Genome-Wide Regression and Prediction with the BGLR Statistical Package

Many modern genomic data analyses require implementing regressions where the number of parameters (p, e.g., the number of marker effects) exceeds sample size (n). Implementing these large-p-with-small-n regressions poses several statistical and computational challenges, some of which can be confronted using Bayesian methods. This approach allows integrating various parametric and nonparametric shrinkage and variable selection procedures in a unified and consistent manner. The BGLR R-package implements a large collection of Bayesian regression models, including parametric variable selection and shrinkage methods and semiparametric procedures (Bayesian reproducing kernel Hilbert spaces regressions, RKHS). The software was originally developed for genomic applications; however, the methods implemented are useful for many nongenomic applications as well. The response can be continuous (censored or not) or categorical (either binary or ordinal). The algorithm is based on a Gibbs sampler with scalar updates and the implementation takes advantage of efficient compiled C and Fortran routines. In this article we describe the methods implemented in BGLR, present examples of the use of the package, and discuss practical issues emerging in real-data analysis.     

A microsatellite genetic map of the turbot (Scophthalmus maximus)

A consensus microsatellite-based linkage map of the turbot (Scophthalmus maximus) was constructed from two unrelated families. The mapping panel was derived from a gynogenetic family of 96 haploid embryos and a biparental diploid family of 85 full-sib progeny with known linkage phase. A total of 242 microsatellites were mapped in 26 linkage groups, six markers remaining unlinked. The consensus map length was 1343.2 cM, with an average distance between markers of 6.5 +/- 0.5 cM. Similar length of female and male maps was evidenced. However, the mean recombination at common intervals throughout the genome revealed significant differences between sexes, approximately 1.6 times higher in the female than in the male. The comparison of turbot microsatellite flanking sequences against the Tetraodon nigroviridis genome revealed 55 significant matches, with a mean length of 102 bp and high sequence similarity (81-100%). The comparative mapping revealed significant syntenic regions among fish species. This study represents the first linkage map in the turbot, one of the most important flatfish in European aquaculture. This map will be suitable for QTL identification of productive traits in this species and for further evolutionary studies in fish and vertebrate species.     

Genomewide analysis of admixture and adaptation in the Africanized honeybee

Genetic exchange by hybridization or admixture can make an important contribution to evolution, and introgression of favourable alleles can facilitate adaptation to new environments. A small number of honeybees (Apis mellifera) with African ancestry were introduced to Brazil ~60 years ago, which dispersed and hybridized with existing managed populations of European origin, quickly spreading across much of the Americas in an example of a massive biological invasion. Here, we analyse whole‐genome sequences of 32 Africanized honeybees sampled from throughout Brazil to study the effect of this process on genome diversity. By comparison with ancestral populations from Europe and Africa, we infer that these samples have 84% African ancestry, with the remainder from western European populations. However, this proportion varies across the genome and we identify signals of positive selection in regions with high European ancestry proportions. These observations are largely driven by one large gene‐rich 1.4‐Mbp segment on chromosome 11 where European haplotypes are present at a significantly elevated frequency and likely confer an adaptive advantage in the Africanized honeybee population. This region has previously been implicated in reproductive traits and foraging behaviour in worker bees. Finally, by analysing the distribution of ancestry tract lengths in the context of the known time of the admixture event, we are able to infer an average generation time of 2.0 years. Our analysis highlights the processes by which populations of mixed genetic ancestry form and adapt to new environments.     

Effects of an adjunctive,chronotype-based light therapy in hospitalized patients with severe burnout symptoms - a pilot study

Light therapy is a well-established treatment option for seasonal affective disorders and is effective in reducing sleep problems and daytime fatigue. Symptoms of severe burnout include feelings of exhaustion and impaired sleep and mood. Thus, light therapy seems promising for burnout treatment. So far, light therapy effects in burnout were investigated in outpatient settings only, with inconclusive results. The present study targeted light therapy effects in an inpatient setting. Participants with severe burnout were recruited in two psychosomatic clinics and randomly assigned to a control group with multimodal psychiatric treatment or an add-on light treatment group. Participants in the latter group were additionally exposed to morning bright light (illuminance: 4246 lux, irradiance: 1802.81 µW.cm^?2) for 3 weeks, 30 minutes a day, timed to their chronotypes. Light effects on burnout symptoms, depression, well-being, daytime sleepiness, sleep quality, and attentional performance were measured twice (pre-/postintervention design). Adjunctive chronotype-based bright light therapy was well tolerated and improved burnout symptoms and well-being without additional effect on severity of depression. Furthermore, reduced daytime sleepiness, improved nighttime sleep quality, a sleep phase advance of 25 minutes, shortened sleep latency, less sleep disturbances and increased sleep duration were observed in the light treatment group. No group differences were found in attentional performance. Chronotype-based bright light therapy seems to be effective in improving burnout symptoms and sleep problems in patients with severe burnout symptoms. Further studies with larger sample sizes and objective measures of sleep are necessary to confirm these preliminary results before practical recommendations can be made.     

Multitrait Bayesian shrinkage and variable selection models with the BGLR-R package

The BGLR-R package implements various types of single-trait shrinkage/variable selection Bayesian regressions. The package was first released in 2014, since then it has become a software very often used in genomic studies. We recently develop functionality for multitrait models. The implementation allows users to include an arbitrary number of random-effects terms. For each set of predictors, users can choose diffuse, Gaussian, and Gaussian–spike–slab multivariate priors. Unlike other software packages for multitrait genomic regressions, BGLR offers many specifications for (co)variance parameters (unstructured, diagonal, factor analytic, and recursive). Samples from the posterior distribution of the models implemented in the multitrait function are generated using a Gibbs sampler, which is implemented by combining code written in the R and C programming languages. In this article, we provide an overview of the models and methods implemented BGLR’s multitrait function, present examples that illustrate the use of the package, and benchmark the performance of the software.     

Binomial regression with misclassification

Motivated by a study of human papillomavirus infection in women, we present a Bayesian binomial regression analysis in which the response is subject to an unconstrained misclassification process. Our iterative approach provides inferences for the parameters that describe the relationships of the covariates with the response and for the misclassification probabilities. Furthermore, our approach applies to any meaningful generalized linear model, making model selection possible. Finally, it is straightforward to extend it to multinomial settings.     

Modeling tight junction dynamics and oscillations

Tight junction (TJ) permeability responds to changes of extracellular Ca(2+) concentration. This can be gauged through changes of the transepithelial electrical conductance (G) determined in the absence of apical Na(+). The early events of TJ dynamics were evaluated by the fast Ca(2+) switch assay (FCSA) (Lacaz-Vieira, 2000), which consists of opening the TJs by removing basal calcium (Ca(2+)(bl)) and closing by returning Ca(2+)(bl) to normal values. Oscillations of TJ permeability were observed when Ca(2+)(bl) is removed in the presence of apical calcium (Ca(2+)(ap)) and were interpreted as resulting from oscillations of a feedback control loop which involves: (a) a sensor (the Ca(2+) binding sites of zonula adhaerens), (b) a control unit (the cell signaling machinery), and (c) an effector (the TJs). A mathematical model to explain the dynamical behavior of the TJs and oscillations was developed. The extracellular route (ER), which comprises the paracellular space in series with the submucosal interstitial fluid, was modeled as a continuous aqueous medium having the TJ as a controlled barrier located at its apical end. The ER was approximated as a linear array of cells. The most apical cell is separated from the apical solution by the TJ and this cell bears the Ca(2+) binding sites of zonula adhaerens that control the TJs. According to the model, the control unit receives information from the Ca(2+) binding sites and delivers a signal that regulates the TJ barrier. Ca(2+) moves along the ER according to one-dimensional diffusion following Fick's second law. Across the TJ, Ca(2+) diffusion follows Fick's first law. Our first approach was to simulate the experimental results in a semiquantitative way. The model tested against experiment results performed in the frog urinary bladder adequately predicts the responses obtained in different experimental conditions, such as: (a) TJ opening and closing in a FCSA, (b) opening by the presence of apical Ca(2+) and attainment of a new steady-state, (c) the escape phase which follows the halt of TJ opening induced by apical Ca(2+), (d) the oscillations of TJ permeability, and (e) the effect of Ca(2+)(ap) concentration on the frequency of oscillations.     

Comparative cytogenetics of three species of Dichotomius (Coleoptera, Scarabaeidae)

Meiotic and mitotic chromosomes of Dichotomius nisus, D. semisquamosus and D. sericeus were analyzed after conventional staining, C-banding and silver nitrate staining. In addition, Dichotomius nisus and D. semisquamosus chromosomes were also analyzed after fluorescent in situ hybridization (FISH) with an rDNA probe. The species analyzed had an asymmetrical karyotype with 2n = 18 and meta-submetacentric chromosomes. The sex determination mechanism was of the Xy(p) type in D. nisus and D. semisquamosus and of the Xy (r) type in D. sericeus. C-banding revealed the presence of pericentromeric blocks of constitutive heterochromatin (CH) in all the chromosomes of the three species. After silver staining, the nucleolar organizer regions (NORs) were located in autosomes of D. semisquamosus and D. sericeus and in the sexual bivalent of D. nisus. FISH with an rDNA probe confirmed NORs location in D. semisquamosus and in D. nisus. Our results suggest that chromosome inversions and fusions occurred during the evolution of the group.