Bifurcation Analysis of Single-cell Gene Expression Data
主 题: Bifurcation Analysis of Single-cell Gene Expression Data
报告人: Prof. Guo-Cheng Yuan (Dana-Farber Cancer Institute and Harvard School of Public Health)
时 间: 2012-06-06 16:00-17:00
地 点: 理科一号楼1114 （数学所活动）
One of the fundamental questions in biology is how a complex organism
develops from a single cell. At the systems-level, this dynamic process
has been described as an \"epigenetic landscape\". Stuart Kauffman
proposed the hypothesis that each observed cell type is an \"attractor\"
of the dynamic gene regulatory network. The recent development of
single-gene expression profiling technology has provided an
unprecedented opportunity to test this hypothesis.
We have developed a new method, called SCUBA, to analyze single-cell
gene expression data. Our method is based on a novel combination of
dynamic clustering and the mathematical theory of bifurcations. First,
we extended traditional clustering approaches by further incorporating
temporal variations including bifurcations. Second, we apply the
mathematical theory on bifurcations to characterize the dynamical
changes of clustering patterns. Using this new method to analyze a
public dataset, we were able to obtain a glimpse of the epigenetic
landscape for early mouse embryonic development.
If time allows, I will also discuss our recent work in predict
epigenetic patterns from DNA sequences. Our work suggests that the
genome and epigenome are not independent but highly associated with each
other.
