Systems Biology for Smart-Cities
Last month I attended the retirement ceremony of my PhD supervisor — Prof Hans Westerhoff — from his professorships in Amsterdam. His invitation came as a nice and unexpected surprise — even more so, considering my not so recent carrier change that marked my departure from the world of Systems Biology (where Hans is one of the world’s most renown experts) and my arrival to the as-exciting field of Smart-Cities.
As all brilliant scientists, Hans likes to recognise patterns and commonalities across different venues of scientific enquiry. It is not by chance that his academic farewell celebration was also the occasion for him to host a scientific conference where the theme was how the Systems Biology approach may be applied to understand and address the complexities of our society and stir its development towards desirable outcomes.
This is indeed a very interesting point that I didn’t fail myself to ponder. However, while Hans’ contribution to this conversation had more of a societal and political flavour, my personal take is that similar considerations can be applied to the current notions of smart-cities and urban development more in general (notions that also contain — by the way — a strong societal element). In this article I make a first attempt to hint in that direction, drawing some parallelisms between the study of cities and the study of biological organisms through the lens of Systems Biology.
What is Systems Biology?
As Aristotle once said, “The whole is greater than the sum of its parts”. Systems Biology applies this concept to the study of living organisms, trying to understand how the interaction of their molecular constituents gives rise to the system-level properties of life.
To do this, the corpus of knowledge available for a given biological system is gathered and organised in a coherent computational model, where the simple components of such system and their interactions are described as explicitly as possible. If this model is able to accurately describe and predict the behaviour of the underlying living organism, then the researcher can infer how that behaviour emerges from the interactions of the system’s simpler components.
Are cities an extension of biology? Are they just very large organisms?
It is difficult not to notice the similarities between cities and living organisms. Cities — like cells — are characterised by stocks and flows of matter, energy and information; they depend on external resources for their survival; they need to transform these resources in consumable goods; they have to adapt to changes in their external environment; and the list could go on.
Some — like Jeffrey West — go even further by asserting that such similarities might not be merely a useful conceptual tool, but could rather reflect a more literal truth, as cities would be actual living organisms (although not biological in nature). I leave to the reader to decide whether this assertion is far fetched or not. For now I would simply point out that, if there are indeed similarities between cities and — say — cells, then it seems reasonable to address the organisational, logistic and governmental complexities of the urban environment using tools and approaches borrowed from the computational element of Systems Biology.
Urban metabolism
Some terms that designate specific functions of living systems have already made their appearance in the discourse about smart-cities. The most notable example is probably metabolism.
In biology the term metabolism refers to all chemical reactions involved in maintaining the living state of an organism. This network of reactions is very complex but the basic functions it fulfils are: (i) the conversion of food into energy to run cellular processes; (ii) the conversion of food/energy into building blocks (like proteins) that constitute the physical structure of the organism; (iii) the elimination of waste.
This seems pretty much what a city does — if we care to replace food with the more generic term resources and possibly be more specific about what a city’s building blocks are. Tools like Material Flow Analysis (MFA) — which examine the stock and flows of materials through metabolic lenses — have proved to be useful in providing policy makers with important insights about the priorities pertaining to material and waste management. When applied to an entire urban environment, MFA provides the means to carry out what is known as Urban Metabolism Analysis (UMA). However, these tools are more descriptive than analytic in nature, in that they implement what is essentially an accounting approach.
The Systems Biology counterpart of MFA is called Metabolic Control Analysis (MCA), a theoretical and computational tool that considers not only the topology of the network of interactions (i.e. what component interacts with what other component) but also the dynamics that characterises each of those interactions¹. The result is a very sophisticated conceptual framework able to predict how the control on specific processes is distributed among the different system’s components, and what component should be targeted in order to foster the maximal effect on the process of interest. MCA is then far more advanced than MFA in its ability to analyse a metabolic network, as it allows for (and in fact requires) an explicit mathematical description of the system under study.
Different levels of complexity
Non-linearity
Because MFA implements what is essentially an accounting approach, the effect on the state of the system is implicitly assumed to vary linearly with changes of the adjustable parameters. However, if there is anything that Systems Biology teaches us, it is that living systems (and more in general complex systems characterised by a network of highly interconnected components) are anything but linear in their behaviour.
Perhaps inspired by Systems Biology (?), the study of urban metabolism has recently started to also include mathematical descriptions of the city system’s dynamics, so as to improve the predictive power of its models. This allows to estimate how the stocks and flows of material change over time in a quantitative (or semi-quantitative) fashion and to account for any non-linearity that emerges in the network’s response to specific perturbations.
Possible applications of these mathematical models are pollution prevention programs, mobility optimisation schemes and waste minimisation solutions.
Hierarchy of organisational and logistic layers
Another source of complexity in living organisms (and cities) emerges from the interplay of different hierarchical layers, all exerting some degree of control upon the system’s processes.
In cells, for example, metabolic fluxes are controlled by different mechanisms that are part of a hierarchically organised structure. Genes contain the encoded information — the blueprint, if you wish — to build the machinery that makes metabolism run. The production of this machinery is obviously a prerequisite for any meaningful metabolic process to occur, and it is itself a very complex process regulated by many factors. This machinery, in turn, allows the different metabolic reactions to occur at different rates, depending on the interplay of internal and external conditions.
Similarly, in cities the provision of services and infrastructure is not a simple process, where demand and supply meet at a single point of contact. There are many stakeholders involved in the design and delivery of any process that underpins the overall functioning of our cities. This results in a hierarchical structure that reflects the scope and range of influence of the different participants in their capacity to supply / consume a variety of services. In addition to this, it is not uncommon to run into conflicting interests and trade-offs, highlighting the importance of laying down an action plan that identifies strategic priorities at different levels of organisation.
To my knowledge, no computational tool has been devised yet to emulate the hierarchical organisational structure that emerges within a city². Systems Biology — again — seems quite ahead in its ability to explain the contribution that different hierarchical layers exert on the control of metabolic processes³. In fact the term Hyerarchical Control Analysis (HCA) was coined to describe an extension of MCA capable of capturing precisely those aspects.
Challenges and future prospects
In the previous section I suggested that the study of cities and their dynamics could draw inspiration from some of the conceptual and computational tools used in Systems Biology, such as MCA and HCA.
Although such tools seem to lend themselves nicely to urban-related studies (given the similarities between cities and living organisms), one should keep in mind that their transposition from the purely biological domain to its urban-scale counterpart is not obvious.
The study of cell metabolism is facilitated by a wide range of well established, experimental procedures. These allow researchers to fully characterise the set of metabolic reactions occurring in a given cell-type or organism, and to quantify their activity under different conditions. The whole process is surely laborious and not as straight forward as it may sound; however the pipeline that goes from culturing the cells of interest to modelling their metabolism in silico has been successfully adopted countless times.
On the contrary, harvesting the necessary information to fully characterise the dynamics that regulates urban metabolism is something that has never been done before. There are different reason for that:
- there is not an established set of protocols that can be used to gather all the necessary information across the different sectors contributing to the provision, delivery and consumption of services;
- it is not obvious (at least at this stage) what all this necessary information would consist of, e.g. what is necessary what is not;
- there are reservations (mostly from the private sector) to fully disclose the detailed chain of processes that characterise specific and potentially critical workflows.
- there tend to be overlaps of competence / sphere of influence between different entities, as well as conflicting interests that do not necessarily resolve in a predictable manner.
There are also other reasons that are more elusive in nature and difficult — if not impossible — to quantify. The most obvious is perhaps the cultural background of the place, which shapes how resources are used, business is done and decisions are made. In addition to that, the dynamics of today’s cities (mostly if they embrace innovation and technological experimentation) evolves much quicker than biological organisms do, making any attempt to capture its overall functioning even more arduous.
A possible way to mitigate these obstacles may consists of designing (rather than capturing) the metabolism of a city, in a sort of urban-scale metabolic engineering experiment. In this way the dynamics of the system could be more easily described in mathematical terms, as it would follow precise design principles. This would make the tools of Systems Biology more readily viable for urban-related studies. However such an approach to urban metabolism would be more easily implemented in cities that are built from scratch, and would entail quite an extensive control by the city governance over the different organisational aspects and hierarchical layers at play⁴.
In conclusion, although a fully fledged conceptual framework — such as MCA or HCA— may be out of reach in the study of urban metabolism for the time being, the basic principles of control theory as applied to biological organisms by Systems Biology may easily be the way forward.
Resources and Footnotes
- See “Metabolic control analysis: a survey of its theoretical and experimental development.” by David Fell.
- Different models have been devices to emulate the hierarchy in size that emerges between cities when they compete for resources. This, however, is a different kind of hierarchy from the one we are focusing on this article.
- See “(Im)Perfect robustness and adaptation of metabolic networks subject to metabolic and gene-expression regulation: marrying control engineering with metabolic control analysis.” by Fei He, Vincent Fromion and Hans V. Westerhoff.
- This is the approach currently adopted in Asia, where new cities are designed around what are considered “best-practice principles” of efficiency, resilience and sustainability.
