Four open Master’s Theses

Supervisor: Jürgen Zanghellini | Department of Analytical Chemistry, University of Vienna

Background

Genome-scale metabolic models (GEMs) are computational representations of an organism’smetabolic network, linking genes, reactions, and metabolites to enable simulation, hypothesis testing, and metabolic engineering. To ensure usability and reproducibility, GEMs are typically published in standardized formats such as SBML or MATLAB (.mat) files.

However, many metabolic reconstructions are instead shared as Excel spreadsheets, which lack the structure required for computational analysis. Converting these files into functional GEMs requires extensive manual curation, including data cleaning, standardization, and validation. This process is time-consuming and often neglected, rendering many models effectively unusable for the broader scientific community.

This project aims to develop an agentic AI-based workflow to automate the reconstruction of GEMs from Excel files. By leveraging intelligent agents to interpret, standardize, and validate metabolic data, the approach will restore accessibility to previously unusable models and enable their integration into existing computational frameworks.

Aims of the project

This project includes the following aims:

1. Collecting GEMs published in unusable formats, like excel, from literature
2. Set up an agentic AI based workflow to reconstruct usable GEMs
3. Validate curated GEMs using state of the art tools in metabolic modeling
4. Provide an open source database with all curated GEMs

The duration of this project is set to 9 months.

What we offer

• Great working atmosphere in a young and dynamic group working on computational biology
• Access to the state-of-the-art computational environment
• Co-supervision by a PhD student and a postdoctoral researcher
• Participation in group meetings and discussions about projects within the group or beyond

For applications, please contact juergen.zanghellini@univie.ac.at.

Supervisor: Jürgen Zanghellini | Department of Analytical Chemistry, University of Vienna

Background

Adipocytes are broadly classified into white, beige, and brown cells. While white adipocytes primarily function in lipid storage, beige and brown adipocytes exhibit increased metabolic activity through thermogenesis. Although significant progress has been made in studying the conversion of white to beige adipocytes, a systems-level understanding of their metabolic differences using genome-scale metabolic models (GEMs) remains limited.

GEMs are computational frameworks that represent the full set of metabolic reactions within a cell, linking genes, enzymes, and metabolites. They enable the simulation and analysis of cellular metabolism under different conditions, providing insights into metabolic states and regulatory mechanisms.

This project aims to address the current knowledge gap by reconstructing adipocyte-specific GEMs using published time-resolved multi-omics datasets. Building on themost recent human GEM, the approach will integrate multi-omics data to capture dynamic metabolic changes during adipocyte differentiation. This will provide a systems-level understanding of adipocyte metabolism and its role in thermogenic activation.

Aims of the project

This project includes the following aims:

1. Reconstruct state of the art adipocyte-specific GEMs
2. Comparison to already established GEMs of adipocytes
3. Use dynamic flux balance analysis to simulate time-dependent changes of adipocytes

The duration of this project is set to 9 months.

What we offer

• Great working atmosphere in a young and dynamic group working on computational biology
• Access to the state-of-the-art computational environment
• Co-supervision by a postdoctoral researcher
• Participation in group meetings and discussions about projects within the group or beyond

For applications, please contact juergen.zanghellini@univie.ac.at.

Supervisor: Jürgen Zanghellini | Department of Analytical Chemistry, University of Vienna

Background

Food authentication has become increasingly important due to widespread mislabeling and adulteration, particularly in high-value products such as honey and olive oil. These products are often diluted with arti- ficial substances or falsely labeled in terms of geographical origin, undermining consumer trust and regu- latory standards.

Omics technologies provide a powerful approach to address this challenge by enabling detailed molecular characterization of food samples. Genomics can reveal phylogenetic signatures linked to production environments, while metabolomics captures chemical profiles reflective of composition and origin.

This project aims to leverage these data types in combination with machine learning methods to de- velop robust computational tools for food authentication. By training models on publicly available datasets, the approach will enable automated classification of authenticity and geographic origin for honey and olive oil samples.

Ultimately, the project seeks to provide scalable, data-driven solutions to detect food fraud and support quality control in the food industry.

Aims of the project

This project includes the following aims:

1. Collect publicly available multi omics data of honey and olive oil samples
2. Set up libraries for authentic honey and olive oil based on the geographic origin using genomics and metabolomics
3. Develop ML tools for classification based on origin and authenticity

The duration of this project is set to 9 months.

What we offer

• Great working atmosphere in a young and dynamic group working on computational biology
• Access to the state-of-the-art computational environment
• Co-supervision by a postdoctoral researcher
• Participation in group meetings and discussions about projects within the group or beyond

For applications, please contact juergen.zanghellini@univie.ac.at.

Supervisor: Jürgen Zanghellini | Department of Analytical Chemistry, University of Vienna

Background

Genome-scale metabolic models (GEMs) are computational representations of an organism’s metabolic network, integrating known biochemical reactions and gene–protein–reaction associations to enable systemlevel analysis of metabolism.

Within a GEM, flux sampling is the gold-standard approach, used to explore the range of feasible flux distributions under given constraints. Despite its potential, flux sampling faces several challenges: existing methods can produce thermodynamically infeasible loops in flux distributions, suffer from poor convergence, and often depend strongly on the structure and size of the solution space.

In addition, flux sampling for community GEMs remains less established and requires systematic benchmarking. The aim of this project is to implement and evaluate currently available flux sampling tools and to investigate alternative sampling approaches using well-established GEMs.

Aims of the project

This project includes the following aims:

1. Collecting GEMs published in unusable formats, like excel, from literature
2. Set up an agentic AI based workflow to reconstruct usable GEMs
3. Validate curated GEMs using state of the art tools in metabolic modeling
4. Provide an open source database with all curated GEMs

The duration of this project is set to 9 months.

What we offer

• Great working atmosphere in a young and dynamic group working on computational biology
• Access to the state-of-the-art computational environment
• Co-supervision by a PhD student and a postdoctoral researcher
• Participation in group meetings and discussions about projects within the group or beyond

For applications, please contact juergen.zanghellini@univie.ac.at.

Join us!

Join us!

We extend an invitation to Postdoctoral, Doctoral, Master's, and Bachelor's students with interest in bio/cheminformatics, network analysis, and computational biology.

• Bio/cheminformatics
  • computational mass spectrometry
  • high-performance computing
• Network analysis (constraint-based)
  • modeling of metabolism
  • multi-omics integration
  • rational cell factory design
  • optimal fermentation design
• Computational biology
  • theory of microbial (community) growth

Broaden your expertise, create new knowledge, and excel as a member of our research group. To take the next step, please send your application to juergen.zanghellini@univie.ac.at.