3rd Nordic Workshop on Statistical Physics: Biological, Complex and Non-Equilibrium Systems

28 Mar 2012, 08:00 → 30 Mar 2012, 18:00 Europe/Stockholm

132:028 (Nordita)

Alberto Imparato (University of Aarhus), Ralf Eichhorn (Nordita)

Venue

Nordita, Stockholm, Sweden

Scope

This workshop is the third one in a series which has been initiated in 2010 at Nordita. The first two editions in 2010 and 2011 were highly appreciated by the participants, what encouraged us to continue, in order to provide a “forum” where scientists in the Nordic countries working in the area of Statistical Physics can meet regularly. The workshop series brings together experts interested in the broad spectrum of timely problems in (classical) Statistical Physics, ranging from fundamental aspects in the theory of non-equilibrium processes to modern applications in biophysics.

Topics covered include diffusion problems, physics of DNA and biomolecules, population dynamics, pattern formation, non-equilibrium transport, bacterial motility, single-molecule kinetics, dynamics and structure of networks, statistical inference, Monte-Carlo simulation techniques, self-assembly, soft condensed matter (colloids, liquid crystals etc.), work relations and fluctuation theorems, and many more.

[Timetable - available shortly before start of the workshop]

Invited Participants

Mikko Alava (Helsinki University of Technology) Tapio Ala-Nissilä (Helsinki University of Technology) Erik Aurell (KTH, Stockholm) Gaute T. Einvoll (Norwegian University of Life Sciences, Aas) Hans Fogedby (University of Aarhus) Thomas Heimburg (University of Copenhagen) Thordur Jonsson (University of Iceland) Heiner Linke (Lund University)

Oksana Manyuhina (NORDITA, Stockholm) Bernhard Mehlig (Göteborg University) Namiko Mitarai (Niels Bohr Institute, Copenhagen) Alessandro Mossa (University of Aarhus) Lene Oddershede (Niels Bohr Institute, Copenhagen) Kim Sneppen (Niels Bohr Institute, Copenhagen) Jonas Tegenfeldt (Lund and Göteborg) Mats Wallin (KTH, Stockholm)

Special guest

Chris van den Broeck, Hasselt University Belgium

Application

If you want to participate in the workshop, please fill in the registration form.

Registration deadline: 20 March 2012

There is no registration fee.

Sponsored by:

Wednesday 28 March

09:30 Welcome coffee & Registration

1 Welcome and Opening Remarks

2 Symmetry breaking in soft matter: from self-assembled vesicles to motile cells

Soft matter essentially differs from hard matter and thus the approaches to study it are also different. Since the typical energy scale between the components of soft matter system is of the order of k_BT, the fluctuating and curved geometries are energetically accessible, and the language of differential geometry becomes useful. When we consider soft materials with characteristic length scale of the order of microns (thousands of molecules) the phenomenology turns out to be an adequate tool to describe the collective behaviour of the system. In this talk I will consider two examples: shape transformation of self-assembled spherical vesicles in presence of magnetic fields and spontaneous crawling of keratocyte cells. Our theoretical findings give insight into recent experimental observations.

Speaker: Dr Oksana Manyuhina (Nordita, Stockholm)

3 Physics of crawling bacteria

I will report on some new theoretical and experimental results for certain species of crawling bacteria. New experiments show that the motion of crawling bacteria display a multitude of different regimes. Both linear motion, random walk, subdiffusion, and superdiffusion can be observed under different conditions. I will discuss how these regimes can be related to the peculiar pilus retraction mechanism for the motion of these bacteria.

Speaker: Prof. Mats Wallin (Theoretical Physics, KTH, Stockholm)

12:00 Lunch

4 Control translations in bacteria

Bacteria controls the levels of proteins via gene regulation. In this talk we focus on the regulation at the translation level, i.e., the process to convert information on mRNAs to proteins. We present three different ways of regulating translation, which are (i) small RNA regulation (ii) regulation of ribosome traffic jam (iii) toxin-antitoxin system. We construct mathematical models of each systems based on experimental knowledge to demonstrate the pros and cons of each regulations.

Speaker: Dr Namiko Mitarai (Niels Bohr Institute, Copenhagen)

5 How and Why for bacteriophages

The talk will discuss survival strategies of bacteriphages, with emphasis on temperate phages and their lysis-lysogeny decision. I will discuss regulatory circuits that count and makes decisions, and it will discuss why stochastic behaviour can be better than deterministic strategies.

Speaker: Prof. Kim Sneppen (Niels Bohr Institute, Copenhagen)

15:00 Coffee break

6 The many faces of the second law

I review some spectacular recent advances in statistical mechanics, including universal features of efficiency of machines at maximum power, Brownian refrigerators, the fluctuation and work theorems, a deeper formulation of the second law, and the splitting of the second law. C. Van den Broeck, "Thermodynamic Efficiency at Maximum Power", Phys. Rev. Lett. 95, 190602, 1-3 (2005). B. Cleuren, C. Van den Broeck, and R. Kawai, "Fluctuation and Dissipation of Work in a Joule experiment", Phys. Rev. Lett. 96, 050601, 1-4 (2006). C. Van den Broeck and R. Kawai, "Brownian Refrigerator", Phys. Rev. Lett. 96, 210601, 1-4 (2006). R. Kawai, J.M.R. Parrondo, and C. Van den Broeck, "Dissipation: The Phase-Space Perspective", Phys. Rev. Lett. 98, 080602, 1-4 (2007). M. van den Broek and C. Van den Broeck, "Chiral Brownian Heat Pump", Phys. Rev. Lett. 100, 130601, 1-4 (2008). M. Esposito, K. Lindenberg, and C. Van den Broeck, "Universality of Efficiency at Maximum Power", Phys. Rev. Lett. 102, 130602 (2009). M. Esposito and C. Van den Broeck, "Three Detailed Fluctuation Theorems", Phys. Rev. Lett. 104, 090601 (2010). M. Esposito, R. Kawai, K. Lindenberg, and C. Van den Broeck, "Efficiency at Maximum Power of Low-Dissipation Carnot Engines", Phys. Rev. Lett. 105, 150603 (2010).

Speaker: Prof. Chris van den Broeck (Hasselt University)

7 Deciding about manuscripts: the dynamics of refereeing

Speaker: Prof. Mikko Alava (Aalto University School of Science, Espoo)

18:00 Reception

Thursday 29 March

8 Evolutionary branching in a stochastic population model

Speaker: Prof. Bernhard Mehlig (University of Gothenburg)

10:30 Coffee break

9 Heat flow in chains driven by noise

We consider the large deviation function for a harmonic chain composed of N particles driven at the end points by heat reservoirs, first derived by Saito and Dhar and Kundu et al. Within a Langevin description we carry out a standard path integral calculation in Fourier space. The large deviation function is given in terms of a transmission Green's function and is, moreover, consistent with the fluctuation theorem. We, moreover, consider an extension of a single particle model suggested by Derrida and Brunet and also discuss the two-particle case. We find a simple expression for the tails of the heat distribution which turn out to decay exponentially. We also discuss the limit of large N and present a closed expression for the large deviation function. Finally, we present a derivation of the fluctuation theorem on the basis of a Fokker-Planck description. This result is not restricted to the harmonic case but is valid for a general interaction potential between the particles.

Speaker: Prof. Hans Fogedby (Aarhus University)

10 Anomalous diffusion on random graphs

We present some results about diffusion on random combs, random trees and related structures. In particular we describe results about the spectral dimension of such random graphs.

Speaker: Prof. Thordur Jonsson (University of Iceland)

12:30 Lunch

11 On properties of optimal heat and work in stochastic thermodynamics

I will consider the problem of minimized (expected) dissipated work or released heat in systems described by over-damped Langevin equation. The problem can be mathematically stated as a standard stochastic optimization problem, but turns out to have a suprisingly simple solution in turns of Burgers equation (or nonlinear diffusion equation) for an auxiliary field, and mass transport by the corresponding velocity field [1]. One application of these results is an improvement of Landauer's bound on the heat released when setting one bit, if it has to be done in a finite time [2]. The refined bound has the form of T log2 + K/t, where T log 2 is the Landauer bound, t is the time of the process and K can be computed from the initial and final states and an appropriate solution of Burgers equation. If temperature and/or the friction coefficient are not constant in time and/or space a similar almost closed formula can be derived, not from the released heat but for the entropy production in the environment [3]. I will discuss the conceptual issues we have encountered in this direction. [1] Erik Aurell, Carlos Mejia-Monasterio, Paolo Muratore-Ginanneschi, Phys. Rev. Lett. 106, 250601 (2011) [2] Erik Aurell, Krzysztof Gawȩdzki, Carlos Mejía-Monasterio, Roya Mohayaee, Paolo Muratore-Ginanneschi [arXiv:1201.3207] [3] Stefano Bo, Erik Aurell, Antonio Celani and Ralf Eichhorn (2012, in preparation).

Speaker: Prof. Erik Aurell (Computational Biology, KTH, Stockholm)

12 Multiscale modeling of cortical columns

Until now most studies of biological neural networks have focused on generic properties, for example, conditions for obtaining various types of spike-train statistics in homogeneous structureless networks (regular vs. irregular, synchronous vs. asynchronous) or formation of coherent structures such as stationary bumps or traveling waves and pulses of neural activity. Now the ambition must be to go beyond this and also model structured networks mimicking particular biological systems, thus allowing for more direct and comprehensive comparison with experiments. Sensory cortical columns in mammals, comprising about 10000-100000 neurons, are prime candidates as model systems as (i) the physiological properties of these cortical neurons and their connections are fairly well mapped out, and (ii) their direct involvement in sensory processing makes them conceptually and technically easier to probe experimentally. Here a multiscale modeling approach for the signal processing properties of such cortical columns is presented and discussed. The approach is multilevel in that the same system is modeled at different levels of detail, just like a gas of molecules both can be modeled at the microscopic molecular level (using Newton’s laws) and at the macroscopic level (using thermodynamics). To allow for model testing, the set of interconnected models must be able to predict what is measured with the various available experimental techniques, and multimodal modeling, i.e., “modeling of what you can measure”, is thus a key part of the approach. As an example we focus on stimulus-evoked responses in the rat barrel cortex, a part of cortex involved in the processing of whisking stimuli. From extracellular potentials recorded with a linear (“laminar”) electrode array spanning the column of the barrel cortex [1], physics-type “multimodal” modeling of the recorded potentials [2,3] are used to extract population firing rates of the salient cortical populations [1]. These are in turn used to estimate population network models for the signal processing done in the column [4]. Finally, preliminary results from attempts to “reverse engineer”, i.e., represent the same dynamics with spiking-neuron network models with thousand of neurons instead of population firing rates, are presented. [1] GT Einevoll et al, J Neurophysiol 97:2174 (2007) [2] KH Pettersen, GT Einevoll, Biophys J 94:784 (2008) [3] KH Pettersen et al, J Comp Neurosci 24:291 (2008) [4] P Blomquist et al, PLoS Comp Biol 5:e1000328 (2009)

Speaker: Prof. Gaute Einevoll (Norwegian University of Life Sciences)

15:30 Coffee break

13 Lipid ion channels and critical opalescence in biomembranes

In the recent years, our group has explored the possibility that electromechanical pulses (solitons) can travel along nerve axons that share many similarities with the action potential of nerves. The above model does not explicitly require ion channel proteins, which are the central elements in the textbook models for the explanation of the nervous impulse. When a voltage is applied across the membrane, ion channel proteins can be recognized in electrophysiological experiments by a quantized change of current intensities in the range of a few Pico-amperes. Thus, these currents exist and are a proven fact. It is therefore important to address the question of the origin of the quantized currents in the soliton model. Here, we show that changes in the conductance of membrane can be the result of critical fluctuations in the lipid membrane that leads to pore formation in the membrane. These pores generate a current signature that is indistinguishable from that of protein channels both in amplitude and channel lifetime. Within a thermodynamic treatment, these phenomena are well explainable by application of the fluctuation-dissipation theorem. The channel lifetimes are shown to correspond to the fluctuation timescales, and the increased channel likelihood is explained by the large increase of the magnitude of the fluctuations close to transitions in the membrane, which lead to a large increase in membrane compressibility. This phenomenon resembles that of critical opalescence in binary mixtures of fluids. We compare experimental traces from biological cells with current recordings from synthetic membranes.

Speaker: Prof. Thomas Heimburg (Niels Bohr Institute, Copenhagen)

14 Optimizing the performance of an artificial protein motor

Biomolecular motors are typically studied in a top-down approach, by observing the function, kinetics, and structure of existing motors. Once one has developed a basic understanding of motor function in this way, it is desirable to test this understanding by attempting to construct a motor from the bottom up. Of particular interest is the use of proteins as building blocks, like biology. Here we present such an ongoing approach. The ‘Tumbleweed’, a synthetic protein motor designed to move along a linear track [1]. This concept uses three discrete ligand-dependent DNA-binding domains to perform rectified diffusion along a synthesized DNA molecule. I will present the motor concept and give an overview on its experimental realization. Then, I will focus on modelling efforts that were used to understand the expected motor performance, and to guide its optimization. [1] B. Bromley, N. Kuwada, M. Zuckermann, R. Donadini, L. Samii, G. Blab, G. Gemmen, B. Lopez, P. Curmi, N. R. Forde, D. N. Woolfson, and H. Linke, The Tumbleweed: Towards a synthetic protein motor. HFSP J. 3, 204 (2009). [2] N. Kuwada, G. Blab, and H. Linke, A Master equation approach to modeling an articial protein motor arxiv.org/abs/1004.1114, accepted by J. Chem. Phys. (2010). [3] Kuwada et al. Tuning the performance of an artificial protein motor. Phys Rev E (2011) vol. 84 (3) pp. 031922

Speaker: Prof. Heiner Linke (The Nanometer Structure Consortium and Division of Solid State Physics, Lund University)

18:00 Conference dinner

Friday 30 March

15 Probing confined DNA

In my presentation I will touch on three topics with fundamental interest and with relevance to biomedical applications. All involve DNA stretched by confinement in nanofluidic channels: (1) orientational correlations, (2) force-extension measurements, and (3) barcode labeling. The behavior of DNA confined into a nanochannel with a effective cross sectional diameter less than the radius of gyration of the polymer have been described by two theories. For effective diameters less than the persistence length Odijk has devised a model where the DNA undulates between the walls. For effective diameters greater than the persistence length deGenne's blob theory predicts a powerlaw dependence of the extension as a function of the effective diameter with an exponent of -2/3. However, although deGennes blob theory correctly predict some of the characteristics of experimental results, the predicted exponent is not reproduced and with careful measurements we show that the behavior does not even follow a power law. Using Monte Carlo simulations together with a mean-field theory we develop a model based on local orientational correlations that better predict the global properties of the DNA that we observe in our measurements. Using force-extension measurements we probe the effect of confinement on the elastic properties of DNA. The results have strong biological relevance due to the high degree of crowding in natural environments. The measurements take place in a device with two microscale channels connected by a nano-slit. The DNA is attached to a magnetic bead and introduced into one of the microchannels. The DNA is subsequently allowed to pass through the nanoslit and finally bound to the surface of the opposite microchannel. With a magnetic tweezers a force is exerted on the DNA while the total extension is observed. For deep slits and for high forces the force extension curve follows the bulk Marco-Siggia model. However for strong confinement and for low forces the fluctuations in one dimension are suppressed, lowering the required force for a given extension. A modified Marco_Siggia model is developed and is found to be consistent with our observations. The direct visualization of DNA stretched in nanochannels opens up for interesting applications in genomics, oncology and infectious disease. We have developed a simple labeling technique that results in a pattern along the DNA that is based on the local melting and thus a function of the underlying sequence. I will discuss the prospects of this technique along with recent results.

Speaker: Prof. Jonas Tegenfeldt (Department of Physics, Lund University and Department of Physics, University of Gothenburg)

16 Unifying model of driven polymer translocation

We present a Brownian dynamics model of driven polymer translocation, in which non-equilibrium memory effects arising from tension propagation (TP) along the cis side subchain are incorporated as a time-dependent friction. To solve the effective friction, we develop a finite chain length TP formalism, based on the idea suggested by Sakaue [Sakaue, PRE 76, 021803 (2007)]. We validate the model by numerical comparisons with high-accuracy molecular dynamics simulations, showing excellent agreement in a wide range of parameters. Our results show that the dynamics of driven translocation is dominated by the non-equilibrium TP along the cis side subchain. Furthermore, by solving the model for chain lengths up to 10^10 monomers, we show that the chain lengths probed by experiments and simulations are typically orders of magnitude below the asymptotic scaling limit. This explains both the considerable scatter in the observed scaling of translocation time w.r.t. chain length, and some of the shortcomings of other present theories. Our study shows that for a quantitative theory of polymer translocation, explicit consideration of finite chain length effects is required.

Speaker: Prof. Tapio Ala-Nissilä (Aalto University School of Science, Espoo and Brown University, Providence, Rhode Island)

11:30 Lunch

17 Non-equilibrium forcing of DNA / RNA structures to twist, stretch, open, or melt

In a living cell DNA and RNA are constantly subject to forces causing the nucleic acid structures to twist, bend, stretch, open, or melt. In order to understand important cellular processes and the physical mechanisms that nucleic acids obey, it is important to know how these nucleic acid structures comply to mechanical stress. By applying forces on individual DNA and RNA molecules, we investigated their response while being forced to undergo a non-equilibrium structural transition. Precise force-extension measurements performed on DNA by optical tweezers prompted a re-formulation of the celebrated worm-like chain model, our new model is denoted the 'twistable worm-like chain' and takes into account the observed twist-stretch coupling [1]. In the DNA overstretching regime, the so-called 'force-plateau', we consistently observed a reproducible rip-like structure in the data which originated from a mechanical unpeeling of the two strands [1]. RNA is another nucleic acid of poss ibly even more importance than DNA.. During translation mRNA pseudoknots are subject to a mechanical force acted upon the structure by the translating ribosome. This force causes the structure to open and occasionally the ribosome to shift reading frame. Using an optical trapping assay we mimicked the action of the ribosome by forcing mRNA pseudoknots to unfold in a non-equilibrium fashion. We found that the frameshifting efficiency correlates with the mechanical strength of the structure [2], however, the relation is not trivial as an inversion of the structure has a tremendous effect both on frameshifting rates, on Gibbs free energies, and on its mechanical strength. Occasionally, the mRNA pseudoknot is so strong that it efficiently acts as a roadblock for e the translating polymerase [3]. [1] Twist, stretch and melt: quantifying how DNA complies to tension. P. Gross, N. Laurens, L.B. Oddershede, U. Bockelmann, E.J.G. Peterman, G.J.L. Wuite, Nature Physics, vol.7 p.731 (2011). [2] Correlation between mechanical strength of messenger RNA pesudoknots and ribosomal frameshifting, T.M. Hansen, S.N.S. Reihani, L.B.Oddershede, M.A. Sørensen, PNAS vol.104 p.530 (2007). [3] mRNA pseudoknots act as ribosomal roadblocks, J. Tholstrup, L.B. Oddershede, M.A. Sørensen, Nucleic Acids Research vol.40 p.303-313 (2012).

Speaker: Prof. Lene Oddershede (Niels Bohr Institute, Copenhagen)

18 Hidden Markov methods for free energy landscape reconstruction

Hidden Markov models (HMM) have been developed during the '60s and '70s in the context of artificial intelligence, and are widely used for complex tasks like speech recognition. In fact, they are a powerful statistical tool for analyzing time series of different nature. After a general overview, I'll focus on some recent success in the application of HMM to the paramount problem of free energy landscape reconstruction in biophysics.

Speaker: Dr Alessandro Mossa (Aarhus University)

19 Concluding remarks