Vaccine and Infectious Disease Division
Cell growth is a dynamic, stochastic process. In any one homogenous cell population whether in vitro or in vivo the individual cells are not actually all in the same phase of the cell cycle. This phenomenon can theoretically distort gene expression analyses at the single cell level and be a confounding factor in single cell transcriptional profiling.
VIDD Staff Scientist Dr. Greg Finak, Associate Member Dr. Raphael Gottardo, and Andrew McDavid, a PhD student in Statistics at the University of Washington working under the supervision of Dr. Gottardo, recently developed a statistical method to measure how temporal differences in the cell cycle affect gene expression variability at the single cell level.
The authors analyzed gene expression of three different mammalian cells lines during three stages of the cell cycle. For each of these subgroups, individual cells were isolated from homogenous populations and researchers measured expression levels of a total of 253 genes. Their analytical method called the Hurdle model was able to delineate cell cycle phase specific differences in gene expression. Their model identified several genes that were ‘highly ranked,’ meaning their expression correlated with a specific phase of the cell cycle (e.g., those involved with mitosis), and ‘unranked,’ meaning the expression pattern was not dependent on cell cycle at all. As it turned out, only a small fraction of gene expression variability could be contributed to cell cycle asynchronicity. The authors propose the utility of this tool to characterize transcriptional patterns for any two-way comparison, such as the effect of a drug on cell viability. They are currently working on using this new computational framework for characterizing antigen-specific T cells in the context of HIV and malaria vaccination.
McDavid A, Dennis L, Danaher P, Finak G, Krouse M, Wang A, Webster P, Beechem J, Gottardo R. Modeling bi-modality improves characterization of cell cycle on gene expression in single cells. PLoS Comput Biol. 2014 Jul 17;10(7):e1003696. PMCID: PMC4102402.
Full-text Article: http://scholars.fhcrc.org/4571/
PubMed Record: http://www.ncbi.nlm.nih.gov/pubmed/25032992?otool=fhcrclib