Dr. rer. nat. Hristo Todorov
Phone: +49 6131 39 20141
Statistical analysis of high-dimensional data
Small sample sizes pose a major challenge for the statistical analysis of data in basic research. Depending on the type of data (e.g. omics studies), sometimes the number of outcome measures is much greater than the number of measurements per variable (“great p, small n “ problem). Analysis of such data requires methods which are able to deal with their multi-dimensional nature. While principal component analysis (PCA) has emerged as the most popular dimensionality reduction technique, numerous other methods might be more appropriate depending on the type of data and research question. In my research, I am interested in which ordination techniques lead to robust inferences in data sets with more variables than measurements and which methods offer increased sensitivity to detect group differences.
I mainly focus on analyzing behavioral outcome measures to identify endophenotypes after stress exposure, community profiling of commensal microbiota using 16S rRNA sequencing, and applying and adapting statistical techniques for integrating multi-omics data.