CCBIO Seminar – Franziska Görtler
Welcome to the CCBIO seminar series in the spring term of 2025. Open to all in the auditorium in Armauer Hansens Hus. No registration necessary. Speaker is Franziska Görtler from the Computational Biology Unit, UiB/Department of Oncology and Medical Physics, Haukeland University Hospital. Topic is "Digital deconvolution of bulk gene expression data."
Main content
Speaker: Franziska Görtler, bioinformatician from the Computational Biology Unit, UiB / Department of Oncology and Medical Physics, Haukeland University Hospital
Title: Digital deconvolution of bulk gene expression data.
Host: Vladan Milosevic
Where: the auditorium in Armauer Hansens Hus
When: June 12, 2025 at 14.30-15.30
No registration necessary.
Abstract: The talk will focus on the challenges in determining the cell type populations in bulk RNASeq probes like closely related cell types, small cell type proportions, hidden background contributions as well as cell type-specific regulations due to disease. We will present our three deconvolution algorithms DTD, ADTD and HIDE, providing machine learning solutions. We will verify our algorithms both on simulated data as well as on bulk RNASeq data from the Cancer Genome Atlas.
Finally, we will introduce Deconomix – a comprehensive toolbox for the cell-type deconvolution of bulk transcriptomics data based on our algorithms, available as a Python package and as a standalone graphical user interface for easy and user-friendly access.
Franziska Görtler, PhD, is since 2023 a researcher at the Department of Oncology and Medical Physics, Haukeland University Hospital, with Prof. O. Straume's and Prof. J. Lorens' groups. Her research mainly focuses on algorithm development for immune cell deconvolution in tissues. She was 2024 awarded FRIPRO funding for early career researchers from the Research Council of Norway. Through loss-function learning optimisation, she investigates the dissection of the cellular distributions and the cell-type specific regulation patterns in spatial transcriptomics data.
She is the responsible bioinformatician in a randomised phase II clinical trial study on melanoma, focused on biomarker detection, machine learning algorithm development, and single-cell and spatial transcriptomics data. She did her postdoc in the Computational Biology Unit, Department of Biological Sciences, UiB, and her PhD in Bioinformatics at the University of Regensburg, Germany, on Loss-Function Learning for Digital Tissue Deconvolution.