|
Abstract:
|
Single Cell Analysis (SCA), a set of techniques for gathering genomic-level data from single cells, has recently developed as a field of enormous promise. In cancer research, it offers the ability to understand tumor cell heterogeneity at an unprecedented level of resolution. However, several issues of a statistical nature have not been resolved. The small amount of nucleic acids in single cells make it necessary to use nucleic acid amplification, and this introduces biases and variability which are not fully characterized. This affects the analysis of SCA data, starting from the design stage: what sample sizes (number of cells) are appropriate? What sequencing depth is needed? In the trade-off between these two factors, where is the optimal design?
At the same time, SCA offers unexpected opportunities for studying processes that vary continuously in time. I developed CRESCA (Cycle-Regulated expression via SCA) as a method for extracting dynamic information about the cell cycle from static single-cell gene expression data. This holds great potential for cancer genetics, opening a direct window into the dysregulation of the cell division cycle, one of the hallmarks of cancer.
|
