Interfakultäres Institut für Biomedizinische Informatik (IBMI)

Bachelorarbeiten

Generell können Sie sich bei unseren Arbeitsgruppen direkt nach möglichen Themen für Ihre Bachelorarbeit erkundigen. Die meisten Themen werden in direktem Gespräch entwickelt.
Darüber hinaus sind derzeit folgende konkrete BSc-Themen unter folgendem Link im Angebot: https://caura.cs.uni-tuebingen.de/index.php/s/X6pxCExfJB6sNrx?dir=/&editing=false&openfile=true

Prof. Dr. Carsten Eickhoff

Lab Name: Health NLP Lab
Webseite: https://health-nlp.org

Kontakt: carsten.eickhoff (at) uni-tuebingen.de

Forschungsstichpunkte:

  • Natural Language Processing
  • Explainable AI
  • AI Safety
  • Clinical AI

Abschlussarbeiten in 2025:

  • Agentic Document Textualization with Small Vision-Language Models: An Agentic Pipeline for Scientific Visual Content
  • Information Extraction from Breast Cancer Reports for Tumor Boards
  • Iterative Retrieval Augmented Generation for Multi-Hop Question Answering
  • Using Machine Learning to predict adverse events in patients over 65 years undergoing cardiac surgery
  • Christmas is coming bald: Exploring Representations of True Cognates and False Friends in Multilingual LLMs
  • Machine learning based investigation of physical activity in mid-life parents: A multi-factorial approach
  • Privacy in Medical AI: Evaluating Risks and Countermeasures for Protecting Patient Information
  • Information Extraction from Radiation Oncology Reports
  • From Geometry to Performance - Analyzing Embedding Space Structure in Large Language Models
  • Understanding Unreliability of Steering Vectors in Language Models
  • TruthAlign: Enhancing Fact-Checking Using Conflicting Words
  • SPOT: Sparse Parameter-efficient Outlier-guided Tuning
  • Contrastive Attribution Learning in Explainable Neural Text Classification
  • Developing an Intelligent Search Engine for Cancer Therapy and Medication Documents Using Lucene and Advanced Relevance Algorithms

Prof. Dr. Daniel Huson

Kontakt: daniel.huson (at) uni-tuebingen.de

Forschungsstichpunkte:

  • Phylogenetics
  • Metagenomics
  • Algorithms
  • Software

Bachelorarbeiten in 2025:

  • WebSplits - Platform for Phylogenetic Trees and Networks
  • Extracting phylogenies from images using AI
  • Development of pipeline for gene tree calculation from annotated long read MAGs and their visualization
  • Evaluating the utility of DuckDB for on-demand querying and efficient storage of metagenomic alignments
  • Assessing the viability of SQLite in the context of on-demand alignment for large genomic datasets
  • Visualization of Phylogenetic Trees in Phylogenetic Networks

Prof. Dr.-Ing. Oliver Kohlbacher

Kontakt: oliver.kohlbacher (at) uni-tuebingen.de

Forschungsstichpunkte:

  • Translational bioinformatics
  • Computational mass spectrometry
  • Personalized medicine
  • Research data infrastructures

Bachelorarbeiten in 2025:

  • PDB2Net: Automated extraction of protein interaction networks from Three-Dimensional Structure
  • Pipeline Entwicklung für Local Energy Decomposition (LED) mit Orca

Dr. Thales Kronenberger

Our group would be fine with either bachelor's or master's theses, as long as the students write their thesis in English.
We offer thesis topics on infection biology, computational drug discovery and method development for molecular dynamics and simulations. We are happy to discuss specifics over coffee and would appreciate it if the students sent a short motivation letter and CV.

Contact: thales.kronenberger (at) uni-tuebingen.de

Jun.-Prof. Dr. Thomas Küstner

Unsere Forschungsthemen:

  • AI for Medical Imaging (MRI, CT, PET)
  • Multimodal Data Integration (Imaging + Clinical Data)
  • Clinical Applications & Large-Scale Cohort Studies (NAKO, UK Biobank) 

Informationen über aktuelle Themen finden Sie hier: https://www.medizin.uni-tuebingen.de/de/das-klinikum/einrichtungen/kliniken/radiologie/allgemeine-radiologie/forschung/ag-midas/teaching

Bitte senden Sie Ihre Bewerbung mit Angabe Ihres gewünschten Fachgebiets / Ihrer gewünschten Fachgebiete, Lebenslauf und Transkript an: midaslabthesis (at) gmail.com

Bachelorarbeiten in 2025:

  • Investigation of biological age in large-scale population studies
  • The aging heart: Associations between cardiac structure, function, and demographic factors in a population-based study

Prof. Dr. Andrei Lupas

Forschungsstichpunkte:

  • Protein sequence and structure analyses
  • Machine learning approaches to protein structure prediction and design
  • Prediction and classification of protein-protein interactions

Kontakt: andrei.lupas (at) tuebingen.mpg.de

Bitte fügen Sie bei Ihrer Anfrage nach einem Thema für Ihre Abschlussarbeit Ihren Lebenslauf, Transkripte und eine Liste der Methoden bei, die Sie ohne intensive Betreuung durchführen können.

Bachelorarbeiten in 2025:

  • Integrating Evolutionary Coupling and AlphaFold for the Discovery of Protein–Protein Interactions
  • Assessing AlphaFold For Protein-Protein Interactions Involving Intrinsically Disordered Regions
  • Functional Contextualization of the Signal Transduction Model Protein Af1503 by Interaction and Homolog Screening

Prof. Dr. Kay Nieselt

Link: https://uni-tuebingen.de/de/140699

Forschungsstichpunkte:

  • Analysis and visualization of genomics
  • Transcriptomics and/or proteomics data
  • Analysis of (ancient) (pan-)genomic data, of expression data and of the transcriptome's architecture of prokaryotes
  • Development of machine learning methods for dimensionality reduction and for analysis of proteomics data.

Vergangene Bachelorarbeiten

Prof. Dr. Stephan Ossowski

Prof. Dr. Detlef Weigel

Forschungsstichpunkte:

  • Genome graphs
  • Pangenomes
  • Hyperdiverse regions of the genome

Abschlussarbeiten in 2025:

  • Genomic diversity of a local blast fungus population in Italy
  • Genomic hypervariability of the essential gene P5CR in Arabidopsis thaliana
  • Identification of novel gene duplication events in Arabidopsis thaliana genomes

Bewerbung ist lt. Webseite des Max-Planck-Instituts direkt an die betreffende Gruppenleiterin / den betreffenden Gruppenleiter zu senden.