Abstract

Symposium Sessions

Day 2, June 11（Tue.）　10:15-10:35　Room B (Convention Hall 200)

Metabolomics plays a pivotal role in the study of biological systems, enabling the analysis of metabolites as direct phenotypic outcomes of genetic information. Despite the existence of large volumes of public metabolomics data, comprehensive reanalysis remains largely unexplored due to the challenges in integrating data collected under diverse analytical conditions.
To address this bottleneck, we are working on developing a network-based data analysis method that enables integration and reanalysis of cross-study large scale metabolomics data. By employing the iDMET approach which quantifies similarities in metabolic change trends between different studies, large number of study datasets acquired from a metabolomics data repository have been integrated as a network structure. In addition to the metabolite-based data integration, we extract standardized biomedical concepts from the textual information accompanying each study. This enables the networking of studies based on conceptual similarities, offering a multidimensional perspective that integrates both metabolite-level data and broader research concepts.
The integration of vast amounts of metabolomic datasets unveils a holistic view of metabolic processes, significantly enhancing our understanding of complex biological phenomena. The network integrating various study datasets from various backgrounds, offer us opportunities for deep reanalysis, including novel discoveries and hypothesis generation.

---