“Works-in-Progress” Postdocs Seminar

The Department of Epidemiology presents “Works-in-Progress” postdoctoral research fellows seminar

Open to the public

Talk 1:

Metabolomic Signature of Ultra-Processed Food Intake in Association with Colorectal Cancer Risk

Mengxi Du
Postdoctoral Research Fellow
Department of Epidemiology
Harvard T.H. Chan School of Public Health

Abstract: Growing evidence links high ultra-processed food (UPF) intake to colorectal cancer (CRC) risk. However, the metabolomic changes underlying this relationship remain unknown. Integrating dietary and high throughput metabolomic profiling data from large US cohorts – Nurses’ Health Study and Health Professionals Follow-up Study, we derived a metabolomic signature of UPF intake among 1,715 participants and prospectively evaluated the signature in pre-diagnostic blood specimens among a separate group of 686 matched pairs of incident CRC cases and controls. Our analysis showed a strong positive association between the metabolomic signature and CRC risk. Among individual metabolites, C4-OH carnitine, trimethylbenzene, and N2, N2-dimethylguanosine were positively associated, while 21-deoxycortisol and proline betaine were inversely associated with CRC risk. These findings provide biological insight into the potential metabolic pathways linking UPF intake to CRC risk.

Talk 2:

Establishing Multi-Omics Risk Scores for Inflammation Markers and Their Utility in Epidemiological Research

Anat Yaskolka Meir
Postdoctoral Research Fellow
Department of Epidemiology
Harvard T.H. Chan School of Public Health

Abstract: Germline genetic components of disease risk and phenotypes remain the same throughout life, while epigenetic factors vary dynamically throughout the life course owing to a complex set of endogenous biological processes and environmental exposures. Blood metabolomics are chemically diverse and may reflect short-term exposures at any moment. Integrating different time-scope omics from different functional layers (“multi-omics”) may characterize the burden on chronic inflammation at different ranges of time, from immediate and acute status (e.g., metabolomics) to lifetime and low-grade impact (e.g., genetics). We used an established Polygenic Risk Score (PRS) and developed a Metabolomic Risk Score (MRS) and Epigenetic Risk Score (ERS) as single scores and integrated the different scores using a hierarchical approach to create multi-omics scores, to predict chronic inflammation status (reflected by circulating C-reactive protein (CRP), interleukin 6 (IL6), and tumor necrosis factor alpha (TNFa)). Data from the Canadian Longitudinal Study on Aging (CLSA) was used to train MRS and ERS models. The Nurses’ Health Studies (NHS) and II (NHSII) and the Health Professionals Follow-up Study (HPFS) were used to validate the models. This work-in-progress seminar will present the methodology of integrating different omics to build single and multi-omics risk scores for inflammation markers and demonstrate their utility in predicting morbidity and all-cause mortality. Using omics from different functional layers may enhance our ability to predict and explore inflammatory-related pathophysiology, requiring a multi-faceted approach.