Focus
The group's research is organized into the following key areas, which are reflected in our projects and publications.
RNA Bioinformatics | Cheminformatics | Synthetic Biology | Models of Evolution | AI in Bioinformatics
RNA Bioinformatics
For a long time, RNA was primarily viewed as a simple information carrier for protein synthesis. However, recent years have revealed a "zoo" of diverse, functional RNA molecules that play critical roles in regulating essential cellular processes like transcription, translation, mRNA processing, and modification.
Furthermore, RNA is now recognized as a promising therapeutic target or drug itself, exemplified by mRNA vaccines. Our research focuses on developing sophisticated algorithms to detect RNA genes and to predict their structure and dynamics across various levels of abstraction. Current projects address, e.g., the role of posttranscriptional RNA modification and determinants of mRNA translation efficiency.
Cheminformatics
Cheminformatics applies advanced computer science methodologies to predict the properties of molecules, chemical reactions, and reaction networks. We specialize in developing graph grammar approaches to formalize chemical reactions. We apply this methodology to a broad set of (bio)chemical problems, including synthesis planning and the evolution of catalyzed reaction networks.
Synthetic Biology
Synthetic biology aims to manipulate existing biological systems to achieve desired properties or to design entirely novel biological building blocks and systems that do not occur in nature. For instance, one might modulate a metabolic pathway to significantly enhance the production of a desired enzyme or metabolite. Leveraging our expertise in RNA structure, we design artificial RNAs with regulatory functions, such as synthetic riboswitches.
Models of Evolution
We develop and apply computational models to understand the mechanisms and dynamics of biological and chemical evolution. Our research ranges from modeling the evolution of sequences under structural constraints, such as in structural RNA families, to simulating the prebiotic precursors of metabolism.
AI in Bioinformatics
Across all our research topics, we employ a dual approach, utilizing both traditional, often biophysics-based algorithms and cutting-edge machine learning techniques, including deep learning. Our focus goes beyond maximizing prediction performance on benchmarks. We aim to understand what is easy or hard for a neural network to learn and use these learning approaches as a tool to uncover novel biological mechanisms.