Spring 2015 Schedule

Date Speaker
15 January Karl Lundquist — Gumbart Lab
22 January Kerry McGill — Schmidt-Krey
29 January Michael Tennenbaum — Fernandez-Nieves Lab
5 February Feifei Qian — Goldman Lab
12 February open
19 February Farzan Beroz — Princeton University
26 February Dr. Anton Souslov — Goldbart Lab — Snow Day, Campus Closed
5 March APS Meeting — San Antonio, TX (No PoLS)
12 March Diandian “Diana” Chen — Fenton Group
19 March Spring Break (No PoLS)
26 March James “Jim” Waters — Kim Lab
2 April Travis Tune — Sponberg Lab
9 April Douglas ‘Bo’ Broadwater — Kim Lab
16 April Dr. Joey Leung — Weitz Lab
23 April Casey Trimble — Fenton Lab
30 April Post-doc Candidate — Sponberg Lab

Spring 2015 Abstracts

16 April 2015

Chung Yin “Joey” Leung

Conflicting Attachment and the Growth of Bipartite Networks

Abstract: Network growth models have become an indispensable part of the study of networks. In these models, the network grows by sequential addition of new nodes, where the connectivity of each newly added node is determined by the existing network connectivity and a set of predefined rules describing the network growth mechanisms. A number of network growth mechanisms has been proposed to explain the emergence of structural properties observed in real networks. For example, the widely-studied preferential attachment mechanism assumes that a newly added node is more likely to link to an existing node that is already well-connected. Such a mechanism can explain the power-law degree distribution observed in complex scale-free networks [1]. However, the growth mechanisms of many networks are incompatible with preferential attachment. In many bipartite systems composed of two types of agents that interact, there exists a fundamental conflict of interests as to the preferred level of connectivity. For example, in an interaction network between virus (phage) and microbe (bacteria) where a link denotes the ability for a given phage to infect a bacteria, a high level of connectivity may be beneficial for a phage but detrimental to a bacteria. Recent studies have revealed intriguing structural properties in these infection networks. They exhibit a nested pattern where relatively highly-resistant bacteria are infected by generalist phages and relatively susceptible bacteria are infected by generalist and specialist phages [2]. In addition, they can exhibit modular patterns with groups consisting of phages and bacteria that interact strongly within the group but not between groups [3]. The mechanism for the emergence of these properties is not well understood. We propose a network growth model with a conflicting attachment mechanism inspired by the evolution of phage-bacteria interactions. We find that the resulting networks reproduce realistic patterns found in empirical data including nestedness, modularity, and a nested-modular pattern where modularity is observed at the network scale and nestedness is observed at the module scale. We study the role of conflicting interests in shaping network structure and discuss ways to incorporate greater realism in linking growth process to pattern.

(1) A. L. Barabási and R. Albert, “Emergence of scaling in random networks” Science 1999, 286(5439), 509-512.

(2) C. Flores, J. Meyer, L. Farr, S. Valverde, and J. S. Weitz, “The statistical structure of host-phage interactions” Proceedings of the National Academy of Sciences 2011, 108, E288-E297.

(3) C. Flores, S. Valverde, and J. S. Weitz, “Multi-scale structure and geographic drivers of cross-infection within marine bacteria and phages” ISME Journal 2013, 7, 520-532.


9 April 2015

Douglas “Bo” Broadwater — Kim Lab

The Effects of Mismatch Position on DNA Strand Displacement Rate

DNA strand displacement is a reaction involving three single strands of DNA. It occurs when one single strand of DNA fully hybridizes to a partially duplexed complement. DNA strand displacement is essential to two of the most critical operations in biology, DNA repair and homologous recombination. It is also utilized in the field of nanotechnology in diverse areas such as DNA motors and DNA computation. In this week’s PoLS meeting, I will discuss our work towards tuning the overall rate of strand displacement by varying the position of a single nucleotide mismatch. The ability to tune the displacement rate is important because it has applications in identifying single nucleotide variants in mRNA transcripts. I will provide experimental evidence of a non-trivial relationship between strand displacement rate and mismatch position. Finally, I will posit a model that explains the origins of the complex rate behavior as being due to both secondary structure and mismatch position.


2 April 2015

Travis Tune — Sponberg Lab

Energetic Versatility of Muscle

Muscle is a unique material which has the ability to operate as a spring, motor, brake, or strut which are characterized by a muscle’s workloop. In a typical workloop experiment, force and length of a muscle are plotted against each other while the muscle undergoes oscillation while being activated under physiological conditions. A muscle’s workloop can be altered by changing any of a number of physiological inputs such as temperature, phase or frequency of activation, or operating length. This is what is meant by “Energetic Versatility of Muscle”- the ability of muscle to alter its role quickly and with low energy cost. In this talk I will give an overview of what parameters are known to alter muscle workloops and then will introduce a system of two muscles whose workloops are not explained by our current understanding of muscle physiology. I will then discuss the structure of the muscle as a potential source of variation of workloops. Because skeletal muscle is a highly ordered structure, we can use the technique of x-ray diffraction to examine the lattice structure of muscle. To do this, we traveled to Argonne National Lab to make use of the Advanced Photon Source x-ray beamline. Our early results show what may be a difference in the lattice spacing of the two muscles. We also will perform a frequency analysis on muscle, from which we can extract rates which are related to the molecular processes which are responsible for muscle’s force generation.


26 March 2015

James Waters — Kim Lab

Simulation of fluctuating entropic force by kinetic sampling

Abstract: A polymer chain when pinned to a point or a plane exerts a force on the pinning point in thermal equilibrium. From a thermodynamic point of view, this force arises as the chain tries to minimize its free energy at constant temperature. Due to the hard constraint at the pinning point, calculation of this entropic force is not a trivial problem in statistical mechanics. Here, we introduce a novel kinetic sampling method that can produce the equilibrium distribution of instantaneous forces exerted by a terminally pinned polymer. We represent a generic flexible chain using the canonical bead-rod model and perturb both the positions and momenta of the beads compatible with the spatial constraints and the Boltzmann distribution of total energy. We then calculated the constraint force for each conformation using Lagrangian dynamics. The mean of this force distribution is consistent with the force estimated from the partition function. Unlike the mean that quickly plateaus at tens of monomers, the skewness of the distribution keeps increasing in the range of chain lengths tested. This result suggests that longer chains exert large pulling forces at the pinning point more frequently than short chains. Our work provides insights into the mechanistic origin of entropic forces, and a computational tool applicable to simulations of constrained systems


12 March 2015

Diandian Chen — Fenton Lab

Cellular mechanisms for QRS alternans in the heart

Abstract: Alternans in the electrocardiogram of the heart often precede sudden cardiac death. To analyze the conditions that allow for alternans, scientists examine dynamics at the cellular level. For T-wave alternans, it has been confirmed that the source is due to an alternating action potential duration at the cellular level. However, the cellular mechanisms for QRS alternans have not been well studied. We propose that the source of QRS alternans is due to action potential amplitude alternation at the cellular level. First, we created a simple model that reproduces this effect. Next, we derived a method to analyze the criteria for amplitude alternation in this simple model. Our results show that our method of analysis is correct in predicting the conditions for alternans.


5 February 2015

Feifei Qian — Goldman Lab (Physics)

Legged locomotion on homogeneous and heterogeneous granular ground

Flowable substrates like sand, snow, leaf litter and gravel can be especially challenging for terrestrial locomotors. The objective of our research is to discover principles of ambulatory locomotion on homogeneous and heterogeneous granular substrates and create models of animal and robot interaction with such environments. Since interaction with natural substrates is complicated to model, we take a robophysics approach – we create a terrain generation system that allows systematic exploration of substrate properties, and we use bio-inspired robots as simplified model locomotors to vary morphological and kinematic parameters. We find that terradynamic performance on homogeneous deformable substrates was determined by the leg penetration ratio. Locomotors with smaller foot pressure were less sensitive to ground stiffness variation, and thus could passively maintain relatively effective performance on low resistance ground. At low gait frequencies, hydrostatic-like forces generated during the yielding of the granular material ultimately led to solidification of the material, allowing locomotors to move as if they were walking on a solid surface. At high frequencies, the inertia of the grains being accelerated became important and granular forces became hydrodynamic-like, allowing lightweight locomotors to achieve high locomotor performance by exploiting fluid properties of the granular material.

29 January 2015

Michael Tennenbaum — Fernandez-Nieves Lab (Physics)

Rheological Properties of Fire Ant Aggregations

Fire ants, Solenopsis invicta, form aggregations that are able to drip and spread like simple liquids, but that can also store energy and maintain a shape like elastic solids. They are an active material where the constituent particles constantly transform chemical energy into work. We find that fire ant aggregations shear thin and exhibit a stress cutoff below which they are able to oppose the applied stress. In the linear regime, the dynamics is fractal-like with both storage and shear moduli that overlap for over three orders of magnitude and that are power law with frequency. This dynamic behavior, characteristic of polymer gels and the gelation point, gives way to a predominantly elastic regime at higher ant densities. In comparison, dead ants are always solid-like.


22 January 2015

Kerry McGill — Schmidt-Krey Lab (Biology)

Electron crystallography: A method for protein structure determination

Membrane proteins occupy a variety of functional niches in the cell including structural support, regulation, and communication; understanding how these proteins accomplish their role in the cell can be enhanced with the knowledge of the membrane protein structure. One method of membrane protein structure determination is 2D crystallography. In this method, solubilized membrane proteins are crystallized in lipid membranes via manipulation of the dialysis conditions, like lipid to protein ratio (LPR), salt type and concentration, dialysis length, lipid type and concentration, etc. As these conditions are changed, 2D crystals are screened for via uranyl acetate negative staining and imaging with a transmission electron microscope. Once crystallization conditions are optimized for membrane formation, high resolution images are taken via electron cryo-microscopy (cryo-EM). A crystal structure is determined with the help of image processing. Both membrane bound and membrane associated membranes can be studied by 2D crystallography. Understanding the structure of membrane proteins can lead to insight on how proteins function within the cell.

15 January 2015

Karl Lundquist — Gumbart Lab (Physics)

Insertion of β-barrel proteins in gram-negative bacteria

Gram-negative bacteria possess two phospholipid membranes, the inner and outer of which are composed primarily α-helix and β-barrel proteins respectively. In recent years, significant progress has been made in understanding insertion and assembly of proteins into the inner membrane, while the same process in the outer membrane has remained elusive. In 2013, the crystal structure of BamA, the central and essential component of the β-barrel assembly machinery (BAM), was released, paving the way for rapid progress in understanding the insertion and assembly process. All-atom molecular dynamics simulations have been performed, revealing many novel features including lateral gate opening between the first and last barrel strands, and a destabilized membrane region near the putative insertion site caused by a decreased hydrophobic band. However, many questions remain, including the role of the periplasmic domains, the mode of substrate recognition, and the energetic factors driving function in the absence of both ATP and an electrochemical gradient. Currently, we are performing novel equilibrium simulations of the protein in its native LPS environment, and calculating the free energy of lateral-gate opening for native systems, as well as systems with strand modifications and/or augmentations designed to yield insight into driving energetics and substrate recognition.