Research

Metabolomics is an emerging field of research aimed at characterizing and quantitating a variety of small molecule compounds across various biological samples. In addition to overseeing operations of the Metabolomics Shared Resource in the Lombardi Comprehensive Cancer Center at Georgetown University Medical Center, our core research group focuses on a number of metabolomics studies in a variety of areas including radiation biology, cancer and neurodegenerative disease.

One area of research in which we are active is in pancreatic cancer. Using human pancreatic ductal adenocarcinoma (PDAC) serum patient samples, we are working to identify biomarkers of disease that can be used for earlier diagnosis. Additionally, we validate the specificity of this markers from patients with pancreatitis and intraductal papillary mucinous neoplasms (IPMN). We also implement metabolic flux analysis studies in induced pluripotent stem cell (iPSC) and 3D-organoid models, to identify early metabolic markers of disease and novel therapeutic targets in the progression of pancreatic cancer from pancreatic intraepithelial neoplasia (PanIN) to late-stage PDAC. Furthermore, we are interested in the analytical and functional characterization of exosomes and other extracellular vesicles (EVs) through pancreatic cancer progression.

A second area of research involves understanding the molecular changes associated with ionizing radiation exposure. Using a variety of in vivo models, we examine the effects of radiation induced tissue toxicity on the molecular profile of these models, elucidating mediators of radiation damage response. This work seeks to understand the mechanism behind the radiation damage response, and is applicable to cancer patients who receive radiation therapy, humans exposed to space radiation and for planning for mass-exposure events.

Finally, we work to create computational tools to aid in unknown metabolite identification. This includes developing algorithms for metabolite identification, structuring and populating MS/MS libraries, optimizing peak-picking parameters for untargeted metabolomics studies, designing new visualization techniques for high-dimensional metabolomics data and integrating machine-learning methods into metabolite recognition workflows.

Our goal is to join analytical techniques (LC-MS, GC-MS and NMR) with cell and molecular biology techniques, and computational and statistical methods, to identify signaling mechanisms that are responsible for phenotypic changes within a system, and to orthogonally validate potentially abrogated pathways.