Physics of epigenetic landscapes and statistical inference by cells
Lang, Alex Hunter
MetadataShow full item record
Biology is currently in the midst of a revolution. Great technological advances have led to unprecedented quantitative data at the whole genome level. However, new techniques are needed to deal with this deluge of high-dimensional data. Therefore, statistical physics has the potential to help develop systems biology level models that can incorporate complex data. Additionally, physicists have made great strides in understanding non-equilibrium thermodynamics. However, the consequences of these advances have yet to be fully incorporated into biology. There are three specific problems that I address in my dissertation. First, a common metaphor for describing development is a rugged "epigenetic landscape" where cell fates are represented as attracting valleys resulting from a complex regulatory network. I introduce a framework for explicitly constructing epigenetic landscapes that combines genomic data with techniques from spin-glass physics. The model reproduces known reprogramming protocols and identifies candidate transcription factors for reprogramming to novel cell fates, suggesting epigenetic landscapes are a powerful paradigm for understanding cellular identity. Second, I examine the dynamics of cellular reprogramming. By reanalyzing all available time-series data, I show that gene expression dynamics during reprogramming follow a simple one-dimensional reaction coordinate that is independent of both the time and details of experimental protocol used. I show that such a reaction coordinate emerges naturally from epigenetic landscape models of cell identity where cellular reprogramming is viewed as a "barrier-crossing" between the starting and ending cell fates. Overall, the analysis and model suggest that gene expression dynamics during reprogramming follow a canonical trajectory consistent with the idea of an "optimal path"' in gene expression space for reprogramming. Third, an important task of cells is to perform complex computations in response to external signals. Intricate networks are required to sense and process signals, and since cells are inherently non-equilibrium systems, these networks naturally consume energy. Since there is a deep connection between thermodynamics, computation, and information, a natural question is what constraints does thermodynamics place on statistical estimation and learning. I modeled a single chemical receptor and established the first fundamental relationship between the energy consumption and statistical accuracy of a receptor in a cell.