Critical Dynamics in Coupled Boolean Networks With Applications in Pluripotent Stem Cells

Chris Kang

doi:10.7273/000005191

The spontaneous emergence of order is one of the fundamental features in complex systems. Embryogenesis and stem cell differentiation, for example, are highlighted by spontaneous phase transitions of disorder-order in its developmental cycle, resulting in a new functional life form. In this thesis, we examine the existence of characteristic signatures of disorder-order phase transitions in populations of isogenic cells whose gene regulatory networks are modeled as Boolean networks, motivated by the studies in pluripotent stem cells. We present a model for coupled random Boolean network whose interaction rules are governed by the mulitilayer Ising Hamiltonian. Our approach allows for modeling multiple, biologically plausible intercellular signaling effects (paracrine, autocrine, and external fields). The model demonstrates an emergence of cell types in populations, which can be verified by three modes of analysis: (1) spectral decomposition of cell type distributions, (2) linear optimization method, and (3) a machine learning approach based on non-negative matrix factorization. Statistical analyses show that coupled random Boolean networks exhibit signatures of a second-order phase transition in cell type composition due to a combination of cell-cell cooperativity and intrinsic noise in its population. Near critical states in its parameters, stem cell populations undergo a spontaneous phenotypic transition, characterized by the symmetry-breaking events. Here, we show that this transition is possible through proper interplay of cell-cell cooperativity and intrinsic noise. Moreover, the system displays a first-order phase transition in the presence of external stimuli. We consider the effects of different sizes in control genes, specifically, the control kernel (CK) set. We see that dynamically pinning CKs with the mulitilayer Ising Hamiltonian can generate new cell types and demonstrate cell-to-cell variability in model simulations. Finally, we present that cells can collectively self-tune through a critical transition, which allows them to decide their fate. This behavior is seen with an internal dynamical system of a negative feedback between tissue heterogeneity and intrinsic noise, and the result is compared to recent experimental studies of mouse embryonic stem cells. Under strict conditions, the model captures experimentally-observed qualitative behaviors of multilevel transitions in cell type heterogeneity; that is, a unimodal-bimodal transition of cell states at the cellular and colony level.

Critical Dynamics in Coupled Boolean Networks With Applications in Pluripotent Stem Cells

Files and links (1)

Abstract

Metrics

Details