A5072285340- Statistical Methods and Inference
- Gene expression and cancer classification
- Statistical Methods in Clinical Trials
- Statistical Methods and Bayesian Inference
- Genetic and phenotypic traits in livestock
- Machine Learning and Data Classification
- Data Analysis with R
- Sepsis Diagnosis and Treatment
- Advanced Statistical Methods and Models
- Health Systems, Economic Evaluations, Quality of Life
- Genomics and Phylogenetic Studies
- Ethics in Clinical Research
- Nutrition and Health in Aging
- Computational Drug Discovery Methods
- Economic and Environmental Valuation
- Hemodynamic Monitoring and Therapy
- Statistics Education and Methodologies
- Bioinformatics and Genomic Networks
- Simulation Techniques and Applications
- Head and Neck Cancer Studies
- Machine Learning and Algorithms
- Genetics, Bioinformatics, and Biomedical Research
- Cardiac, Anesthesia and Surgical Outcomes
- Advanced Causal Inference Techniques
- Environmental Impact and Sustainability
Friedrich-Alexander-Universität Erlangen-Nürnberg
2012-2024
Paul Ehrlich Institut
2017-2024
University of Southern Denmark
2023
Natural and Medical Sciences Institute
2023
University of Tübingen
2023
Zimmer Biomet (Germany)
2014-2020
Heinrich Heine University Düsseldorf
2019
Düsseldorf University Hospital
2019
University Hospital Münster
2019
Leibniz Institute DSMZ – German Collection of Microorganisms and Cell Cultures
2013-2015
Abstract Summary: opm is an R package designed to analyse multidimensional OmniLog® phenotype microarray (PM) data. provides management, visualization and statistical analysis of PM data, including curve-parameter estimation discretization, dedicated customizable plots, metadata automated generation textual tabular reports, mapping substrates databases, batch conversion files export phylogenetic software in the YAML markup language. Availability: distributed under GPL through Comprehensive...
Summary Generalized additive models for location, scale and shape (GAMLSSs) are a popular semiparametric modelling approach that, in contrast with conventional generalized models, regress not only the expected mean but also every distribution parameter (e.g. shape) to set of covariates. Current fitting procedures GAMLSSs infeasible high dimensional data set-ups require variable selection based on (potentially problematic) information criteria. The present work describes boosting algorithm...
Modern biotechnologies often result in high-dimensional data sets with many more variables than observations (n≪p). These pose new challenges to statistical analysis: Variable selection becomes one of the most important tasks this setting. Similar arise if modern from observational studies, e.g., ecology, where flexible, non-linear models are fitted data. We assess recently proposed flexible framework for variable called stability selection. By use resampling procedures, adds a finite sample...
The gathering of clinical data on fractures dental restorations through prospective trials is a labor- and time-consuming enterprise. Here, we propose an unconventional approach for collecting large datasets, from which information indirect can be retrospectively analyzed. authors accessed the database industry-scale machining center in Germany obtained 34,911 computer-aided design (CAD)/computer-aided manufacturing (CAM) all-ceramic posterior restorations. bridges, crowns, onlays, inlays...
Abstract Variable and model selection are of major concern in many statistical applications, especially high-dimensional regression models. Boosting is a convenient method that combines fitting with intrinsic selection. We investigate the impact base-learner specification on performance boosting as procedure. show variable may be biased if covariates different nature. Important examples models combining continuous categorical covariates, number categories large. In this case, least squares...
Summary Objectives: Component-wise boosting algorithms have evolved into a popular estimation scheme in biomedical regression settings. The iteration number of these is the most important tuning parameter to optimize their performance. To date, no fully automated strategy for determining optimal stopping has been proposed. Methods: We propose data-driven sequential rule algorithms. It combines resampling methods with modified version an earlier approach that depends on AIC-based information...
Generalized additive models for location, scale and shape are a flexible class of regression that allow to model multiple parameters distribution function, such as the mean standard deviation, simultaneously. With R package gamboostLSS, we provide boosting method fit these models. Variable selection choice naturally available within this regularized framework. To introduce illustrate gamboostLSS its infrastructure, use data set on stunted growth in India. In addition specification...
Spatial and temporal processes shaping microbial communities are inseparably linked but rarely studied together. By Illumina 16S rRNA sequencing, we monitored soil bacteria in 360 stations on a 100 square meter plot distributed across six intra-annual samplings managed, temperate grassland. Using multi-tiered approach, tested the extent to which stochastic or deterministic influenced composition of local communities. A combination phylogenetic turnover analysis null modeling demonstrated...
Little is known about the long-term success of non-drug therapies for treating dementia, especially whether effects are sustained after therapy ends. Here, we examined a one-year multimodal 10 months patients completed therapy. This randomised, controlled, single-blind, longitudinal trial involved 61 (catamnesis: n = 52) with primary degenerative dementia in five nursing homes Bavaria, Germany. The highly standardised intervention, MAKS, consisted motor stimulation, practice activities daily...
A novel Gram-reaction-positive, aerobic actinobacterium, tolerant to mitomycin C, heavy metals, metalloids, hydrogen peroxide, desiccation, and ionizing- UV-radiation, designated G18 T , was isolated from dolomitic marble collected outcrops in Samara (Namibia). The growth range 15–35°C, at pH 5.5–9.5 presence of 1% NaCl, forming greenish-black coloured colonies on GYM Streptomyces agar. Chemotaxonomic molecular characteristics the isolate matched those described for other representatives...
Boosting algorithms were originally developed for machine learning but later adapted to estimate statistical models—offering various practical advantages such as automated variable selection and implicit regularization of effect estimates. The interpretation the resulting models, however, remains same if they had been fitted by classical methods. Boosting, hence, allows use an advanced scheme types models. This tutorial aims highlight how boosting can be used semi-parametric modelling, what...
Flexible modeling frameworks for species distribution models based on generalized additive that allow smooth, nonlinear effects and interactions are of increasing importance in ecology. Commonly, the flexibility such smooth function estimates is controlled by means penalized estimation procedures. However, actual shape remains unspecified. In many applications, this not desirable as researchers have a priori assumptions estimated effects, with monotonicity being most important. Here we...
Bouldering psychotherapy (BPT) combines psychotherapeutic elements with physical activity (PA). It might be effective for reducing symptoms of depression, but so far, no study has assessed individuals' levels PA to control whether positive effects on depression can also found when adjusting participants' PA. This is important because itself been proven in and therefore an variable account - especially therapies using sport as one therapeutic mechanism.Using a waitlist group design,...