Thema der Dissertation:
Systematic Analysis of Enhancer RNAs for revealing enhancermediated gene expression regulation in metastatic melanoma Thema der Disputation:
Graph-based clustering: from data to communities
Systematic Analysis of Enhancer RNAs for revealing enhancermediated gene expression regulation in metastatic melanoma Thema der Disputation:
Graph-based clustering: from data to communities
Abstract: Graph-based clustering provides a flexible mathematical framework for detecting structure in high-dimensional biological data. In this talk, I will introduce the basic principles behind representing transcriptomic data as graphs, where vertices correspond to genes or samples and edges encode similarity, for example through correlation measures or nearest-neighbor relationships. I will then discuss how clustering can be formulated as a graph partitioning problem and explain the main ideas behind fundamental graph-clustering algorithms. The main application context will be bulk RNA sequencing, where graph-based approaches can be used to identify co-expressed gene modules, characterize tumor subtypes, and relate transcriptional patterns to biological pathways or clinical phenotypes. Alongside the theoretical concepts, I will briefly discuss practical aspects of the analysis workflow, including data normalization, graph construction, parameter choice, cluster validation, and biological interpretation. The aim is to show how graph-theoretic ideas can support robust and reproducible analysis of transcriptomic data while remaining connected to real biological questions.
Zeit & Ort
19.06.2026 | 10:00
Seminarraum 120
(Fachbereich Mathematik und Informatik, Arnimallee 3, 14195 Berlin)