Biology, Networks and Control Theory

The Institute for Mathematics and its Applications (or IMA, in Minneapolis, Minnesota), is teaming up with the Mathematical Biosciences Institute (or MBI, in Columbus, Ohio). They’re having a big program on control theory and networks:

IMA-MBI coordinated program on network dynamics and control: Fall 2015 and Spring 2016.

MBI emphasis semester on dynamics of biologically inspired networks: Spring 2016.

At the IMA

Here’s what’s happening at the Institute for Mathematics and its Applications:

Concepts and techniques from control theory are becoming increasingly interdisciplinary. At the same time, trends in modern control theory are influenced and inspired by other disciplines. As a result, the systems and control community is rapidly broadening its scope in a variety of directions. The IMA program is designed to encourage true interdisciplinary research and the cross fertilization of ideas. An important element for success is that ideas flow across disciplines in a timely manner and that the cross-fertilization takes place in unison.

Due to the usefulness of control, talent from control theory is drawn and often migrates to other important areas, such as biology, computer science, and biomedical research, to apply its mathematical tools and expertise. It is vital that while the links are strong, we bring together researchers who have successfully bridged into other disciplines to promote the role of control theory and to focus on the efforts of the controls community. An IMA investment in this area will be a catalyst for many advances and will provide the controls community with a cohesive research agenda.

In all topics of the program the need for research is pressing. For instance, viable implementations of control algorithms for smart grids are an urgent and clearly recognized need with considerable implications for the environment and quality of life. The mathematics of control will undoubtedly influence technology and vice-versa. The urgency for these new technologies suggests that the greatest impact of the program is to have it sooner rather than later.

First trimester (Fall 2015): Networks, whether social, biological, swarms of animals or vehicles, the Internet, etc., constitute an increasingly important subject in science and engineering. Their connectivity and feedback pathways affect robustness and functionality. Such concepts are at the core of a new and rapidly evolving frontier in the theory of dynamical systems and control. Embedded systems and networks are already pervasive in automotive, biological, aerospace, and telecommunications technologies and soon are expected to impact the power infrastructure (smart grids). In this new technological and scientific realm, the modeling and representation of systems, the role of feedback, and the value and cost of information need to be re-evaluated and understood. Traditional thinking that is relevant to a limited number of feedback loops with practically unlimited bandwidth is no longer applicable. Feedback control and stability of network dynamics is a relatively new endeavor. Analysis and control of network dynamics will occupy mostly the first trimester while applications to power networks will be a separate theme during the third trimester. The first trimester will be divided into three workshops on the topics of analysis of network dynamics and regulation, communication and cooperative control over networks, and a separate one on biological systems and networks.

The second trimester (Winter 2016) will have two workshops. The first will be on modeling and estimation (Workshop 4) and the second one on distributed parameter systems and partial differential equations (Workshop 5). The theme of Workshop 4 will be on structure and parsimony in models. The goal is to explore recent relevant theories and techniques that allow sparse representations, rank constrained optimization, and structural constraints in models and control designs. Our intent is to blend a group of researchers in the aforementioned topics with a select group of researchers with interests in a statistical viewpoint. Workshop 5 will focus on distributed systems and related computational issues. One of our aims is to bring control theorists with an interest in optimal control of distributed parameter systems together with mathematicians working on optimal transport theory (in essence an optimal control problem). The subject of optimal transport is rapidly developing with ramifications in probability and statistics (of essence in system modeling and hence of interest to participants in Workshop 4 as well) and in distributed control of PDE’s. Emphasis will also be placed on new tools and new mathematical developments (in PDE’s, computational methods, optimization). The workshops will be closely spaced to facilitate participation in more than one.

The third trimester (Spring 2016) will target applications where the mathematics of systems and control may soon prove to have a timely impact. From the invention of atomic force microscopy at the nanoscale to micro-mirror arrays for a next generation of telescopes, control has played a critical role in sensing and imaging of challenging new realms. At present, thanks to recent technological advances of AFM and optical tweezers, great strides are taking place making it possible to manipulate the biological transport of protein molecules as well as the control of individual atoms. Two intertwined subject areas, quantum and nano control and scientific instrumentation, are seen to blend together (Workshop 6) with partial focus on the role of feedback control and optimal filtering in achieving resolution and performance at such scales. A second theme (Workshop 7) will aim at control issues in distributed hybrid systems, at a macro scale, with a specific focus the “smart grid” and energy applications.

For more information on individual workshops, go here:

• Workshop 1, Distributed Control and Decision Making Over Networks, 28 September – 2 October 2015.

• Workshop 2, Analysis and Control of Network Dynamics, 19-23 October 2015.

• Workshop 3, Biological Systems and Networks, 11-16 November 2015.

• Workshop 4, Optimization and Parsimonious Modeling, 25-29 January 2016.

• Workshop 5, Computational Methods for Control of Infinite-dimensional Systems, 14-18 March 2016.

• Workshop 6, Quantum and Nano Control, 11-15 April 2016.

• Workshop 7, Control at Large Scales: Energy Markets and Responsive Grids, 9-13 March 2016.

At the MBI

Here’s what’s going on at the Mathematical Biology Institute:

The MBI network program is part of a yearlong cooperative program with IMA.

Networks and deterministic and stochastic dynamical systems on networks are used as models in many areas of biology. This underscores the importance of developing tools to understand the interplay between network structures and dynamical processes, as well as how network dynamics can be controlled. The dynamics associated with such models are often different from what one might traditionally expect from a large system of equations, and these differences present the opportunity to develop exciting new theories and methods that should facilitate the analysis of specific models. Moreover, a nascent area of research is the dynamics of networks in which the networks themselves change in time, which occurs, for example, in plasticity in neuroscience and in up regulation and down regulation of enzymes in biochemical systems.

There are many areas in biology (including neuroscience, gene networks, and epidemiology) in which network analysis is now standard. Techniques from network science have yielded many biological insights in these fields and their study has yielded many theorems. Moreover, these areas continue to be exciting areas that contain both concrete and general mathematical problems. Workshop 1 explores the mathematics behind the applications in which restrictions on general coupled systems are important. Examples of such restrictions include symmetry, Boolean dynamics, and mass-action kinetics; and each of these special properties permits the proof of theorems about dynamics on these special networks.

Workshop 2 focuses on the interplay between stochastic and deterministic behavior in biological networks. An important related problem is to understand how stochasticity affects parameter estimation. Analyzing the relationship between stochastic changes, network structure, and network dynamics poses mathematical questions that are new, difficult, and fascinating.

In recent years, an increasing number of biological systems have been modeled using networks whose structure changes in time or which use multiple kinds of couplings between the same nodes or couplings that are not just pairwise. General theories such as groupoids and hypergraphs have been developed to handle the structure in some of these more general coupled systems, and specific application models have been studied by simulation. Workshop 3 will bring together theorists, modelers, and experimentalists to address the modeling of biological systems using new network structures and the analysis of such structures.

Biological systems use control to achieve desired dynamics and prevent undesirable behaviors. Consequently, the study of network control is important both to reveal naturally evolved control mechanisms that underlie the functioning of biological systems and to develop human-designed control interventions to recover lost function, mitigate failures, or repurpose biological networks. Workshop 4 will address the challenging subjects of control and observability of network dynamics.

Events

Workshop 1: Dynamics in Networks with Special Properties, 25-29 January, 2016.

Workshop 2: The Interplay of Stochastic and Deterministic Dynamics in Networks, 22-26 February, 2016.

Workshop 3: Generalized Network Structures and Dynamics, 21-15 March, 2016.

Workshop 4: Control and Observability of Network Dynamics, 11-15 April, 2016.

You can get more schedule information on these posters:

12 Responses to Biology, Networks and Control Theory

  1. Vanessa Schweizer says:

    These look great! Thanks for spreading the word. These are new fields for me, but I want to know more. What are the best papers/books (perhaps even sections of your blog or particular posts) for getting started? Speak to me like I’m a freshman. :)

    • John Baez says:

      Hi! What do you want to get started on? This bunch of workshops seems to span a huge range, from classical control theory to the control of quantum processes, to the study of chemical reactions, to neuroscience and neural networks, to the design of power grids and energy markets. I’m not an expert on all this stuff, obviously, but if you pick something, maybe I can recommend some reading — with help, I hope, from other readers.

    • Toussaint says:

      I’m interested in references as well but in my case just biological networks, specifically gene regulatory networks. Since I am new to the mathematical aspects of this field, introductory texts would be most beneficial. My background is in biotechnology. On the math side, I can understand texts that involve differential equations and probability theory (up to Markov processes) but not much beyond or aside those. Thanks

      • John Baez says:

        I know very little about gene regulatory networks. So, I started by looking at this:

        • Wikipedia, Gene regulatory networks.

        Here I read

        Mathematical models of GRNs have been developed to capture the behavior of the system being modeled, and in some cases generate predictions corresponding with experimental observations. In some other cases, models have proven to make accurate novel predictions, which can be tested experimentally, thus suggesting new approaches to explore in an experiment that sometimes wouldn’t be considered in the design of the protocol of an experimental laboratory. The most common modeling technique involves the use of coupled ordinary differential equations (ODEs). Several other promising modeling techniques have been used, including Boolean networks, Petri nets, Bayesian networks, graphical Gaussian models, Stochastic, and Process Calculi.

        It turns out I know some of the mathematical models here. I helped write a book on Petri nets, which explains in detail how these are used to describe networks of chemical reactions:

        • John Baez and Jacob Biamonte, Quantum techniques for stochastic mechanics.

        However, if you wish to avoid hearing about cool analogies to quantum mechanics and focus on biological aspects of Petri nets, I recommend this friendly book:

        • Darren James Wilkinson, Stochastic Modelling for Systems Biology, Taylor and Francis, New York, 2006.

        or for something a bit more terse but still expository, these articles:

        • Peter J. E. Goss and Jean Peccoud, Quantitative modeling of stochastic systems in molecular biology by using stochastic Petri nets, Proc. Natl. Acad. Sci. USA 95 (1998), 6750–6755

        • Ina Koch, Petri nets—a mathematical formalism to analyze chemical reaction networks, Molecular Informatics 29 (2010), 838–843.

        Chemists use a different-looking but very similar formalism, ‘chemical reaction networks’, which has the advantage that one can deduce lots of things about the differential equations described by these networks simply by looking at the networks! My book discusses a lot of those results.

    • Vanessa Schweizer says:

      I’m with Toussaint, where I am new to the mathematical details. His description of what he’s familiar with is similar to my background. I have been using various techniques that seem to overlap with networks to model socio-technical systems (e.g. users + infrastructure).

      • John Baez says:

        Besides the references I gave Touissant, you might also like this introduction to network theory, which is less about using networks as tools to model dynamical systems, and more about ways to take a complicated network and extract information from it:

        • Mark Newman, Networks: An Introduction, Oxford U. Press, Oxford, 2010.

        There’s also this, which I haven’t looked at, but sounds interesting and has good reviews:

        • Matthew O. Jackson, Social and Economic Networks, Princeton U. Press, Princeton, 2010.

        Good luck, and let me know how it goes!

  2. Giampy says:

    The 4th link (to workshop #4) it’s wrong, should be 899 instead of 896.

You can use Markdown or HTML in your comments. You can also use LaTeX, like this: $latex E = m c^2 $. The word 'latex' comes right after the first dollar sign, with a space after it.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s