For Math. Biol. Seminar of Fall, 2000, see

 Math. Biol. Seminar

Mathematical Biology Seminar, Spring, 2000

Seminars in Dept. of Biol., Spring, 2000

Department of Mathematics,Arizona State University

CO-SPONSORED BY THE CENTER FOR SYSTEMS SCIENCE AND ENGINEERING

Default time: Friday 12:40p.m. - 1:30p.m. Time and place may change for some talks 
Default place: PSA 106
Organizer: Steven Baer, Frank Hoppensteadt, Yang Kuang (kuang@asu.edu), Hal Smith (halsmith@math.la.asu.edu), Horst Thieme (thieme@math.la.asu.edu)



COMING TALKS

************************************************************************
************************************************************************
MONDAY, May 1, 2000
************************************************************************
        MATHEMATICAL BIOLOGY SEMINAR                 PSA 307  12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        S.A.L.M. Kooijman, Free University, Amsterdam
          "The Synthesizing Unit as Model for the Interaction of
          Substrates in the Uptake by Organisms"
************************************************************************
TUESDAY, May 2, 2000
************************************************************************
        COMPUTATIONAL AND APPLIED MATH PROSEMINAR    GWC 604  12:30 p.m.
        Rosemary Renaut, Department of Mathematics
          "Kinetic Parametric Estimation Using Positron Emission
          Tomographic Data"
        ABSTRACT:  In this talk I will present a simple model that
        describes FDG dynamics in the brain. Using PET data to estimate
        the output from this model, we obtain an inverse problem for
        the determination of kinetic rate constants for different
        tissue types in the brain. I will describe several solution
        techniques based on least squares, and total least squares
        algorithms. This is initial work on which the development of a
        whole brain parametric imaging technique will be based. The
        research is joint with Cristina Negoita in our department, and
        Dr. Kewei Chen of the Good Samaritan PET Center in Phoenix.
See http://plato.la.asu.edu/compsem.html
 

Talks Given

***********************************************************************

FRIDAY, January 28, 2000
************************************************************************
        MATHEMATICAL BIOLOGY SEMINAR                 PSA 106  12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Yang Kuang, Department of Mathematics
          "Theoretical Frameworks for Ecological Dynamics Subject to
          Stoichiometric Constraints"
        ABSTRACT:  All organisms are composed of multiple chemical
        elements such as carbon, nitrogen, and phosphorus.  Although
        the relative abundance of these chemical constituents is known
        to vary considerably among species and across trophic levels,
        most ecological studies have until very recently ignored the
        sources and consequences of this chemical heterogeneity.
        However, rapidly accumulating evidence suggests that the
        dynamic implications of chemical heterogeneity among species
        deserves much more study than it has yet received. This body of
        research, which is to date chiefly empirical in nature, places
        major emphasis on the consequences of chemical heterogeneity
        among species for consumer-resource dynamics and nutrient
        recycling in ecosystems.
          In this talk, we will outline a theoretical framework for
        ecological dynamics that explicitly incorporates stoichiometric
        constraints. The base model involves a stoichiometric counterpart
        of the familiar Rosenzweig-MacArthur equations in which the
        effective carrying capacity of the resource species and the
        transfer efficiency of the consumer species are constrained by
        stoichiometric principles. Introduction of stoichiometric
        considerations in these equations (here, akin to "food quality")
        allows for a rich array of ecologically realistic dynamics,
        including deterministic extinction of the consumer species when
        resources are abundant but of poor quality.
          We will expand this model in several different directions, to
        explore ecological realities whose consideration has proved
        illuminating in other, non-stoichiometric settings.
************************************************************************

FRIDAY, February 4, 2000
************************************************************************
     MATHEMATICAL BIOLOGY SEMINAR                PSA 106   12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Horst Thieme, Department of Mathematics
          "What can we Learn from the Most Elementary Stage-Structured
          Population Model?"
************************************************************************MONDAY, February 14, 2000
***********************************************************************
Note different day and time for this seminar:
        MATHEMATICAL BIOLOGY SEMINAR                PSA 103   3:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Duane Nykamp, Courant Institute of Mathematical Sciences, NYU
          "A Population Density Approach That Facilitates Large-Scale
          Modeling of Neural Networks"
        ABSTRACT:  The neural networks of even small functional units
        in the brain are enormously complex. Conventional simulation
        methods, where one models thousands of individual neurons, can
        take large amounts of computer time even for models of small
        cortical areas.  The population density approach can be used
        to speed up large-scale neural network simulations. In this
        method, ones groups neurons into large populations of similar
        neurons. By calculating the evolution of a probability density
        function for each population, one obtains population firing
        rates and the distribution of neurons over state space.
        I demonstrate a population density method for simulating
        networks of integrate-and-fire neurons with instantaneous
        synapses or with slow inhibitory synapses.  Through comparisons
        with conventional Monte-Carlo simulations for a model of a
        hypercolumn in cat visual cortex, I demonstrate the speed and
        accuracy of the population density method.

FRIDAY, March 3, 2000
************************************************************************
        Joint
        MATHEMATICAL BIOLOGY SEMINAR
        DYNAMICAL SYSTEMS SEMINAR                     PSA 106  12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Eugene Izhikevich, Department of Mathematics
          "Classification of Bursting Dynamics"
        ABSTRACT: A neuron is said to exhibit bursting dynamics when its
        behavior alternates between spiking (oscillation) and rest
        (quiescence). The history of formal classification of bursters
        starts from the seminal paper by Rinzel (1987), who contrasted
        the bifurcation mechanism of the "square-wave", "parabolic", and
        "elliptic" bursters. We use geometric bifurcation theory to
        extend the existing classification of bursters, including many
        new types.  We show how the type of burster affects its
        neuro-computational properties.
        Additional Information:
        http://math.la.asu.edu/~eugene/publications/nesb.htm
************************************************************************
********************************************************************************
FRIDAY, March 10, 2000
        MATHEMATICAL BIOLOGY SEMINAR                 PSA 106  12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Horst Thieme, Department of Mathematics
          "When can a Cannibalistic Strain Invade a Tiger Salamander
          Population?"
****************************************************************************************************************************************
FRIDAY, March 24, 2000
************************************************************************
        MATHEMATICAL BIOLOGY SEMINAR                 PSA 106  12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Diana Verzi, Claremont Graduate University
          "A Nonlinear Cable Model for Activity-Dependent Spine
          Densities"
        ABSTRACT: The spread of electrical activity in a dendritic
        tree is shaped, in part, by its morphology (its branching
        structure, geometry, spine distribution). Conversely,
        experimental evidence is growing that electrical activity can
        slowly shape the morphology of the dendrite and regulate the
        density of dendritic spines. In this talk a nonlinear cable
        model is formulated to explore how activity-dependent changes
        in spine densities (minutes to hours) can influence patterns
        of electrical activity; and how electrical activity due to
        synaptic events and excitable membrane properties can, over
        time, influence the spine distribution and hence the morphology
        of the dendrite. In the model, the distribution of spines is
        treated as a continuum rather than discretely, and the spine
        density is modeled as a slow dynamic variable. The system of
        differential equations is autonomous. A minimal model is
        proposed where the local change in spine density depends on
        local electrical activity, measured by current flow between
        the spine head and the dendritic shaft. For spines with
        passive spine head membrane, we show how spine densities can
        increase in regions of sustained synaptic activity, and how
        existing spines can wither away when deprived of synaptic
        activity. For excitable spines, we demonstrate how pathways
        for impulse propagation can be forged over time, due to the
        recruitment of distal spines.
************************************************************************

************************************************************************
FRIDAY, April 14, 2000
************************************************************************
        MATHEMATICAL BIOLOGY / STATISTICAL GENOMICS SEMINAR
        NOTE DIFFERENT ROOM AND TIME:               ECA 221   12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Laura Salter, University of New Mexico
          "A Stochastic Search Strategy for Estimation of Maximum
          Likelihood Phylogenetic Trees"
        ABSTRACT: A common goal in the analysis of nucleotide sequence
        data sets is the inference of the phylogenetic history of the
        sequences under consideration.  Although many criteria for the
        selection of a phylogenetic representation of the data are
        available, the maximum likelihood method of phylogenetic tree
        construction has several advantages over other criteria.
        These include the interpretability of the underlying models,
        consistency in the statistical sense, and the possibility of
        statistical testing of hypotheses using the likelihood framework.
        However, use of the maximum likelihood method in practice has
        been limited by two significant difficulties in its
        implementation.  The first is that existing implementations of
        the method can be quite time-consuming when the number of
        sequences under consideration is large.  The second is that
        current algorithms used to implement the method seem to be
        prone to entrapment in local maxima when the data are
        sufficiently complex.  To overcome these difficulties, we
        propose a stochastic search strategy for estimation of the ML
        tree which incorporates ideas from both the simulated annealing
        algorithm and the stochastic probing algorithm.  The algorithm
        works by moving through tree space via a "local rearrangement"
        strategy so that topologies that improve the likelihood are
        always accepted, while those that decrease the likelihood are
        accepted with a probability that is related to the
        proportionate decrease in likelihood.  In addition to greatly
        reducing the time required to estimate the ML tree, the
        stochastic search strategy is much less likely to become
        trapped in local optima than are existing algorithms for ML
        tree estimation.  The success of the algorithm will be
        demonstrated by comparison to two existing algorithms
        (Felsenstein's DNAMLK and Swofford's PAUP*) for several
        theoretical and real data examples. Extensions of the current
        algorithm to include simultaneous estimation of the parameters
        in the underlying nucleotide substitution models will also be
        discussed.
          This is joint work with Dennis Pearl (Ohio State University).
************************************************************************
MONDAY, April 24, 2000
************************************************************************
        MATHEMATICAL BIOLOGY SEMINAR                 PSA 307  12:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Dung Le, University of Texas at San Antonio
          "On a Time Dependent Cross Diffusion System"
        ABSTRACT:  Long time dynamics of solutions to a strongly
        coupled system of parabolic equations modelling the
        competition in bio-reactors with chemotaxis will be studied.
        If the parameters of the system are periodic, sufficient
        conditions for positive periodic solutions will be derived
        in terms of parabolic eigenvalue problems.
************************************************************************
 NONLINEAR DYNAMICS / MATHEMATICAL BIOLOGY SEMINAR   PSA 103   3:40 p.m.
        (Co-sponsor: Systems Science and Engineering Research Center)
        Ira Schwartz, Naval Research Laboratory, Washington
          "Noise Induced Chaos in Population Dynamics"
        ABSTRACT:  In many model models of population dynamics, which
        include epidemics and lasers in the semi-classical limit,
        quadratic  nonlinearities play an important role in the global
        behavior of dynamical transients. That is, when population
        models are modeled deterministically  using estimated parameters
        from measurements, predictions are periodic  in contrast to
        observed chaotic oscillations. In this talk, I will review the
        global topology of some simple examples of population models,
        and show how stochastic parameters can generate chaotic
        behavior. In adition, one can use these ideas to generate
        sustained chaos where there is no chaotic attrcator.
************************************************************************
 

Fall, 1997 talks
Spring, 1998 talks,
Fall,  1998 talks
Spring, 1999
Fall, 1999