Neutral forces acting on intragenomic variability shape the Escherichia coli regulatory network topology

Abstract

Cis-regulatory networks (CRNs) play a central role in cellular decision making. Like every other biological system, CRNs undergo evolution, which shapes their properties by a combination of adaptive and nonadaptive evolutionary forces. Teasing apart these forces is an important step toward functional analyses of the different components of CRNs, designing regulatory perturbation experiments, and constructing synthetic networks. Although tests of neutrality and selection based on molecular sequence data exist, no such tests are currently available based on CRNs. In this work, we present a unique genotype model of CRNs that is grounded in a genomic context and demonstrate its use in identifying portions of the CRN with properties explainable by neutral evolutionary forces at the system, subsystem, and operon levels. We leverage our model against experimentally derived data from Escherichia coli. The results of this analysis show statistically significant and substantial neutral trends in properties previously identified as adaptive in origin-degree distribution, clustering coefficient, and motifswithin the E. coli CRN. Our model captures the tightly coupled genome-interactome of an organism and enables analyses of how evolutionary events acting at the genome level, such as mutation, and at the population level, such as genetic drift, give rise to neutral patterns that we can quantify in CRNs. Reconstructing a CRN from experimental data, elucidating its dynamic and topological properties, and understanding how these properties emerge during development and evolution are major endeavors in experimental and computational biology (1-5).The complexity of CRNs, coupled with observed "unexpected" trends in their properties, such as scale-freeness (6), high degree of clustering (7), and overrepresented subgraphs (3,(8)(9)(10), has led to several hypotheses of adaptive origins and explanations of CRNs and their properties. Central to most of these studies was the use of simplistic graph-theoretic models, such as randomly rewiring the connectivity of a biological network, to serve as a null model for CRN connectivity maps and their properties (11). However, it has been shown that when subjecting CRNs to the various neutral evolutionary forces and tracing their trajectory in time, many of these topological patterns may simply arise spontaneously due to the forces of mutation, recombination, gene duplication, and genetic drift (10, 12). These studies call into question arguments that were made in favor of adaptive explanations for the emergence and conservation of CRN properties (8,(13)(14)(15) and identify important parameters that may significantly affect the evolution of CRNs from a neutral perspective. Specifically (12), they highlighted the role that promoter length, binding-site size, and population size may play in forming certain topological patterns known as motifs. Nonetheless, a lingering question remains: Which specific parts of a CRN arise due to nonadaptive forces and, moreover, can we quantify these patterns to...

Evolution After Whole-Genome Duplication: A Network Perspective

“…Furthermore, this negative correlation between divergence and connectivity is reasonable because an increase number of mutations, particularly those in regions involved in the interactions, would result in an increased (albeit not necessarily at the same rate) loss of interactions. This agrees with recent findings on how mutation at the genomic level, combined with neutral evolutionary forces, shape emergent properties at the network level (Ruths and Nakhleh 2013) and can explain correlations between network properties and gene duplicability (Zhu et al 2012). …”

Making Sense of Transcription Networks

“…2 and 3, and thus, the preponderance of feed-forward loops in biological networks may be a result of the same non-adaptive processes that result in large network size and interconnectedness. Indeed, it has been explicitly proposed that many network motifs have arisen as a result of neutral evolutionary processes rather than selection for a particular function of the motif itself (Cordero and Hogeweg, 2006; Ingram et al, 2006; Ruths and Nakhleh, 2013; Ward and Thornton, 2007). These ideas contrast with models where each feed-forward loop in the network possesses optimized parameters that specify a particular transcriptional input-output relationship.…”

Boosting forward-time population genetic simulators through genotype compression

“…For instance, not only the sequence, but also its annotated features like genes or regulatory elements could be simulated as one complex system, facilitating evolutionary questions which investigate the coevolution of these integrated systems. For instance, we leveraged our compression algorithm in a recent study which investigated the neutral evolutionary trends of the E. coli regulatory network by simulating, at scale, the entire regulome and its underlying sequence (595 operons) over long evolutionary time scales [17]. These simulations resulted in a null distribution of system, sub-system, and operon level regulatory properties, allowing for rigorous statistical testing of neutral topological patterns.…”

“…Further, while ARGs model the evolution of genetic sequences in a population setting, the OG is defined for arbitrary genotypes. A case in point is our recent population-level analysis of regulatory networks in E. coli, where the OG was defined over genotypes consisting of regulatory networks [17]. …”