Title: Enabling Translational Proteomics in Human Biological Samples
J. Will Thompson, Ph.D.
Assistant Research Professor, Pharmacology and Cancer Biology
Assistant Director, Duke Proteomics and Metabolomics Shared Resource
New Belgium Brewery
500 Linden St
Fort Collins, CO
Analysis of proteins by mass spectrometry has made extraordinary advances in the past 10 years, to the point where it is now routine to measure several thousand proteins in a single sample in 2 hours or less. Additionally, the ability to perform multiplexed targeted validation by multiple reaction monitoring (MRM) mass spectrometry, when combined with the decrease in cost of stable-isotope reference peptides, now makes the possibility of validation in large cohorts a reality. Nonetheless, the translation of human proteomics studies to validated biomarkers has been limited in general by poor experimental design in the discovery stage. Some typical limitations include low sample numbers and generally a lack of quality control (QC) along the discovery pipeline, including the steps of sample preparation, analytical data collection, and data analysis. Our lab has focused on efforts to improve the quality of proteomics discovery experiments in order to allow biomarker discovery to be performed in clinically relevant sample cohorts, and decrease the rate of biomarker attrition along the translational pipeline. This presentation will use examples from the areas of hepatitis/liver fibrosis, cardiovascular disease, and sepsis to describe important workflows and pitfalls along the biomarker pipeline.