be coupled with less extinction. Although each of the five regions can easily be distinguished by differences in composition, in all regions the 10 most abundantly occurring species exhibit nearly equal proportions of occurrences. No region is dominated by only one or two species. All regions exhibit the log series distribution, have nearly equal proportions for abundant species, and differ only in the number of rare species that coexist. Thus, from the point of view of the distribution of occurrences, the most striking aspect is the similarity among regions.     

Coexistence of a locally undifferentiated foraging guild: avian snatchers in a southeastern Australian forest

Three species of sexually dichromatic, insectivorous birds (Golden Whistler, Pachycephala pectoralis; Rufous Whistler, Pa. rufiventris; Rose Robin, Petroica rosea) employ a characteristic foraging manoeuvre (snatching or sally‐striking) disproportionately often, leading several studies to group them together in one foraging guild. The three species synchronously co‐occupy an extensive eucalypt forest east of Melbourne, Australia (Olinda State Forest). All assemble in relatively high densities in late winter and through spring and early summer to breed. Log‐linear modelling of use of foraging techniques, substrate, plant species and heights within the forest indicate that all six species–gender groups virtually do not differ from one another: a situation in which very similar species coexist at high densities. The results are considered in the context of interspecific competition among ecologically similar species and it is suggested how such a situation might arise and be sustained. The absence of gender‐specific differentiation within species is also addressed.