kommt noch

Achtung: Die Navigationslinks auf dieser Seite funktionieren nicht, weil das nicht der Rahmen ist, der zu diesem Skript gehört. Wenn das Skript ordnungsgemäß dort installiert ist, wo es später laufen soll, funktionieren auch die Links in der dortigen Umgebung.

LINUX Cluster Project

TUM Mikrobiologie


  • Name: Institut für Medizinische Mikrobiologie, Immunologie und Hygiene
  • Address: Trogerstr. 30, 81675 München
  • Project Proposal Date: 2020-03-20 15:23:59


Our research aims to understand the differentiation dynamics of CD8 T cells during different types of infections. We have therefore isolated CD8 T cells from the spleens and lymph nodes from either acute or chronic bacterial and viral infections and performed single-cell RNA sequencing (scRNAseq) with the 10x Genomics technology. Furthermore, we are collaborating with other research groups where we already have performed and also planned additional scRNAseq experiments. We will also do the processing and analysis of the experimental data that we have collected together with our collaboration partners. (In one collaboration project we are following the early developmental stages of natural killer T (NKT) cells. During another collaboration, we try to investigate the immune infiltration and tumor microenvironment during pancreatic ductal adenocarcinoma.) We plan to do the raw data analysis, which includes the generation of fastq files and gene alignments of the scRNAseq data, with the Cellranger program on a Linux computer with high computational power. After the gene alignments on the Linux clusters, we continue performing the downstream analysis on our own computers. 
 We then use the information that we derive from the scRNAseq data, together with information from other types of experimental data (FACS measurement, live cell imaging, etc.) to develop mathematical models that describe the differentiation pathways of CD8 T cells during the different infections, as well as the NKT cell development. To calibrate these mathematical models, we use various parameter and model inference frameworks which require the use of MATLAB and C++. Given the high-dimensionality and complexity of the experimental data on the one hand, and complexity of the underlying mathematical models (potentially comprising many parameters) on the other hand, we also plan to use the Linux Cluster for the task of parameter inference and model comparison.