Beyond stable isotope labeling: A mathematical algorithm capable of extracting biological signal from noise inherent in genetically diverse populations.

摘要

An important goal in proteomics is the identification of protein "biomarkers" to aid in detection, diagnosis and treatment of disease. Efficient identification of such markers requires a technology capable of profiling large numbers of samples and robust enough to reveal the relevant biological structure underlying the "noise" inherent in genetically diverse populations. To this end, we have developed a suite of algorithms capable of high throughput and robust peptide quantitation without resorting to stable isotopic labeling. We validated this technology in three organisms: 1) Francisella novicida (~1,800 ORFs), 2) Saccharomyces cerevisiae (~6,000 ORFs) and 3) human leukemia cells (~25,000 ORFs). Results in these three diverse organisms will show our mathematical algorithm works independent of underlying genetic complexity.