Research

We study rapid adaptation in the waterflea Daphnia. We are following the adaptive process over different organisational levels from the genome over molecular mechanisms to the phenotype, and from the individual animal to populations and ecosystems. We have a particular focus on tracking evolution over time, in natural populations as well as in the laboratory using experimental evolution.

Our current projects address specific questions within the following research themes:

Adaptation to toxic cyanobacteria in the Daphnia longispina species complex

Toxic cyanobacteria, especially when they occur in large abundances during the so-called "algal bloom", pose a threat to aquatic ecosystems and human health. Daphnia are known to rapidly adapt to toxic cyanobacteria. This even has practical applications, as these tolerant Daphnia lines provide a means of containment of such cyanobacteria mass developments.

Here, we investigate the genomic and molecular basis of cyanobacterial tolerance in Daphnia over time and space. We analyse genomic time series and study local adaptation in organisms from habitats with different cyanobacterial loads.

What are important factors determining if a population can evolve rapidly?

We are interested in detecting the genomic base of rapid adaptive processes in natural

populations experiencing environmental change. For this, we depend on time-series data

either generated forward in time by regular sampling of contemporary Daphnia populations,

or by going back in time by analysing resting stages extracted from sediment cores. Here,

the spatial analysis of lakes is also important, as it can serve as a source of replication

in natural systems. We are also interested in the role of fluctuating selection in natural

habitats and its interaction with environmental change. This approach is supplemented by

evolutionary modelling to get an idea about the role of factors like e.g. the number of

genes involved (few vs many) or the shape of the change (constant vs fluctuating) and to

form hypothesis that can then be tested in the laboratory with experimental evolution, i.e. under very controlled conditions.

How do zooplankton communities rapidly adapt to salt stress?

In a collaborative project with colleagues from the Netherlands, we study rapid adaptation to salinisation, another anthropogenic-induced stressor, in a big experimental evolution setup. Here, we analyze the response of three different Daphnia species, both, for individual populations as well as for a community consisting of all three species. We apply population genomics as well as a computer vision machine learning approach to study the genomic and phenotypic bases of salt tolerance in a lake community.

How can we study adaptive trait evolution with high-throughput phenotyping and machine learning?

In addition to genomics and population genomics, studying adaptive phenotypes to understand

ecology and evolution is a key part of our work. However, in contrasts to the recent advances in sequencing for the high-throughput generation of genomic data, high-throughput phenotyping of animals is still in its infancy. To overcome this phenotyping bottleneck, we are working on machine learning approaches for the high-throughput analysis of images and videos to identify, track and count animals as well as to analyse morphological features in more detail.