Predictive computation of genomic logic processing functions in embryonic development

Abstract

Gene regulatory networks (GRNs) control the dynamic spatial patterns of regulatory gene expression in development. Thus, in principle, GRN models may provide system-level, causal explanations of developmental process. To test this assertion, we have transformed a relatively well-established GRN model into a predictive, dynamic Boolean computational model. This Boolean model computes spatial and temporal gene expression according to the regulatory logic and gene interactions specified in a GRN model for embryonic development in the sea urchin. Additional information input into the model included the progressive embryonic geometry and gene expression kinetics. The resulting model predicted gene expression patterns for a large number of individual regulatory genes each hour up to gastrulation (30 h) in four different spatial domains of the embryo. Direct comparison with experimental observations showed that the model predictively computed these patterns with remarkable spatial and temporal accuracy. In addition, we used this model to carry out in silico perturbations of regulatory functions and of embryonic spatial organization. The model computationally reproduced the altered developmental functions observed experimentally. Two major conclusions are that the starting GRN model contains sufficiently complete regulatory information to permit explanation of a complex developmental process of gene expression solely in terms of genomic regulatory code, and that the Boolean model provides a tool with which to test in silico regulatory circuitry and developmental perturbations.gene regulatory logic | transcriptional control system | sea urchin embryogenesis | Boolean gene expression G ene regulatory network (GRN) models formalize the manner in which specification of cellular domains during development is controlled by spatial and temporal gene expression (1). Each cell fate depends on expression of a specific set of regulatory genes, that is, genes encoding transcription factors and signaling molecules. In each domain of the developing organism and at each point in time, the genetic activities and therefore the fates of the cells are determined directly by the regulatory gene products present in the nuclei. The regulatory states constituted by these regulatory gene products are themselves the output of transcriptional control systems encoded in the genome. GRNs thus capture the transcriptional control functions that specify the spatial regulatory states of the embryo. GRNs consist of regulatory genes and the transcriptional interactions that determine their specific patterns of expression. Models derived from experimental studies of developmental GRNs conceptually relate genomic regulatory sequence information to developmental process. Every node in such network models represents a regulatory gene, which is controlled by interactions encoded in genomic cis-regulatory binding sites.GRN models represent intellectual syntheses of experimental gene expression and cis-and trans-perturbation data, as well as information regard...

Encoding regulatory state boundaries in the pregastrular oral ectoderm of the sea urchin embryo

“…The Bio-Tapestry model presents the predicted topology of the oral ectoderm regulatory system, displaying its modular circuit features (4), such as double-negative gates, community effect circuits, exclusion circuits, feedbacks, etc. Space does not permit discussion of these individual features and the logic operations that they execute; suffice it to say, the pregastrular oral ectoderm GRN models will soon support a global logic analysis similar to that recently applied to the endomesoderm GRN model (7). Fig.…”

“…S1 can be abstracted to provide the dynamically changing Boolean expression matrices shown in Fig. 2, where the contributions of the 12 genes to the regulatory state of each domain can be read horizontally (6,7). These are the specific patterns of expression for which we seek causal explanation in the encoded architecture of the ectodermal GRNs.…”

“…This GRN model encompasses about half of the embryo, covers 30 h of development (18 h in the nonskeletogenic mesoderm), and all or almost all relevant regionally expressed regulatory genes. A recent study (7) shows that the endomesoderm GRN model contains sufficient regulatory relationships to generate a computational automaton that successfully predicts almost all spatial and temporal regulatory gene expression in this phase of embryogenesis.…”

Evolutionary rewiring of gene regulatory network linkages at divergence of the echinoid subclasses

“…The explanatory power of these networks was demonstrated, in these pages, by a predictive computational analysis that showed that they contain sufficient information to regenerate the developmental course of events in silico, in automaton-like fashion (6). The present work stems from the almost irresistible opportunities that these same GRNs offer for approaching the basic evolutionary mechanisms of GRN divergence.…”

Introduction to Philosophy of Systems Biology

“…The revival of the interest for general design or organizing principles has also given rise to discussions about the extent to which living systems, despite the complexity of intertwined processes, are constituted by individual functional units that exhibit modularity (cf. Green 2015a; Gross, Chapter 10; Isalan et al 2008, Peter et al 2012Peter, Chapter 22). Moreover, the renewed interest in mathematical analysis of organizational features has led to discussions about the implications of abstraction from molecular details for the sake of identifying generalizable organizational features Levy and Bechtel 2013;Wolkenhauer, Chapter 24).…”