Systems Biology of Phenotypic Robustness and Plasticity

Abstract

Synopsis Gene regulatory networks, cellular biochemistry, tissue function, and whole body physiology are imbued with myriad overlapping and interacting homeostatic mechanisms that ensure that many phenotypes are robust to genetic and environmental variation. Animals also often have plastic responses to environmental variables, which means that many different phenotypes can correspond to a single genotype. Since natural selection acts on phenotypes, this raises the question of how selection can act on the genome if genotypes are decoupled from phenotypes by robustness and plasticity mechanisms. The answer can be found in the systems biology of the homeostatic mechanisms themselves. First, all such mechanisms operate over a limited range and outside that range the controlled variable changes rapidly allowing natural selection to act. Second, mutations and environmental stressors can disrupt homeostatic mechanisms, exposing cryptic genetic variation and allowing natural selection to act. We illustrate these ideas by examining the systems biology of four specific examples. We show how it is possible to analyze and visualize the roles of specific genes and specific polymorphisms in robustness in the context of large and realistic nonlinear systems. We also describe a new method, system population models, that allows one to connect causal dynamics to the variable outcomes that one sees in biological populations with large variation.

Are trade‐offs really the key drivers of ageing and life span?

“…This has led to the proposal of a modified version of the mutation accumulation theory of ageing, in which the deleterious effects of mutations are present throughout life, but increase in magnitude with age (Maklakov et al, 2015). Interestingly, this idea is concordant with recent thinking on the evolution of cancer (DeGregori, 2011), as well as with complex systems explanations of ageing as a breakdown in homeostasis (Cohen, 2012(Cohen, , 2016: physiological regulatory networks are highly buffered and redundant, and loss of homeostasis in a given subnetwork thus is most likely to be expressed in the presence of other problems in the same or connected subnetworks (Nijhout, Sadre-Marandi, Best, & Reed, 2017). Early in life, buffering is likely to better mask the effects of mutations than later on when increasing dysregulation exposes the consequences of the mutation.…”

Systems biology of robustness and homeostatic mechanisms

“…As before, an analysis of the virtual population may identify the particular combinations of parameter values that make individuals more or less susceptible to develop low serotonin. An example of such a study, for individual variation of dopamine levels, is shown in Figure of Nijhout et al (). Relating deterministic models to biological data. Imagine an experiment where a experimenter changes the input to a biochemical network for a short period of time and then measures the values of a particular concentration over time. If the experiment is repeated several times, somewhat different curves will be obtained, and the experimenter will report the mean curve and the standard deviations.…”

“…It is this family of model output curves that should be compared to the family of experimental curves, not just the average model curve to the experimental mean curve, because there is real biological information in the dispersion of the curves. Thus, system population models are the right way to compare mechanistic ODE models to real biological data. Identifying important subpopulations. A systems population model can be “treated” with a drug, or given a specific nutrient or vitamin deficiency, and the resulting population can be compared with a untreated population (Nijhout et al, ). Not all individuals will respond similarly, and statistical analyses of the two populations can be used to identify genetic make‐ups (i.e., specific combinations of parameter values) that make individuals particularly sensitive, or resistant, to the treatment.…”

Novel approach to quantitative spatial gene expression uncovers genetic stochasticity in the developing Drosophila eye

“…Indeed, the genetic basis of adaptation is often inferred to be due to differences in gene expression (Chan et al, ; Gompel, Prud'homme, Wittkopp, Kassner, & Carroll, ; Hoekstra & Coyne, ; Jeong et al, ; Pai & Gilad, ; Signor, Liu, Rebeiz, & Kopp, ; Wray, ; Yassin et al, ). This balance between stability and lability is a question that has been addressed philosophically, but little experimental evidence exists to suggest the mechanisms underlying these phenomena (Casci, ; Gibson, ; Green et al, ; Heranz & Cohen, ; Hermisson & Wagner, ; Nijhout et al, ; Rutherford et al, ; Stern, ; True & Haag, ).…”

“…Pai & Gilad, 2014;Signor, Liu, Rebeiz, & Kopp, 2016;Wray, 2007;Yassin et al, 2016). This balance between stability and lability is a question that has been addressed philosophically, but little experimental evidence exists to suggest the mechanisms underlying these phenomena (Casci, 2005;Gibson, 2009;Green et al, 2017;Heranz & Cohen, 2010;Hermisson & Wagner, 2004;Nijhout et al, 2017;Rutherford et al, 2007;Stern, 2000;True & Haag, 2001).…”