Computational Tools for Genome Mining of Antibiotics in bacteria and fungi

Advisor: Nadine Ziemert

Location: University of Tübingen

The Ziemert Lab at the University of Tübingen investigates the genomic diversity and evolution of natural product biosynthesis in bacteria and fungi. The group combines computational and experimental approaches to discover new antibiotics and other bioactive compounds. Its research integrates comparative genomics, pangenomics, metagenomics, and genome mining with molecular biology and analytical chemistry to link biosynthetic gene clusters (BGCs) to their chemical products and biological functions. The lab develops and maintains widely used computational tools and databases for genome mining and collaborates closely with microbiologists and chemists to validate predictions and explore the ecological and therapeutic roles of microbial natural products

Project description

Natural products remain one of our most powerful sources of antibiotics, yet the vast biosynthetic potential encoded in bacterial and fungal genomes is still largely untapped. This PhD project aims to develop and modernize computational tools that enhance the discovery of novel bioactive compounds through genome mining.

A central focus will be the further development of ARTS (Antibiotic Resistant Target Seeker), a tool that applies target-directed genome mining to identify biosynthetic gene clusters (BGCs) associated with self-resistance mechanisms. The project will update and expand ARTS using modern computational methods, including:

Microbes as Chemists. Microbes produce a variety of chemical compounds as secondary metabolites. We use them as so called natural products in agriculture, biotechnology, and especially in medicine. — **Microbes as Chemists.** Microbes produce a variety of chemical compounds as secondary metabolites. We use them as so called natural products in agriculture, biotechnology, and especially in medicine.

**Microbes as Chemists.** Microbes produce a variety of chemical compounds as secondary metabolites. We use them as so called natural products in agriculture, biotechnology, and especially in medicine.

Pangenome-based models to refine the detection of core and accessory genes and to better identify lineage-specific resistance signatures;
Protein language models and structure-based comparisons (e.g., AlphaFold-derived metrics) to detect functional divergence beyond sequence similarity;
Machine-learning approaches for improved prediction of resistance genes, targets, and potential modes of action;
Integration of metagenomic and fungal data to broaden applicability across microbial domains.

The project contributes to an interdisciplinary effort to create next-generation genome-mining pipelines that integrate evolutionary, structural, and functional information for more accurate and interpretable prediction of antibiotic-producing gene clusters.

More information about the research of Nadine Ziemert and a selection of recent publications can be found on her faculty page.

To apply

Check that you meet our requirements
Familiarize yourself with the application process and have a look at the Frequently Asked Questions
Register to apply and complete the application form

Application deadline: 19 January 2026