On Fundamental Implications of Systems and Synthetic Biology

Abstract

The classical scientific paradigm consists in the study of simple physical systems of either a few components and of many components at random, supporting linear causal analysis and reduction to linear causal mechanisms, with the real being what was stable, especially invariant. The philosophy of science was shaped to suit, focusing on determinism, universal a-temporal (hence a-contextual) causal laws, analysis into fundamental constituents then yielding bottom-up mechanical synthesis, a simple deductive model of explanation and prediction with reduction to fundamental laws and separate contingent initial conditions becoming the basic explanatory requirement. In biology reduction to molecular genetics thus became a defining issue where the simplified modelling of the phenotype as a bundle of gene-trait pairs that determined fitness sufficed for population genetics and molecular biology at the time. Life is radically reduced to simple chemical mechanisms and then to physics or it has to be taken outside the paradigm altogether and asserted as metaphysically sui generis. Yet all the while scientific work itself was quietly and often unintentionally laying the groundwork for superseding these approaches, both scientifically and philosophically, through the emergence of new orthodoxy-breaking complex systems concepts focused around instability and symmetry breaking – far-from-equilibrium irreversible systems, feedback, dynamical stability, cybernetic control, bifurcations, servomechanism, and so on. These are the ideas that, allied to the development of generalised network analysis emerging from circuit theory, chemical process engineering and elsewhere, and initially applied in developmental biology, physiology and ecology, would later underlie contemporary systems and synthetic biological modelling and its experimental practice Correlatively, a new philosophy of science is required focusing on dealing with multiple, phase-lagged, multi-scale interdependencies, top-down as well as bottom-up analysis, the entwinement of emergence and reduction, domain-bound (context-dependent) dynamical laws that accept historically unique individuals as the norm, and limited knowability and controllability. In short, a substantially revised philosophy of science is required. A. Setting systems and synthetic biology in context. 1. Systems and synthetic biology in context Systems and synthetic biology promise to revolutionise our understanding of biology, blur the boundaries between the living and the engineered in a vital new bioengineering, and transform our daily relationship to the living world. Their emergence thus deserves to be understood in a wider intellectual perspective. Close attention to their relationship to the larger scientific intellectual frameworks within which they function reveals that systems and synthetic biology raise fundamental challenges to scientific orthodoxy and stand instead in the vanguard of an emerging new complex dynamical systems paradigm now sweeping across science. They emerge from a preceding developmental stage of science where, sketching crudely, biology was divided between molecular biology on the one side and, on the other, physiology (functional biology) and, on a larger scale, population genetics (evolutionary biology) and there was relatively little commerce among these approaches. Molecular biology and evolutionary population biology effectively agreed on assuming simple rules for gene expression that had the effect of reducing organism complexity to genetic complexity and so of treating the organism (reduced to a phenotype) as if it consisted simply of a bundle of genes. Whence, with genes directly related to produced phenotypes through the simple gene-trait rules, population gene frequencies could be constructed and the diversity of complex organic processes could be explained in terms of evolutionary natural selection expressed in population frequency shifts. This left molecular biology to focus on the genes, aka DNA, and evolutionary theory to focus on gene population statistics. Caught between them, physiology focused on its own functional descriptions, cast in terms of organism features like energy fluxes and tissue densities, as, in different ways, did its sister domains of embryology and developmental biology. Though somewhat a caricature, this division of conception and labour leaves the treatment of biosynthetic pathways out of the picture; however, they are essential for biological understanding. For they are the linkages connecting gene activity through intraand then intercellular formation and functioning to organism formation and functioning, and on, finally, to an enriched multi-layered conception of evolutionary process (see below at n.22). It is exactly at this locus that systems and synthetic biology intervene. These sub-disciplines act, severally and together, as an inter-level bridge between molecular biology and physiology, precisely by developing the treatment of biosynthetic pathways, and in this way create a lively, reinvigorating integration to biology. Despite the complexity of biosynthetic pathways, scientists have been able to study them by carrying over into biology certain engineering modelling tools, such as control theory and electrical circuit theory and its generalisation to dynamical network theory. With genes, proteins and metabolites as components and replication, self-assembly, metabolism, repair, growth/death, signalling and regulation as process elements, systems and synthetic biology using these tools to model the complexes of processes that constitute cells and interacting multi-cellular bodies like organs in ways analogous to those in which engineers model and regulate fighter jet See, among many recent texts, the nicely diagrammed overview in [1], chapter 1. 2 See Mitsuro Itaya, chapter 5 herein and, for example, [2]. 3 Cf. Joyce and Palsson, chapter 6 herein for deliberate development of this approach as a constraints4 based delineation of possibilities. Ralph Steuer (Humboldt University, Berlin) “From topology to dynamics of metabolitic networks”, 5 lecture to the Bio-Modelling Network, Manchester University, UK, August 29, 2007. For instance, [3], noting the capacity to directly observe functional units, remarks that “By linking 6 genes and proteins to higher level biological functions, the molecular fluxes through metabolic networks (the fluxome) determine the cellular phenotype. Quantitative monitoring of such whole network operations by methods of metabolic flux analysis, thus bridges the gap by providing a global perspective of the integrated regulation at the transcriptional, translational and metabolic level.” aerodynamics and multi-stage industrial processes. 2 Of the two, synthetic biology has a wider scope than systems biology since, beyond the actual life forms of systems biology, the domain of synthetic biology also includes novel viable life forms and bio-engineering complexes in which specialised organisms and/or bio-materials/processes play important roles. However, the hope underlying work in both studies is that a cell can be adequately modelled as a dynamical pathway network and a multi-celled organism can be adequately modelled as a super-network of these (and so on up). Adding inanimate engineering network components then suffices to encompass all the wider domain of synthetic biology. Methodologically, systems biology and synthetic biology are mutually beneficial (symbiotic); systems biology employs to advantage the perturbational and measurement methods developed by synthetic biology while systems biology provides knowledge of dynamical models of various useful organisms from which synthetic biology may work. The key to the rise of these two interrelated subdisciplines has been the (accelerating) emergence over the past 50 years of highthroughput experimental technologies capable of amplifying trace chemical presences to reliably measurable quantities in practicable times and of doing so simultaneously with increasingly many cellular components. Starting with recombinant DNA techniques for single genes in the 1970’s, today the techniques are crossing the threshold of being able to simultaneously monitor all the ‘omics’ for entire, or nearly entire, cellular genomes. As Palsson says (op. cit. n.1) the arrival of this data both forces and enables the study of the cell as a system. Whilst the earlier experimental stages were appreciated for their capacity to identify the lists of components involved, once this had been achieved the vast quantities of simultaneous data now available can only be usefully simplified and comprehended in terms the interrelationships they reveal, that is, in terms of a network model. Method is as yet at a relatively early stage of development compared to engineering theory, confined in many cases to topological considerations backed by stoichiometric considerations like flux measurements. Beyond this “kinetic modelling is still 4 severely hampered by inadequate knowledge of the enzyme-kinetic rate laws and their associated parameter values.” and is only recently beginning to enhance 5 stoichiometry with direct dynamical modelling. This is partly because data of the kind and quality required is only recently becoming available , and partly because the 6 dynamical operations of very complex networks are still being only indirectly studied,