We are interested in understanding the role of microbiomes in the evolution of animals

Adaptation to the environment is a major driving force in evolution and shapes the traits and properties of all organisms. Therefore, understanding adaptation is crucial to understand why organisms are the way they are.

For a better understanding of how adaptation shapes life on our planet, it is essential to consider the role microbes play in the evolution of multicellular organisms because all organisms live in a world immersed in microbes. On the one hand, the strong selection pressures exerted by pathogenic microbes drive the evolution of host defense. On the other hand, microbes can also facilitate adaptation of the host to the environment. A prominent example are the wood digesting microbes in the termite gut that allow termites to live and thrive on wood.

Although the ubiquitous importance of microbes in the evolution of higher organisms is starting to be realized, we still know very little about it. This is why we are interested in studying the role of microbes in evolution and especially adaptation of animals. We do this in two model systems:


Bacterial thiamine affects fly fitness. The genes for thiamine production vary between closely related bacterial strains and show evidence for horizontal gene transfer BMC Biology

Drosophila melanogaster lives in a microbe rich environment: rotting fruit. It is known that microbes influence growth rate and cold tolerance in fruit flies. Obviously, these traits can be relevant for fly fitness. Because microbes impact fitness relevant traits, natural selection can act on fruit flies to preferentially associate with microbes that influence these traits in a way that is beneficial for them. On the other hand, there is evidence that the evolution of microorganisms could also be affected by fruit flies: fruit flies are highly mobile and can disperse microorganisms. Dispersal is highly relevant for microorganisms that colonize the same natural food source as fruit flies because rotting fruit can easily dry up or be consumed and hence become a dead end for less mobile microbes. Therefore, associating with flies can be an adaptive trait. However, this might be a double-edged sword: microbes also serve as fly food.

We would like to find out how the interaction between fly hosts and the microbiome shape the adaptive dynamics of the partners. To this end, we combine genetic, population genomic, metagenomic and metatranscriptomic approaches. Our focus lies on studying the evolutionary dynamics of natural host populations and microbiomes because the natural environment is where species evolve. Therefore, we collaborate with the European Drosophila Population Genomics Consortium (

(A) The European Drosophila melanogaster microbiome is structured on a continental scale. (B) Host genetic differences can predict bacterial community dissimilarity


Cryptotermes secudus in a block of wood

The association of termites with their gut symbionts has attracted interest for over a century and serves as a text book example of mutualistic symbiosis. It is also an important example of microbiome mediated adaptation to a mainly wood-based diet. Termites have further adapted to different lifestyles and diets. The termite microbiome is directly involved in nutrient acquisition from the diet. Hence, it seems only reasonable to assume that the microbiome could have played a role in lifestyle and diet adaptation. We use comparative genomic and metagenomic analyses to search for signals of microbiome mediated adaptation.

Furthermore, the termite-microbe symbiosis represents a rare case of a symbiosis between a metazoan and a eukaryotic microorganism. While we have learned a lot about the molecular mechanisms underlying the recognition of prokaryotic symbionts by their hosts in recent years, very little is still known about signals and recognition in metazoan-eukaryote symbioses. We combine metagenomic, transcriptomic, and comparative genomic techniques as well as functional genetic analysis (e.g. RNAi) to better understand the symbiont recognition mechanisms and their evolution. We collaborate with Judith Korb, a world leading expert in termite biology, for our termite projects.

The metagenome of foraging termite species from the genus Reticulitermes is enriched for pathways that allow the acquisition of nitrogen through nitrate from soil

Human genetic variation in forensics

A recent cabinet decision to reform the ‘Strafprozessordnung’ as well as the revision of the ‘Polizeiaufgabengesetz’ in Bavaria propose to allow almost unrestricted use of crime scene DNA. One goal is to determine phenotypic traits (skin color, eye color, hair color) and biogeographic ancestry of potential suspects from DNA. Reading about very high predictive power of these methods in the law bill and in the media (e.g. 99.9% for biogeographic ancestry), I became interested in the scientific basis for these applications. Together with, Veronika Lipphardt (Science and Technology Studies, Uni-Freiburg), Peter Pfaffelhuber (Department of Statistics, Uni-Freiburg), and others, I work in a multidisciplinary team ( to address the scientific basis as well as the societal consequences of the technology. The goal is to help to develop a scientific road-map for a responsible use of DNA phenotyping and biogeographic ancestry testing in forensics.

Inference of recent admixture using genotype data

The inference of biogeographic ancestry (BGA) has become a focus of forensic genetics. Misinference of BGA can have profound unwanted consequences for investigations and society. We show that recent admixture can lead to misclassification and erroneous inference of ancestry proportions. Subsequently, we extended existing tools for estimation of individual ancestry (IA) by allowing for different IA in both parents, leading to estimates of parental individual ancestry (PIA).

(A) population model (B) top panel: inferred ancestry lower panel: maternal and paternal ancestry

Analysis of off-target phenotypes for markers used in forensic genetics

We analyzed which health-related phenotypes can be inferred from standard methods for DNA phenotyping and biogeographic ancestry analysis that were not intended to be analyzed by these methods and are thus considered ‘off-target’ . We are also currently developing methods to analyze human genetic variation to avoid some of the pitfalls of biogeographic ancestry analysis.

Some DNA markers used in forensic biogeographic ancestry analysis and DNA phenotyping reveal information on health-related, off-target phenotypes like cancer predisposition or asthma predisposition.