Project Details
Projekt
Bioinformatic data integration and multi-omic pathway analysis
Applicants
Professor Dr. Jan Baumbach; Dr. Olga Tsoy
Subject Area
Orthopaedics, Traumatology, Reconstructive Surgery
Term
since 2024
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 517063424
ProBone will generate a unique data set, including a wide range of clinical measurements for more than 800 patients, as well as omics profiling of cellular systems and mouse models. The central project 2 (CP2), in close collaboration with the other ProBone projects, will address the computational challenges for data preprocessing and analysis, stratification of patients with early-onset low BMD disorders, and disease mechanism extraction. Complex or multifactorial diseases present a high level of heterogeneity, since patients with similar symptoms might have different underlying disease mechanisms. Thus, the treatment and diagnostics of complex diseases should be based on understanding these mechanisms rather than on symptoms or single biomarkers. CP2 will tackle this challenge in the context of early-onset BMD disorders. Thus, CP2 will follow three main objectives. First, we aim to detect subgroups of patients with the same disease mechanisms based on the wide variety of clinical measurements as well as immunophenotyping, metabolomics, and genomics profiling provided by the ProBone projects. This type of analysis is called a disease or patient stratification, and the subgroups of patients are referred to as subtypes. We will use both supervised and unsupervised approaches to detect the subtypes. In a supervised manner, we will apply machine learning approaches to predict clinically relevant subtypes and extract features important for the prediction. In an unsupervised manner, we will cluster patients de novo. Groups of patients identified in an unsupervised approach will be further associated with detected gene variants. As ProBone will generate different types of omics profiling, the second objective will be to survey and evaluate the best practices of preprocessing and data analysis for each type of omics profiling. CP2 will provide guidelines for data analysis and analyze the omics data according to the guidelines. For the projects that provide several types of omics profiling, CP2 will apply multi-omics integration approaches and investigate the impact of different layers of data. Finally, the results of the omics data analysis will be further mined for disease mechanisms using gene prioritization, and gene enrichment approaches based on gene sets or molecular interactions networks. The extracted mechanisms will be analyzed further in order to find candidates for drug repurposing. Disease stratification, candidate disease mechanisms, and repurposed drugs will finally elucidate the heterogeneity of early-onset low BMD disorders and might guide personalized treatment recommendations.
DFG Programme
Clinical Research Units
