Undergraduate Honors Thesis


Interpretation of Hemes: A Human Metabolomic Serum Biomarker of Insufficient Sleep Public Deposited

Downloadable Content

Download PDF
  • Advances in computational power and molecular detection techniques allow investigators unparalleled insight into the metabolome, the collection of small-molecule metabolism products, called metabolites, found in organisms. Metabolomics, the study of such molecules, attempts to discern how various biological conditions such as disease or stress alters metabolite concentration in the body. Investigators can also leverage metabolomics to create biomarkers, tools used to monitor and diagnose biological conditions. Metabolomic biomarkers are often created using metabolites found in accessible biofluids, like blood or urine, and are usually developed to diagnose a disease. Insufficient sleep impacts over a third of Americans, and results in metabolic changes, making it a notable target for biomarker development. The current study uses a cohort of 16 human participants experiencing a protocol of sufficient and insufficient sleep to develop a biomarker using blood metabolite concentration data. Principal component analysis, partial least squares discriminant analysis, and area under the receiver operating characteristic curve analysis aided in the creation of a final biomarker score consisting of six metabolites, weighted by their contribution to overall performance. Biomarker performance was assessed by predictive accuracy at classifying samples into sufficient and insufficient sleep conditions, and area under the receiver operating characteristic curve analysis. These tests were performed in both the original dataset and an independent dataset consisting of 36 individuals experiencing a different protocol of insufficient sleep. Independent verification is essential for effective biomarker development to ensure the results are applicable to a wide range of individuals.
Date Awarded
  • 2018-01-01
Academic Affiliation
Committee Member
Granting Institution
Last Modified
  • 2019-12-02
Resource Type
Rights Statement


In Collection: