The overarching goal of our laboratory is to develop an integrated understanding of mammalian cell fate decisions, and their connections to cancer and aging. We study such mechanisms at two levels: gene pathway level and genome level.
Understanding the cellular state of quiescence (versus proliferation, differentiation, and senescence)
Currently, we are particularly interested in the control mechanisms of cellular quiescence. Quiescence is a reversible, non-proliferative, "sleep-like" cellular state that plays a critical function in the health of higher organisms. The activation and proliferation of quiescent cell types (e.g., fibroblasts, lymphocytes, and stem cells) are fundamental to tissue repair and regeneration. The balance between cellular quiescence and proliferation is tightly regulated; dysregulation of this balance leads to many diseases, including fibrosis, autoimmune disease, cancer, and aging. Despite its importance, cellular quiescence is poorly understood.
The "reactability" of quiescent cells distinguishes them from cells at irreversibly arrested non-proliferative states such as senescence and terminal differentiation. Strikingly, these distinct cellular states (quiescence, proliferation, differentiation, and senescence) involve a same set of molecular regulators in the Rb-E2F gene network, suggesting that different Rb-E2F network outputs in response to different input signals may yield distinct cell fates. Our earlier work has shown that the Rb-E2F pathway network functions as a "bistable switch"; the Rb-E2F bistable switch converts graded and transient growth signals into an all-or-none E2F activation, which controls the irreversible entry of the cell cycle. Our recent work suggests that this Rb-E2F bistable switch controls cellular quiescence and its heterogeneity. We now further study detailed quiescence control mechanisms in cell models such as fibroblasts and mesenchymal stem cells. To this end, we use mathematical modeling to help us dissect and visualize the dynamic properties of the Rb-E2F pathway network, which are hard to grasp intuitively. We test model-generated hypothesis with carefully designed experiments. In particular, we use single-cell measurements to uncover dynamic characteristics that often get buried by population-average measurements. Experimental results enable us to update and improve our model, to achieve a better integrated understanding of the cellular state of quiescence.
Identifying regulatory networks of cancer pathways at the genomic level
We have previously developed bioinformatics tools, coupled with statistical modeling, to identify cancer pathway signatures to guide better understanding of caner heterogeneity and targeted cancer therapies. Working with our collaborators, we now develop statistical and natural language modeling approaches to reconstruct gene regulatory networks of cancer pathways that are responsible for specific cellular states or responses. In a recent collaborative project, we apply our previously established and newly developed network modeling approaches to mine large-scale experimental data from next-generation sequencing, to identify disrupted cancer pathways that may serve as potential biomarkers for ovarian cancer.