This subproject provides the central storage and data analysis hub of the OUTLIVE-CRC project. Our goal is to establish a central database where all raw and analytical data as well as their annotation from all projects will be stored, annotated and distributed. The data will be kept according to FAIR principles to make it available to all involved within the project and also to the general public. 

On the analysis side, the goal is to develop a so-called multi-omics biomarker for the diagnosis and prognosis of colorectal cancer. Multi-Omics means that one does not look for disease markers only in one biological entity, such as the genome, gene expression or at the protein level, but instead combines all information from these different biological entities. In our case, such a marker will be developed from the combined analysis of blood (liquid biopsy) and stool samples, while continuing to analyze patient internal data (P1, P9), genome and epigenome data (P4, P5), as well as information from the proteome (P6), microbiome (P4), and patient metabolome (P3) for their combined predictive power. For modeling, we use detailed functional annotation, as well as unsupervised and supervised dimensionality reduction together with machine learning methods to characterize molecular mechanisms and signaling pathways in tumor progression. Results of this analysis will be evaluated for their sensitivity and specificity to develop a biomarker panel (P8), nutritional intervention (P9), and long-term prediction of patient outcome.