Developing Computational Approaches for Integration of Metabolomics into Systems Biology

点击翻译按钮获取中文摘要

Metabolomics is the comprehensive study of all chemicals and metabolites (i.e. the metabolome) present in a biological system. The metabolome includes both endogenous metabolites and exogenous compounds from food, gut microbes, chemical exposures, etc., representing the final products of complex biological events governed by host genetics and environmental influences. Metabolomics is increasingly applied to study complex diseases, plant physiology, animal nutrition, host-gut microbiota interactions as well as environmental monitoring. Bioinformatics plays an essential role in metabolomics - from raw data processing to biological interpretation. Currently, metabolomics is facing three main computational challenges: i) the spectral processing challenge due to the increasing use of high-resolution mass spectrometry (MS) systems, ii) the functional interpretation challenge due to the generation of metabolomics data from complex environments such as the mammalian gut microbiota (i.e. meta-metabolome), and iii) the challenge for integrating metabolomics data with other omics data for comprehensive biological understanding. Novel computational approaches are urgently needed to address these needs. The long-term goal of my research program is to develop an integrated, comprehensive and predictive systems-biology framework to help explain metabolome variations due to different environmental perturbations. The short-term objective is to develop bioinformatics approaches to address the current data processing and integration challenges within the context of host-gut microbiota metabolic interactions. To this end, I will develop innovative computational algorithms to improve MS-based metabolomics data processing and annotation; integrate with genome mining approaches for comprehensive gut metabolome characterization; and finally model the variations of gut metabolome through community-scale metabolic networks. The performance of these tools will be evaluated using the public data sets as well as the data sets generated in-house and through collaborations. The research results, including tools, databases and algorithms will be publicly available through the latest web technologies and cloud-based computing platforms. The proposed research program addresses the major bottlenecks in current metabolomics, and offers great potential for translational applications through the identification of novel biomarkers, the discovery of key connections and the development of predictive models. The research activities described in this proposal will enable effective training and placement of HQP in a cross-disciplinary environment. **