Computational and Applied Mathematics Seminar at UC Irvine

TIME:     Mondays, 4pm

PLACE:  Multipurpose Science & Technology Building, Room 122

MAP

ORGANIZER:     Natalia Komarova

TRAVEL AND LODGING INFORMATION FOR SPEAKERS



Fall 2004


Sept 20 :   Speaker:   Jack Xin (University of Texas)  

Title:   An Invertible Auditory Transform

Abstract:   Sound signal processing is based on spectral analysis normally done with Fourier type transforms. In this talk, we discuss an invertible transform with built-in auditory filter characteristics, its mathematical properties and applications.



Sept 21 (Tuesday):   SPECIAL SEMINAR!!! 4pm, Room 254. Speaker:   Dieter Bothe (University of Paderborn, Germany)  

Title:   Two-phase flows with soluble surfactant: local existence of strong solutions

Abstract:   The presence of surfactants, ubiquitous at most gas/liquid interfaces, has a pronounced effect on the surface tension, hence on the stress balance at the phase boundary: local variations of the capillary forces induce transport of momentum along the interface - so-called Marangoni effects. Surfactants are often soluble in one of the adjacent bulk phases, in which case there is also exchange of surfactant between the relevant bulk phase and the interface by adsorption and desorption. Along the interface surfactant is transported by convection and diffusion. Further, changes of the interfacial area due to compression or stretching cause corresponding changes in surfactant concentration.

We discuss the mathematical model governing the dynamics of such systems. This leads to the two-phase balances of mass and momentum, complemented by a species equation for both the interface and the relevant bulk phases. Within the model, the motions of the surfactant and of the adjacent bulk fluids are coupled by means of an interfacial momentum source term that represents Marangoni forces. Employing Lp-maximal regularity we obtain local (in time) strong well-posedness of this model for certain initial configurations. The proof is based on recent Lp-theory for two-phase flows without surfactant. Joint work with Gieri Simonett (Vanderbilt University, Nashville TN) and Jan Pruss (Universitat Halle-Wittenberg, Germany)



Sept 27 :   Speaker:   Zoltan Toroczkai (Los Alamos)  

Title:   Gradient Networks:   From Transport Efficiency in Scale-free Graphs to Social Influence Structures

Abstract:   It has recently been recognized that a large number of complex networks are scale-free, having a power-law degree distribution. Here we propose that the emergence of many scale-free networks is tied to the efficiency of transport and flow processing across these structures. In particular, we show that for large networks on which flows are influenced or generated by gradients of a scalar distributed on the nodes, scale-free structures will ensure efficient processing, while non scale-free structures, such as random graphs, will become congested. As an application, we then make a connection to a simple agent-based model of a market and study the effects of the social network on the evolution of the collective/market.



Oct 4 :   Speaker:   Robyn Araujo (NIH)  

Title:   Paradigms of Stress Evolution in Growing Tumors

Abstract:   The evolution and spatial distribution of tissue stresses is of fundamental importance in a number of physiological phenomena. The experimentally-observed collapse of tumor blood vessels, for example, which has been attributed to the elevated tissue stresses resulting from confined proliferation of tumor cells, represents a significant barrier to the delivery of blood-borne therapeutic agents. Such stresses are residual in nature, arising in the tissue in the absence of external loads, and result from the incompatibility of growth strains.

Nevertheless, the underlying phenomenological determinants of residual stresses, as well as their purpose and implications in both normal tissue development and various pathological conditions, are poorly understood since there is currently a paucity of mathematical models to elucidate these phenomena. In this presentation a number new mathematical ideas germane to the study of residual stresses in growing soft tissues will be discussed. Emphasis will be placed on 'solid-multiphase' tissue modeling, which represents a new class of mathematical models in which the concepts of poroelasticity are extended to accommodate continuous volumetric growth.



Oct 11 :   Speaker:   David Wolpert (NASA)  

Title:   Statistical Physics, Bounded Rationality and Distributed Control

Abstract:   Abstract: A long-running difficulty with conventional game theory is how to modify it to accommodate bounded rationality. A recurring issue in statistical physics is how best to approximate joint probability distributions with decoupled (and therefore far more tractable) distributions. A major problem in control theory is how to implement control on (massively) distributed systems, especially in an adaptive manner, with mixed types of control variables.

This talk shows that the same information-theoretic structure, known as Probability Collectives (PC), underpins all three issues. This means that statistical physics, game theory, and distributed control are fundamentally identical. Accordingly techniques and insights from one of those fields can be applied to the others. One example of this, presented here, is the use of the grand canonical ensemble of statistical physics to elaborate game theory in which the number of players is not pre-determined, but varies stochastically. Another example is how to apply steepest descent techniques to optimize/control systems of discrete variables.



Oct 18 :   Speaker:   Arnold Kim (UC Merced)  

Title:   Light transport in two-layer tissues

Abstract:   Light propagation in tissues is governed by the theory of radiative transport. The radiative transport equation takes into account absorption and scattering due to inhomogeneities. A two-layer medium is a useful model for tissues because it accounts for the differences in optical properties between the superficial and deep regions of tissues. We are interested in probing only the superficial layer because most pre-cancerous tissues develop there. To do this we introduce an alternate boundary condition that allows for the removal of the bottom layer from the problem. For the case when the top layer is thin, we compute an asymptotic solution. We validate our results by comparing them with numerical solutions.



Oct 25 :   Speaker:   Shaoqiang Tang (Peking University)  

Title:   Simulating Semiconductor Charge Transport

Abstract:   Computer simulations of charge transport in semiconductor devices (like diodes and micro-chips) are used by the semiconductor industry as a tool for reducing the cost of developing new devices and new process technologies. At the scale of micron or sub-micron, the semiconductor Boltzmann equation is the most exact model. In order to alleviate computing load, macroscopic models have been derived, assuming that the state of the electron gas is described by certain averaged quantities. These models take similar forms as those in fluid mechanics, and we may apply CFD techniques to probe this promising field of academic importance and commercial value.

We shall present some of our recent results. First, by simulating a hydrodynamic model, we demonstrate the (direct) applicability of CFD techniques. Secondly, the continuing trend of scaling-down and speed-up makes the modeling and computing of quantum effect and transient behavior among the top issues in semiconductor research. Careful numerical tests helped identifying well-posedness problem in a quantum hydrodynamic (QHD) model. Viscous QHD model derived from a Wigner Fokker-Planck equation yields more reliable numerical results, and demonstrate interesting nonlinear phenomena, such as negative differential resistance and hysteresis.



Nov 1 :   Speaker:   Don Saari (UCI)  

Title:   Evolution of the universe

Abstract:   We know that the Newtonian N-body problem cannot be solved in a normal sense. On the other hand, we can find all possible asymptotic behaviors as time goes to infinity of all possible solutions for all possible values of N. That is, we can describe the evolution of Newton's universe. In doing so, I will introduce some of the history of the problem showing where "chaos" came from, etc.



Nov 8 :   Speaker:   Laura Biven (Max-Planck-Institut fur Physik Komplexer Systeme and Bard High School Early College New York)  

Title:   Weak Wave Turbulence and its Challengers

Abstract:   I will begin my talk with a brief overview of WWT during which I aim to give an intuitive picture of the phenomenon using the example of surface water waves. This example will be revisited throughout the talk. Next I will try to give a welcoming (although selective) introduction to the calculations of WWT. Equipped with the results of these calculations, I will discuss the relationship between WWT, power-law spectra (both Kolmogorov-Zakharov and MMT) and intermittency. The challengers to WWT are highly nonlinear events, breakdown and the alternative symmetries of the governing equation. I will make some remarks which point out the interconnectedness of these phenomena and, simultaneously, the goals of my research interests.



Nov 15 :   Speaker:   Ziyad Muslimani (University of Central Florida)  

Title:   Multiscale Asymptotic Analysis of Wave Propagating in Nonlinear Periodic Media

Abstract:   New models describing wave propagation in transversely modulated optically induced waveguide arrays are proposed. In the weakly guided regime, a discrete nonlinear Schrodinger equation with the addition of bulk diffraction term and an external ``optical trap'' is derived. In the defocusing regime the optical trap induces a stable localized mode. In the limit of strong transverse guidance, the dynamics is governed by a model which represents the optical analogue of wave action.



Nov 22 :   Speaker:   Darryl Holm (Los Alamos and Imperial College London)  

Title:   Introduction and recent results for LANS-alpha, the Lagrangian averaged Navier-Stokes alpha model of turbulence

Abstract (click)  



Nov 29 :   Speaker:   Simon Tavare (University of Southern California)  

Title:   Stochastic models in colon cancer

Abstract:   I'll begin by describing our recent attempts to model the evolution of crypts in the colon, using methylation patterns as markers [see Yatabe et al.]. Mutations in colon crypts are thought to play an important role in pre-tumor progression, and therefore in understanding the time to cancer. One common complaint about such multistage and multihit models is that they require unrealistically high mutation rates to explain the observed incidence of cancer. I'll use our model, together with classical extreme value theory, to show that we can explain the SEER incidence data for colon cancer using typical mutation rates [Calabrese et al.]. A number of corroborative datasets and open problems will be discussed.



Dec 6 :   Speaker:   Natalia Komarova (UCI)  

Title:   Mathematical modeling of cancer

Abstract:   I will give an overview of the recent work I have done on stochastic modeling of cancer. I will first talk about the concept of multistage carcinogenesis and how we can describe cancer as "bad evolution" within an organism. I will introduce some simple models and explain the phenomenon of "stochastic tunneling". Then I will talk about the role of stem cells in cancer initiation and present some hypotheses about the cellular origins of colon cancer.

Finally, I will talk about growing cellular colonies and models of treatment: how does resistance arise and what can we do about it? Therapies which target specific molecular alterations in cancer cells have shown promising results. Resistance, however, poses a problem, especially in advanced disease. An example is the treatment of chronic myeloid leukemia (CML) blast crisis with Gleevec. I will elucidate the principles which underlie the emergence of drug resistance in cancer. The model (a birth-death process on a combinatorial mutation network) is based on measurable parameters: the turnover rate of tumor cells, and the rate at which resistant mutants are generated. In the context of CML, the prediction is that a combination of three drugs can successfully treat blast crisis.



Winter 2005


Jan 10 :   Speaker:   Royce Zia (Virginia Tech)  

Title:   American Football, Barberpoles and Clouds: Pattern Formation in Biased Diffusion of Two Species

Abstract:   Motivated by several physical systems, we study a simple model of driven, two-species lattice gases. Our system consists of only two types of NON-interacting (apart from an excluded volume constraint) particles, diffusing on a periodic lattice and with biased moves in opposite directions. On a square system, increasing the overall particle density leads to a transition - from a homogeneous phase with high particle current to one with spatial structure and minimal current. For rectangular cases, several structures can appear, with relative frequency depending on the aspect ratio of the system. Using a simple continuum theory, we are able to describe much of the novel transitions. Variations and generalizations, as well as the physical systems they model, will be discussed.



Jan 18 (Tuesday):   SPECIAL SEMINAR!!! 4pm, Room 254. Speaker: Donald Nelson (Worcester Polytechnic Institute) 

Title:   Success of a continuum lagrangian in twentieth-century solid state physics

Abstract:   The development of quantum mechanics in the 1920s convinced physicists that future progress in understanding solids would flow from that discipline, and so continuum mechanics of solids was abandoned by physicists. Thus, success of the latter in late twentieth century is regarded as surprising by most.

We present an overview of the construction of a very general Lagrangian of a closed system of a dielectric crystal interacting with the electromagnetic field. The Lagrangian is first constructed for discrete particles, a long-wavelength (continuum) limit is taken in a manner to preserve all of the eigenmodes. The crystal can be of any class of symmetry, have any structural complexity, and have interactions between its various eigenmodes and between them and the electromagnetic field to any order of nonlinearity. All eigenmodes are included: electromagnetic, acoustic and optic modes of vibration, spin, and all polaritonic combinations of them.

The photoelastic effect was show to have been wrongly formulated for 155 years: the independent variable characterizing the deformation had been wrong! Thus, the interaction tensor has a more general symmetry. The accepted relation between the photoelastic effect and electrostriction was shown to have been wrong for almost as long. The elastic stiffness tensor was shown to lose its traditional symmetry when a soft optic mode became involved. All treatments of acoustic harmonic generation in piezoelectrics were shown to be wrong. The best derivation of optical activity was shown to have missed a fundamental contribution having a different dispersion. The Abraham - Minkowski controversy about the momentum of a light wave in a medium was resolved. The most general Poynting vector in a medium was found. Several nonlinear interactions were characterized and interpreted for the first time.



Jan 24 :   Speaker:   Peter Dodds (Columbia University)  

Title:   Models of Social and Biological Contagion: are Puma shoes some kind of virus?

Abstract:   I will discuss two simple models of contagion relevant to the desciption of social and biological spreading processes.

The first model aims to unify existing models of the spread of social influences and infectious diseases. This generalized model of contagion incorporates individual memory of exposure to a contagious entity (e.g., a rumor or disease), variable magnitudes of exposure (dose sizes), and heterogeneity in the susceptibility of individuals. Through analysis and simulation, we have examined in detail the mean-field case where individuals may recover from an infection and then immediately become susceptible again. We identify three basic classes of contagion models: epidemic threshold, vanishing critical mass, and critical mass respectively. The conditions for a particular contagion model to belong to one of the these three classes depend only on memory length and the probabilities of being infected by one and two exposures respectively. (For both models, a key quantity is the fraction of vulnerables, i.e., individuals who are typically infected by one exposure.) These parameters and their elaborations are in principle measurable for real contagious influences or entities, suggesting novel measures for assessing (as well as strategies for altering) the susceptibility of a population to large contagion events. We also study the case where individuals attain permanent immunity once recovered, finding that epidemics inevitably die out but may be surprisingly persistent when individuals possess memory. I will also discuss some related work by others.

The second model describes the spreading of social influences on networks, and is a natural extension of the threshold model due to Granovetter. For this model on various kinds of random networks, analytic results are known for when cascades (epidemics) are possible. In all cases, the density of the network must belong to an intermediate range referred to as the cascade window. When links are scarce, not enough individuals are connected for global spreading to occur, and when links are overly abundant, too few individuals are vulnerable. In our recent work for this model, we have examined the role of influentials (a.k.a. opinion leaders or, for a biological feel, super-spreaders). We examine cascades after they have occurred, as is invariably done for real cascades. Contrary to much ascribed to influentials, we find that highly connected nodes are not the chief determinants of whether or not a cascade will occur. While cascade initiators are typically more connected than the average individual, the discrepancy is not pronounced. We further observe that cascades arise through a multi-step process and that for dense networks, `early adopters' may in fact be less connected than on average. Also, for dense networks, cascades rapidly take off after a long and `quiet' build up period, making them difficult to identify until after they have been realized. In sum, influentials are limited in their effect since the condition for a cascade to occur is really a global one; there must be a sufficient population of vulnerables available, and it is the most influential of these vulnerables that dictate the spread of an influence.



Jan 31 :   Speaker:   Alan Newell (University of Arizona)  

Title:   Fibonacci and Plants

Abstract:   For over four hundred years, natural scientists have been intrigued and mystified by patterns appearing on plants and by the appearance of Fibonacci sequences when one counts the numbers of arms in the families of spirals on which the primordia of the plant surfaces lie. To date, there has been no widely accepted mechanistic explanation for these observations. I hope that this lecture goes some way towards providing answers.



Feb 7 :   Speaker:   Emmanuel Candes (Caltech)  

Title:   Curvelets and Wave Equations: Theory and Potential for Scientific Computing

Abstract:   This talk explores the potential of new geometric multiscale ideas in the area of partial differential equations. We present a recently developed multiscale system - curvelets - based on parabolic scaling, in which basis functions are supported in elongated regions obeying the relation width ~length^2. This system provides optimally sparse representations of the solution operators for a large class of symmetric systems of linear hyperbolic differential equations - such as the wave propagation operator. This has important implications both for analysis, and for numerical applications, where sparsity allows for faster algorithms. In the second part of the talk, we report on preliminary calculations which suggest that it is possible to derive accurate solutions to a wide range of differential equations in O(N log N) where N is the number of voxels; this complexity holds for arbitrary initial conditions. This is joint work with Laurent Demanet (Caltech)



Feb 14 :   Speaker:   Carlos Castillo-Chavez (Cornell University and Arizona State University)  

Title:   Mathematical Model Applications to Disease and Homeland Security

Abstract:   The events of 9/11 in the US changed the way we look at routine activities such as air and mass-transportation travel. We (as a society) are somewhat prepared to respond to natural acts (epidemics, earthquakes, etc.) but have no data or reliable information that would help in the planning or identification of a set of responses if a deliberate act (against unsuspecting population) were to take place. I will highlight some of the challenges that we face and outline the use of mathematical models in our efforts to meet some of them. I will use recent SARS and foot and mouth epidemics to ground some of the ideas. Should we prepare for worst case scenarios? If so, how do we define worst case scenarios mathematically? I will conclude with the use of some of these ideas on the potential impact or consequences associated with the deliberate release of a biological agent in the mass transportation system of a major metropolitan area.

Mathematical Models and Their Application to the Spread and Control of Tuberculosis

Tuberculosis high levels of prevalence in the world have been the norm, particularly in poor and/or developing nations. The impact of travel and immigration as well as the costs associated with the TB treatment and the consequences associated with treatment compliance (antibiotic resistance) will be discussed. The application of mathematical models in the evaluation of epidemiological and sociological factors associated with TB dynamics and its control at the population level will be highlighted.



Feb 21 :   NO SEMINAR  

Title:  

Abstract:  



Feb 28 :   Speaker:   Hans Othmer (University of Minnesota)  

Title:   The Mathematical Challenge of Multiscale Modeling in Biology: From signal transduction to spatial pattern formation

Abstract:   In the last two decades enormous progress has been made on understanding molecular details in a number of cellular processes such as signal transduction and gene control, but frequently the objective in modeling is to understand the population-level behavior of cells. This gives rise to the problem of how to incorporate sufficient microscopic-level information into macroscopic-level descriptions. In this talk we will discuss two systems that involve chemotaxis, one for which this has problem has been more-or-less solved, and one for which a great deal remains to be done.

Chemotaxis in the bacterium E. coli is widely-studied because of its accessibility and because it incorporates processes that are important in the response of numerous sensory systems to stimuli: signal detection and transduction, excitation, adaptation, and a change in behavior. Quantitative data on the change in behavior is available for this system, and the major biochemical steps in the signal transduction/processing pathway have been identified. We will discuss a mathematical model of single cells that can reproduce many of the major features of signal transduction, adaptation and aggregation, and which incorporates the interaction of the chemotactic protein CheY_p with the flagellar motor. We shall then address the problem of how to obtain macroscopic equations for population-level behavior that incorporate certain features of the microscopic model.

Many cells such leukocytes (cells of the immune system) also respond chemotactically to external signals, but the process by which they determine directional information and alter their pattern of movement is much more complex than in bacteria, and the micro-to-macro step is much more difficult. In the remainder of the talk we will discuss recent progress and open questions in this area.



Mar 7 :   Speaker:   David Chambers (Lawrence Livemore National Lab)  

Title:   Target characterization using time reversal symmetry of wave propagation

Abstract:   The fact that wave propagation looks the same whether time is going forward or backward has been know theoretically since the formulation of wave theory. Only recently, however, has array technology and computers been developed to the point that time reversal of waves can actually be performed in real systems. Experiments using ultrasonic and underwater acoustical arrays have shown enhanced focusing, communications, and imaging through complicated media. Better theoretical understanding of the time reversal symmetry for acoustic and electromagnetic waves has motivated new techniques for imaging and characterization of targets applicable to more conventional array technology. In this talk, these new techniques for target characterization and imaging will be discussed along with examples using both experimental and computational data. It is shown that the time reversal properties of an array system can be predicted by performing a singular value decomposition of the multistatic data matrix. The spectrum of singular values and the form of the singular vectors are related to the physical properties of the target in the field of view of the array. This relationship is described for a number of simple cases and imaging techniques that exploit their properties are shown.



Mar 11 (Friday), MSTB 254:   SPECIAL SEMINAR!!! 4pm, Speaker:   Dongho Chae (South Korea)  

Title:   On the Regularity Conditions for the Navier-Stokes and the Related Equations.

Abstract:   In this talk I present my recent results on the regularity conditions for a solution to the 3D Navier-Stokes equations with powers of the Laplacian, which incorporates the vorticity direction and its magnitude simultaneously. For the proof of the we exploit geometric properties of the vortex stretching term as well as the estimate using the Triebel-Lizorkin type of norms.



Mar 14 :   Speaker:   Eric Mjolsness (UCI)  

Title:   Some Mathematical Problems in Computational Systems Biology

Abstract:   Biochemical reaction networks provide a paradigm for many dynamical systems in biology. The paradigm can be generalized to describe "variable-structure systems" in which objects larger than molecules (such as cells) also change in number and in their relationships over time. In the course of building mathematical and software tools for understanding such networks and dynamical systems, we have identified some interesting applied mathematical problems whose reformulation and solution would be very useful in current computational biology. For example, we can identify partial differential equations whose solution would be especially instructive for enzyme kinetics. These problems arise at the level of small reaction networks, multimolecular complexes, and the development of multicellular tissues. Developmental examples include modeling the shoot meristem of a plant.

A common mathematical framework for models at these different spatial scales can be given in terms of "dynamical grammars". In a dynamical grammar, an input/output syntax for an elementary chemical or biological processes is mapped to an operator algebra expression for the generator of the temporal dynamics associated with that process. Many processes act simultaneously (in parallel) if their generators are summed. Contingent spatial relationships are expressed in terms of dynamical graph grammars, whose formulation could perhaps be improved by use of ideas from topology and differential geometry. By solving such problems, we may hope to construct a useful modeling language of sufficient generality to describe multiscale, variable-structure dynamical systems that arise naturally in biology.



Mar 21 :   Speaker:   Alexander Kiselev (Wisconsin)  

Title:   Quenching of reaction by fluid flow

Abstract:   We consider the problem of quenching the flame in a framework of passive reaction-diffusion model. We ask which flows are more efficient in supressing reaction, and prove bounds on the relationship between flow strength and the initial flame size for different classes of flows. The estimates we prove agree very well with numerical experiments carried out in collaboration with astrophysics ASC group at the University of Chicago. The problem is closely related to proving norm bounds for the evloution semigroup corresponding to the passive scalar model. The techniques involve PDE and probability tools, and further natural questions indicate interesting links with spectral theory of elliptic operators and dynamical systems.



Spring 2005


Apr 4 :   Speaker:   Howard Levine (Iowa State University)  

Title:   A mathematical model for the regulation of tumor dormancy based on enzyme kinetics

Abstract:   We present a two compartment model for tumor dormancy based on an idea of Zetter to wit: The vascularization of a secondary (daughter) tumor can be suppressed by inhibitor originating from a larger primary (mother) tumor. We apply this idea at the avascular level to develop a model for the remote suppression of secondary avascular tumors via the secretion of primary avascular tumor inhibitors. The model gives good agreement with experimental observation (Derm. Surg. 29(2003) 664-667). The authors reported on the emergence of a polypoid melanoma at a site remote from a primary polypoid melanoma after excision of the latter . The authors observed no recurrence of the melanoma at the primary site, but did observe secondary tumors at secondary sites five to seven centimeters from the primary site within a period of one month after the excision of the primary site. We attempt to provide a reasonable biochemical/cell biological model for this phenomenon. We show that when the tumors are sufficiently remote, the primary tumor will not influence the secondary tumor while, if they are too close together, the primary tumor can effectively prevent the growth of the secondary tumor, even after it is removed. It should be possible to use the model as the basis for a testable hypothesis which could be checked in a controlled in vitro experiment.



Apr 11 :   Speaker:   Miguel Dumett (USC)  

Title:   Least squares minimization to estimate the transport of alcohol in the human body

Abstract:   Experimental measurements of transdermal vapor alcohol concentration are used to estimate alcohol concentration in the body using an inverse problem approach. First we propose a model for the transport of alcohol from blood compartments to the skin surface and use the transdermal measurements to estimate the signal obtained by a breathalyzer which is the standard for blood alcohol concentration. Later we couple our skin model to a body model of the human body. The human body is divided in several compartments to facilitate the description of the transport of alcohol in the human body from ingestion to elimination. The adjoint method is used for the computation of the least squares functional gradient. Parameters of the model are estimated using real breathalyzer and a transdermal alcohol skin device data applied to individuals in a hospital. The parameter values obtained are used to predict the evolution of alcohol concentration for patients in the field. Kalman filtering techniques can be used to correct predictions in real time.



Apr 18 :   Speaker:   Yang Kuang (Arizona State University)  

Title:   Modeling insulin secretion ultradian oscillations with two time delays

Abstract:   In the glucose-insulin regulatory system, insulin secretion oscillates with a period of 50-150 minutes. Over the past decade, several mathematical models have been proposed to model these ultradian oscillations as well as the metabolic system producing them. However these existing models yield profiles deviant from a normal physiological range. We introduce a DDE (delay differential equation) model with two discrete delays for better understanding and more accurately modeling the glucose-insulin dynamics and the insulin secretory oscillations. With the same set of experimental data used to test other existing models, the simulation profiles obtained from this two time delay model fall within a normal physiological range.



Apr 22 :   Speaker:   Jian-Guo Liu (University of Maryland)  

Title:   Divorcing pressure from viscosity in incompressible Navier-Stokes dynamics

Abstract:   The pressure term has always created difficulties in treating the Navier-Stokes equations of incompressible flow, reflected in the lack of a useful evolution equation or boundary conditions to determine it. In joint work with Bob Pego and Jie Liu, we show that in bounded domains with no-slip boundary conditions, the Navier-Stokes pressure can be determined in a such way that it is strictly dominated by viscosity. As a consequence, in a general domain with no-slip boundary conditions, we can treat the Navier-Stokes equations as a perturbed vector diffusion equation instead of as a perturbed Stokes system. We illustrate the advantages of this view by providing simple proofs of (i) the stability of a difference scheme that is implicit only in viscosity and explicit in both pressure and convection terms, requiring no solutions of stationary Stokes systems or inf-sup conditions, and (ii) existence and uniqueness of strong solutions based on the difference scheme.

A preprint is available at http://arxiv.org/abs/math.AP/0502549



Apr 25 :   Speaker:   Hong Qian (University of Washington)  

Title:   Modeling biochemical systems with differential equations, stochastic processes, and constraint-based optimizations

Abstract:   With the demand from modeling systems level cellular biochemistry as a reaction network, different applied mathematical approaches are now being pursued. I will discuss three approaches based on (1) systems of ODE with nonlinearity, (1) stochastic processes with irreversibility, and (3) constraint-based optimization suggesting an oriented matroid. A unifying theme of these approaches is the nonequilibrium thermodynamics of living (open) systems.



May 2 :   Speaker:   No seminar  

Title:  

Abstract:  



May 9 :   Speaker:   Lenya Ryzhik (Chicago)  

Title:   Reactive fronts in Boussinesq flows

Abstract:   I will describe some recent results on front propagation in a fluid flow in the Boussinesq approximation. A reaction-diffusion-advection equation is coupled to the fluid flow equation by a temperature dependent buoyancy force. We show that the problem admits non-planar travelling front solutions and that the fluid coupling speeds up the fronts.



May 16 :   Speaker:   Ruo Li (Caltech)  

Title:   A Moving Mesh Method Based on Harmonic Mapping and Its Application

Abstract:   In this talk, I will introduce briefly a moving mesh method based on harmonic mapping. As a rare character, the unique existence of the harmonic mapping is the basic motivation for us to develop this method. The method is implemented in finite element and an iterative procedure is adopted to avoid mesh tangling caused by numerical factors. Our method can move the mesh interior the domain and mesh on the boundary in coupling for both 2D and 3D problems. The moving mesh module can be a black box added on the whole solver to the PDE under consideration that it is very convenient for coding - no modifications to the solver of the PDE are required. The inter-mesh mesh updation is implemented by a linear convection equation instead of generally adopted interpolation methods, thus the method can be easily to applied to problem as incompressible Navier-Stokes equation where the divergence free interpolation can be a big problem, and problem as conservation laws where the conservative interpolation is not trival to be implemented. Numerical results including viscos Burgers equation, reaction-diffusion equation, incompressible Navier-Stokes equation and its coupling with level set method will be shown.



May 18 (Wednesday):   SPECIAL SEMINAR!!! 4pm, Room 254. Speaker:   Kathy Lu (UCSD)  

Title:   Distributed Linear System Solvers: Mathematical Algorithms and Biological Applications

Abstract:   Partial differential equations were discretized using the mortar finite element method, where the mortar space contains piecewise quadratic and cubic functions. We first proved the wellposedness of the saddle point system. To solve this saddle point system, the existing domain decomposition (DD) algorithm of the linear system solver requires the communication of both the interface solution and the interface residual of every local problem. We have developed a new algorithm where only the interface solution is communicated, to accommodate globally nonconforming meshes. The resulting communication complexity is reduced. The scalability and parallel efficiency of this new algorithm were tested with a highly adaptive mesh. As the number of processors increases, linear and logarithmic speed-up of the solving time were observed with this new linear system solver, for the convection-diffusion equation and the Poisson equation respectively.

Three-dimensional dynamic simulation in computational biology provides an emerging field for the application of efficient distributed linear system solvers. We developed a 3D continuum model to investigate the role of structural and functional cellular components in regulating synchronized calcium signaling (SCS), characterized by high gradient near the t-tubule membrane and low gradient in the cytoplasm along the transverse direction, which enables ventricular myocyte to respond rapidly and forcefully to electrical andchemical stimuli. The distributed linear system solver improved the simulation speed by ~10 folds. Simulation results suggest that both t-tubule structure and the spatially heterogeneous distribution of calcium-handling-proteins are important for SCS. The model also predicts that two aspects of heterogeneous distribution are required: the concentration of calcium-handling-proteins in the t-tubule membrane to be ~6 times of that in the surface membrane; and the concentration of L-type calcium channels, in the cytoplasmic end of the t-tubule, to be ~2.3 times of that in the surface membrane end. These results have provided a foundation for further studies on the effects of three- dimensional t-tubule geometry and ion channel distribution on calcium dynamics.



May 23 :   Speaker:   Hal Stern (UCI)  

Title:   Assessment of Ancestry Probabilities in the Presence of Genotyping Errors

Abstract:   This talk discusses an extention of a Bayesian approach for estimating the ancestry probability, the probability that an inbred line is an ancestor of a given hybrid, to account for genotyping errors. The effect of such errors on ancestry probability estimates is evaluated through simulation. The simulation study shows that if misclassification is ignored, then ancestry probabilities may be slightly overestimated. The sensitivity of ancestry probability calculations to the assumed genotyping error rate is also assessed. Finally we briefly discuss approaches for estimating the error rate from limited data.



Date to be determined :   Speaker:   Wayne Hayes (UCI)  

Title:   Rigorous Shadowing of Numerical Solutions of Ordinary Differential Equations by Containment

Abstract:   An exact trajectory of a dynamical system lying close to a numerical trajectory is called a shadow. We present a general-purpose method for proving the existence of finite-time shadows of numerical ODE integrations of arbitrary dimension in which some measure of hyperbolicity is present. Much of the rigor is provided automatically by interval arithmetic and validated ODE integration software that is freely available. The method is a generalization of a previously published containment process that was applicable only to two-dimensional maps. We extend it to handle maps of arbitrary dimension, and finally to ODEs. The method involves building $n$-cubes around each point of the discrete numerical trajectory through which the shadow is guaranteed to pass at appropriate times. The proof consists of two steps: first, the rigorous computational verification of a simple geometric property we call the inductive containment property; and second, a simple geometric argument showing that this property implies the existence of a shadow. The computational step is almost entirely automated and easily adaptable to any ODE problem. The method allows for the rescaling of time, which is a necessary ingredient for successfully shadowing ODEs. Finally, the method is local, in the sense that it builds the shadow inductively, requiring information only from the most recent integration step, rather than more global information typical of several other methods. The method produces shadows of comparable length and distance to all currently published results. We will also briefly mention how a cheaper, non-rigorous algorithm can be used to bolster confidence in large numerical simulations of physical systems.



Jun 6 :   Speaker:   Lisette de Pillis (Harvey Mudd College)  

Title:   Modeling Cancer, the Immune System, and Metabolic Function

Abstract:   We will discuss two approaches for modeling cancer growth. One approach is to employ a deterministic space-independent model that includes separate components to represent specific and non-specific immune function. The other approach employs a spatially dependent hybrid cellular-automata (HCA) model whose rules are driven by cellular metabolic function. For the determinstic model, numerical simulations of mixed chemo-immuno therapy and vaccine therapy using both mouse and human parameters are presented. We illustrate situations for which neither chemotherapy nor immunotherapy alone are sufficient to control tumor growth, but in combination the therapies are able to eliminate the entire tumor burden. The HCA model is in its initial stages of development, and preliminary results will be presented.

This is joint work with Ami Radunskaya (Pomona College) and Weiqing Gu (Harvey Mudd College).