ABSTRACTS

 

Abstracts are listed in alphabetic order by the last name of the presenting author.

 

Competitive Exclusion and Coexistence for Pathogens in an Epidemic Model with Variable Population Size

 

Ackleh, Azmy* and Linda Allen

 

DEPARTMENT OF MATHEMATICS AND STATISTICS, TEXAS TECH UNIVERSITY, LUBBOCK, TX  79409-1042

aackleh@math.ttu.edu

 

We study an SIR epidemic model with a variable host population size. We prove that if the model parameters satisfy certain inequalities then competition between n pathogens for a single host leads to exclusion of all pathogens except the one with the largest basic reproduction number. It is shown that the basic reproduction numbers are necessary but not sufficient for determining competitive exclusion. Numerical results corroborating the theoretical ones are presented. An example is given which shows that if such inequalities are not satisfied then coexistence occurs.

 

 

Some Factors Affecting Coexistence of Multiple Pathogen Strains Within a Host Population

 

Allen, Linda J.S.

 

DEPARTMENT OF MATHEMATICS AND STATISTICS, TEXAS TECH UNIVERSITY, LUBBOCK, TEXAS

lallen@math.ttu.edu

 

Introduction of multiple pathogen strains into a host population results in either no disease outbreak, where all pathogen strains are elinminated or a disease outbreak, where either a single strain dominates or multiple pathogen strains coexist.  Many intrinsic and extrinsic factors affecting the host and pathogen determine whether a disease outbreak will occur and whether one or more pathogen strains will coexist.  Some of these factors include the reproduction and mortality rate of the host, the disease-related death rates, and the horizontal and vertical transmission rates for each strain.  Vertical transmission is the direct transfer of a disease from an infective parent to its offspring whereas horizontal transmission is the passage of infection from one individual to another.  The effect these factors have on a disease outbreak are investigated in some basic deterministic and stochastic epidemic models.  Analytical and numerical results are presented and applications to human and wildlife diseases are discussed.

 

 

An Interdisciplinary Course for Biology and Mathematics Majors

 

Andersen,  Janet*, Murray, K.G.

 

DEPARTMENT OF MATHEMATICS, HOPE COLLEGE, HOLLAND, MI 49422-9000

jandersen@hope.edu

 

Mathematical techniques from linear algebra, differential equations, and probability have become an integral part of many areas of biological research.  Yet few undergraduate biology majors take more than the minimum number of required mathematics courses and so have little or no understanding of these techniques. At the same time, mathematics majors are often unaware of the wide array of biological problems to which their expertise can be applied. Furthermore, mathematics majors often lack a fundamental understanding of the differences between biology and physics (where mathematical applications are typically given) and so do not understand the idiosyncrasies involved in biological experimentation. To address both of these problems, we have recently designed and taught a course for a mixed audience of biology and mathematics students.  This course is co-taught by a biologist and mathematician to a mixture of biology and mathematics major, with different prerequisites for each group. The central theme of the course is for both groups of students to develop the ability to speak a common language. Rather than a standard textbook, we are using biology research papers that highlight the interplay of mathematics and biology and how this gives new insight into biological problems. Students work in mixed groups where biology and math majors help one another to understand the concepts, techniques, and limitations of their respective fields. The course includes a laboratory component focusing on demographic studies of several invertebrate and plant species. We review the development of the course, some of the materials we've used, and the results of its first offering during the Spring 2002 semester.  This course has been developed with the support of NSF grant DUE-0089021.

 

 

Valveless Pumping Explained in Terms of Reflected Pressure Waves

 

Babbs*, Charles1 and Jung, Eunok2

 

1DEPARTMENT OF BASIC MEDICAL SCIENCES, PURDUE UNIVERSITY, 1246 LYNN HALL, WEST LAFAYETTE, IN 47907-1246.

babbs@purdue.edu

2OAK RIDGE NATIONAL LABORATORY

 

Problem:  Valveless pumping refers to nonzero mean flow of fluid within a closed loop of elastic tubing without valves, when a compliant section of the loop is rhythmically compressed at particular frequencies.  Valveless pumping is thought to play a role in blood circulation in embryos and in cardiopulmonary resuscitation (CPR).  Heretofore, the physical mechanism causing valveless pumping has remained a mystery.

 

Approach:  We consider closed loops composed of one length of soft, compliant tubing and one length of stiff, non-compliant tubing.  The loop is filled with water and is compressed near one end of the soft section.  To model such pumps we characterize pulse (pressure) waves in the soft section using the classical wave equation with reflection of the pulse waves at the soft/stiff boundaries, and exponential damping of pulse amplitude with distance traveled by the wave.  Pulse wave velocity is specified by the Moens-Korteweg equation.  The resulting instantaneous pressure difference across the fluid in the stiff segment is computed, and Newton’s second law is used to describe the instantaneous and average movement of fluid in the stiff section.

 

Results:  Mean flow equals zero if compression is performed at the midpoint of the soft segment.  However, when changes in pulse wave velocity caused by expansion of the uncompressed regions of the soft segment are properly accounted for, nonzero mean flow develops during asymmetrical compression of the soft segment at particular frequencies.  The direction of flow depends upon the compression frequency, and flow can be reversed simply by changing from one compression frequency to another, as was found in earlier physical and computational models.  Other parameters, such stiffness constants and tubing length also act to determine the direction of flow.

 

Conclusion:  Reflected pulse waves explain the existence and major features of valveless pumping on the basis of classical Newtonian physics

 

 

Structured Treatment Interruptions during HAART: Looking Beyond the Limit

 

Bajaria, Seema H.*1, Glenn Webb2, Denise E. Kirschner1

 

1DEPARTMENT OF MICROBIOLOGY AND IMMUNOLOGY AND DEPARTMENT OF BIOMEDICAL ENGINEERING, THE UNIVERSITY OF MICHIGAN, 6730 MEDICAL SCIENCE BUILDING II, ANN ARBOR, MI 48109-0620

seemieb@umich.edu

2VANDERBILT UNIVERSITY

 

Highly Active Antiretroviral Therapy (HAART) is presently the only available antiviral therapy for HIV-infected patients.  HAART has been used clinically in various administration schemes for several years.  However, due to the development of drug resistance, evolution of viral strains, or poor patient compliance, this combination of drugs fail to effectively contain the virus long-term in the majority of patients.  Our group and others have suggested a change to the usual regimen of continuous HAART through the use of structured treatment interruptions (STIs).  STIs are believed to provide the same benefits as continuous treatment, such as reduced viral loads and return of CD4+ T cells, while allowing the patient a drug holiday.  We explore the use of STIs using a previously published model that accurately represents CD4+ T-cell counts and viral loads both during HIV-1 infection and HAART therapy.  We simulate the effects of different STI regimens including weekly and monthly interruptions together with variations in treatment initiation time.  We show that treatment outcomes depend upon duration of the interruption, CD4+ T-cell counts and plasma HIV-1 RNA levels at initiation of treatment, and the patient's immune response to therapy.  If a patient exhibits a strong response to continuous treatment, an STI strategy could prove to be a major benefit.  This approach may be particularly relevant for underdeveloped countries, where AIDS prevalence and medical costs are high but resources are lacking.

 

 

Spatial heterogeneity in agroecosystems: field experiments and models

 

Banks, John

 

INTERDISCIPLINARY ARTS & SCIENCES, UNIVERSITY OF WASHINGTON, TACOMA, 1900 COMMERCE STREET, TACOMA, WA 98402

banksj@u.washington.edu

 

Recent congressional legislation aimed at reducing the widespread use of harmful chemical pesticides has generated renewed interest in the development of viable alternatives to insect pest control in agroecosystems.  The development of a new suite of selective pesticides designed to target pests while sparing “beneficial” arthropods provides new opportunities to explore the compatibility of chemical and natural controls.  I describe field experiments performed in Washington State designed to assess the potential for incorporating these selective pesticides into an integrated pest management (IPM) regime for controlling herbivorous aphid pests in a broccoli agroecosystem. I discuss insect movement models focused on designing better IPM strategies, as well as an optimization modeling approach aimed at elucidating mechanisms underlying insect distribution patterns recorded in the field.

 

 

Noisy Determinism in Childhood Diseases

 

Bauch, Chris*, David J.D. Earn

 

DEPARTMENT OF MATHEMATICS AND STATISTICS, McMASTER UNIVERSITY, 1280 MAIN STREET WEST, HAMILTON, ONTARIO, CANADA L8S 4K1

bauch@math.mcmaster.ca

 

Understanding complex incidence patterns of childhood diseases during the twentieth century has been a major goal of mathematical modelling of epidemics.  The salient features of the incidence patterns of most pertussis and rubella time series can be explained by stability analysis of the standard SEIR model, however this approach fails for chicken pox and measles.  Conversely, the incidence patterns of measles and chicken pox time series can be explained as asymptotic states of a seasonally forced variant of the SEIR model, however this fails for pertussis and rubella.  In this work we show that the seasonally forced SEIR model can explain the incidence patterns of all four diseases if an appropriate stability analysis of the model is also carried out in addition to the asymptotic analysis.  An implication is that the dynamics of these diseases can be explained in terms of noisy deterministic cycles as opposed to noise-perturbed fixed points or chaotic effects.  Such analysis shows great promise for future studies in epidemiology and ecology.  In particular, it provides a quantitative method to investigate the effects of noise-sustained transient dynamics.

 

 

Sustainability of Surface Water Transport of Goods: An Ecological Economic Investigation of Systems and Policy Implications

 

Baxter, Brian J.

 

UNIVERSITY OF NEW ENGLAND (ARMIDALE, NSW, AUSTRALIA), 250 EVE MILL ROAD, PHILADELPHIA, TN 37846-3407

bjbaxter99@prodigy.net

 

The described research effort is concerned with the central role of dredging operations in the development and expansion of trade in containerized freight at the Port of New York/New Jersey over the past quarter century and discovery of policy implications revealed through exercise of an ecological economic model capturing dynamic spatial-temporal change of this extended port system . Dredging practices are recognized as needed to service the operational requirement of ocean-going vessel access to interior parts of the Ports’ navigation network and the dredged sediment thus produced are treated as a time series or disturbance frequency driving succession dynamics of selected immotile species associated with the benthic community resident at the federally designated sediment disposal site know as the Mud Dump Site. The disturbance event time series is analyzed using spectral methods and succession dynamics modeled according to birth, death, emigration, immigration rates for early and late colonizers. Some interpretive spatial-temporal trade economic data is also provided.

 

 

Modeling the dynamics of natural rotifer populations

 

1*Berezovsky,  Faina, 2Georgy Karev, 3Mark Borodovsky , 3Terry W. Snell

 

1DEPARTMENT OF MATHEMATICS, HOWARD UNIVERSITY, WASHINGTON, DC 20059,

fsberezo@hotmail.com

2NATIONAL INSTITUTE OF HEALTH, BETHESDA MD  20894

3GEORGIA INSTITUTE OF TECHNOLOGY, ATLANTA, GA 30332-0145

 

A model of the dynamics of natural rotifer populations, based on real data for nine species, is described as a discrete nonlinear map depending on three parameters which reflect characteristics of the population (a species g-index) and environment. Model dynamics and their change by variation of these parameters were investigated by methods of bifurcation theory.  It was shown that the phase–parametric portrait consists of 7 domains of qualitatively different behavior of the model. The portrait allow to predict the principal characteristics of the population dynamics of other rotifer species by calculating  a species g-index for given values of the environmental parameter and placing it in the parametrical portrait.

The principal contribution was to show parameter domains where population persistence is possible (in stable equilibrium, periodic or a- periodic oscillations of population size) as well as to analyze  domains of  population getting to extinct. Five of the nine rotifer species investigated had population dynamical behaviors that lead to steady-states in stable environments; for the remaining species the parametric points lie near domain boundaries. In this case, small reductions in environmental quality due to toxicant exposure can dramatically alter the dynamical behavior of a population. A population formerly in steady-state oscillation may enter a domain of a complex behavior and even chaotic dynamics, which is likely to drive the population to extinction.

It seems prudent to explore how much toxicant exposure it takes thrust rotifer populations into a new dynamical regime. If these exposures are lower than those causing classical toxicity, this would have important implications for setting safe thresholds for toxicant exposures in the environment as well as ecological risk assessment.

 

 

Predator-Mediated Permanence

 

Best, Janet

 

DEPARTMENT OF MATHEMATICS, CORNELL UNIVERSITY, MALOTT HALL, ITHACA, NY 14853  USA

janbes@math.cornell.edu

 

A considerable amount of ecological literature has addressed formulations of competitive exclusion principles; that is, when is it true that n species cannot persist on fewer than n resources? A well known mechanism for avoiding such principles is via predator-mediated coexistence: a predator species can facilitate coexistence between prey species that would otherwise exhibit mutual exclusion. In the simplest ODE example, a maximum of two prey species can persist in this manner with the predator, and this bound is shared by a number of models. We analyze types of predators and feeding behaviors, finding sufficient conditions for a single predator species to mediate uniform persistence of more than two prey species. We also consider the number of prey species that may be supported and describe the population dynamics.

 

 

Dynamics of the 'echo' bloom in a plankton system with nitrogen fixation

 

Boushaba, Khalid and Mercedes Pascual

 

DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY, THE UNIVERSITY OF MICHIGAN, NATURAL SCIENCE BUILDING (KRAUS), 830 NORTH UNIVERSITY, ANN ARBOR, MI 48109-1048

boushaba@umich.edu

 

Consideration of nitrogen fixation adds a positive nonlinear feedback to plankton ecosystem models. We investigate the consequences of this feedback for secondary phytoplankton blooms and the response of phytoplankton dynamics to physical forcing.    The dynamics of phytoplankton, Trichodesmium (the nitrogen fixer), and nutrients is modelled with a system of three differential equations.  The model includes two types of nonlinear interactions: the competition of phytoplankton and Trichodesmium for light, and the positive feedback resulting from Trichodesmium recycling.

      A typical simulation of the model in time, with forcing by a varying mixed-layer depth, reveals a clear successional sequence including a secondary phytoplankton bloom known as an `echo' bloom. We explain this sequence of events through the stability analysis of three different steady states of  the model. Our analysis shows the existence of a critical biological parameter, the ratio of normalized growth rates, determining the occurrence of `echo' blooms and the specific sequence of events following a physical perturbation.   The interplay of positive and negative feedbacks appears essential to the timing and the type of events following such a perturbation.

 

 

Undergraduate Applied Mathematics Projects in a Biology Context

 

Burke, Meghan

 

DEPARTMENT OF MATHEMATICS, KENNESAW STATE UNIVERSITY, BOX #0409, 1000 CHASTAIN ROAD, KENNESAW, GA  30144

mburke@kennesaw.edu

 

This presentation will discuss several project ideas to be used for discovery learning or reinforcement of mathematical concepts from precalculus, differential and integral calculus, and differential equations.  The source of these applied projects will be biology, and specifically, epidemiology.  The projects have been used in the mathematics classroom, but could also be made available to biology students interested in brushing up on mathematics skills within a biology context.  The projects would be appropriate for use in  courses aimed specifically at life-science majors, or in courses for all science (including mathematics) majors.   The intuition and comfort many students have in biology that they may not feel with physics or other traditional sources for projects can be used to excite their learning.

 

 

Including Habitat Quality and Density-dependent Migration in Spatially Explicit Metapopulation Models

 

Diego Ruiz Moreno, Graciela Canziani*, Paula Federico

 

GRUPO DE ECOLOGIA MATEMATICA, FACULTAD DE CIENCIAS EXACTAS, UNIVERSIDAD NACIONAL DEL CENTRO DE LA PROVINCIA DE BUENOS AIRES, CAMPUS PARAJE ARROYO SECO, 7000 TANDIL, ARGENTINA

canziani@exa.unicen.edu.ar

Spatially explicit models are becoming very important in Ecology. The spatial distribution of different populations  have a deep effect on their dynamics as well as on the landscape.

Patch occupancy models include two different spatial scales: that of the local neighborhood and that of the global landscape. Among these models are found Cellular Autamata, which permit to visualize the evolution of the spatial dynamics of populations and analyze the strategies of the species in regard to environmental changes. Cellular Aurtomata permit to include spatial heterogeneity is a simple manner.

A standard tool to link spatial heterogeneity to population dynamics is the definition of Habitat Quality Indices, usually respecting the philosophy of Habitat Suitability Evaluation Procedures (HEP). Here we introduce a methodology for the utilization of  remote sensing information for the production of dynamic maps for landscape classification. The corresponding habitat quality indices are then defined and used to feed the discrete mathematical population models that run within the framework of Cellular Automata. These local population models permit to calculate population densities depending on the quality of the patch and to define migration processes.

Models that have these characteristics permit, as a first approach, to monitor the impact of environmental changes on the dynamics of the population under study as well as on the landscape. Moreover, these models are appropriate for the analysis of different strategies that could help improve the habitat conditions of species that suffer the stress of negative  environmental changes.

 

Spatial modeling to evaluate the effects of habitat loss on mammal populations in tallgrass prairie

 

Carr*, Eric1,2, Henriette Jager2, Tanya Kostova3, Rebecca Efroymson2, Tina Carlsen3, Tom Ashwood2, and Jim Kercher3

 

1UNIVERSITY OF TENNESSEE, KNOXVILLE, TN

carrea@ornl.gov

2OAK RIDGE NATIONAL LABORATORY (ORNL), OAK RIDGE, TN

3LAWRENCE LIVERMORE NATIONAL LABORATORY (LLNL), LIVERMORE, CA

 

We developed spatial models to quantify the effects of petroleum-related habitat loss on mammal populations in the Tallgrass Prairie Preserve in Oklahoma.  We focus mainly on a demographic model, developed at ORNL, for the American badger (Taxodea taxus).  The badger is a voracious predator on small burrowing mammals.  Males and females have large, overlapping home ranges, and mate during summer when the probability of encountering a mate is high compared with the winter, when the animals become dormant.  Implantation and fetal development are delayed until early spring in anticipation of more-abundant resources.  Spatial variation in the badger model is mediated by a habitat suitability index that reflects the availability of prey in different vegetation types.  Native grasses, the dominant category of vegetation in the Preserve, are preferred over other categories.  Habitat suitability influences the acquisition of a home ranges and survival probabilities of simulated individuals.  A second model, developed at LLNL, simulates the trophic dynamics of the prairie vole (Microtus ochrogaster).  The trophic model includes seasonal dynamics of vegetation and its depletion by the herbivorous vole.  The model reflects the territorial behavior of the vole, which is predominantly monogamous.  Prairie voles breed throughout the year, but breeding activity and life expectancy have seasonal dependence. The vole model incorporates bioenergetics and movement that depends on population density, availability of resources, and mating opportunity.  For each model, we simulated population size on landscapes with increasing areas of habitat and differing distributions of habitat subjected to brine spills.  In addition to the amount of habitat removed by spills, we compared population-level effects of one large spill with the effects of many small spills summing to the same area.  Simulated population responses to habitat loss were related to the trophic position and spatial life histories of the two species.

 

 

Spatio-Temporal and long term Mathematical Models on Methane Emission from rice fields

 

Chakraborty, Amit*, Dilip Kumar Bhattacharaya**, Nitai Kundu*

 

*INSTITUTE OF WETLAND MANAGEMENT AND ECOLOGICAL DESIGN, B-04, LA-BLOCK, SECTOR-III, SALTLAKE, CALCUTTA-98, WEST BENGAL, INDIA

chakra_a@hotmail.com

**DEPARTMENT OF PURE MATHEMATICS, CALCUTTA UNIVERSITY, 35, BALLYGANGE CIRCULAR ROAD, WEST BENGAL, INDIA

 

Mathematically the phenomenon of methane emission is a continuous process varying with time. Naturally in this connection, attempt has been made towards developing a continuos dynamic model of methane emission from rice fields expressed by a system of ordinary differential equations with time as the independent variable. As much as possible, all factors affecting the growth of methane has been taken into considerations. These factors are taken as depending variables and their growths are expressed by ordinary differential equations. These differential equations taken together form the required dynamic model. Stability analysis of this dynamical system ascertains the long-term behavior of the model. In this connection, there arises an optimal control problem. Secondly, the growth of methane depending on various factors considered above, also changes significantly with change of places. In other words the growth of methane is also spatio-temporal in nature. The change of different factors responsible for the growth of methane with respect to space co-ordinates separately resulting again as three systems of ordinary first order differential equations, each system has been solved separately. Each such solution has given an estimate of how growth of methane changes as a particular co-ordinate of the points of the space (other two co-ordinates remaining the same) changes. A net change with respect to space co-ordinate has determined by total changes in directions taking two of them fixed. Lastly the spatio-temporal and long term results have been combined with those obtained in separate cases.

 

 

Drug-induced amplification of resistant tuberculosis strains

 

Chou, Tom

 

DEPARTMENT OF BIOMATHEMATICS, UCLA

We develop a model that describes the epidemic dynamics of an arbitrary  number of drug resistant strains of Mycobacterium tuberculosis. The model tracks the dynamics of susceptible individuals, individuals latently infected with M. tuberculosis and diseased individuals. The stability criteria R0 corresponding to the growth of each of the drug-resistant strains is found to be analogous to the stability criterion for the growth of a single strain of M. tuberculosis. Numerical simulations of the differential equations using reasonable parameter values reveals a series of sequential epidemics caused by the drug-resistant strains. These epidemics are predicted to continue to increase for hundreds of years (despite continued treatment of tuberculosis) if the sequential treatment of cases of tuberculosis (due to inadequate treatment regimens) leads to the selection of cases of acquired resistance that have progressively lower cure rates..

Host-Parasitoid Systems:  Modeling the Forest Tent Caterpillar

Cobbold, Christina*, Lele Subhash, Lewis Mark, Lutscher Frithjof, Roland Jens

 

DEPARTMENT OF MATHEMATICS AND STATISTICAL SCIENCES, UNIVERSITY OF ALBERTA, EDMONTON, T6G 2G1, ALBERTA, CANADA

ccobbold@math.ualberta.ca

 

Host-Parasitoid systems are often ideal candidates for integro-difference models, with dispersal and dytnamics occurring as disinct events with non-overlapping generations.  The Forest Tent Caterpillar has a life cycle which follows this behavior.  The principle parasites of the caterpillar are fly species which also adopt the required life cycle for and integro-difference model.

 

The caterpillar population is regarded as a pest causing major forest defoliation during outbreak years.  Although not killing the trees they do slow wood growth.  By using an integro-difference model of the host-parasitoid system we hope to examine how habitat fragmentation can effect host-parasitoid populations and duration of the outbreak period.  From a management perspective the goal is not necessarily eradicate the caterpillar population but minimize the number of years at which we see outbreak densities.

 

 

The Roles of Serial Engagement and Kinetic Proofreading in Peptide-Induced

T-Cell Activation

 

Coombs, Daniel*; Wofsy, Carla; Goldstein, Byron

 

T-10, THEORETICAL BIOLOGY AND BIOPHYSICS, LOS ALAMOS NATIONAL LABORATORY, LOS ALAMOS, NM 87545

coombs@lanl.gov

 

The activation of a T-cell requires the formation of a long-lived attachment to an antigen-presenting cell (APC).  APCs present peptide on their surfaces, held by a major-histocompatibility complex (MHC). A given T-cell carries receptors (TCRs) specific for a particular MHC-peptide group, along with other less specific adhesion and costimulatory molecules. The stable region of close apposition (immunological synapse) may facilitate signal transduction, by concentrating the TCR and MHC-peptide together and allowing long-lived bond formation.  A TCR will become activated if it can form a sufficiently long-lived bond to allow multiple biochemical changes to occur (the kinetic proofreading hypothesis).  We have developed a mathematical model to examine the TCR-MHC-peptide interaction within the stable region and use it to study (1)  the competing effects of serial engagement (sequential activation of TCR by one MHC-peptide) and kinetic proofreading, and (2) the possible role of TCR oligomerization in activation of T-cells. In conjunction with the model, recent experimental data indicates that activated TCR must remain active for a period after dissociation from MHC-peptide. Recent extensions of the model to deal with other situations will also be presented.

 

 

Overgrowth competition, fragmentation, and a sub-individual-based model of sex-ratio dynamics

 

*Crowley, Philip H., Christopher Stieha, and D. Nicholas McLetchie

 

DEPARTMENT OF BIOLOGY AND CENTER FOR ECOLOGY, EVOLUTION & BEHAVIOR, UNIVERSITY OF KENTUCKY, LEXINGTON, KY 40506

pcrowley@uky.edu

 

Some sessile organisms are capable of growing over their neighbors or parts of themselves and thereby blocking access to essential resources—resulting in mortality, abscission of covered distal parts, or fragmentation into functionally separate individuals.  Bryophytes, such as the liverwort Marchantia, fragment into separate ramets by sloughing overgrown tissue within mats on substrate patches.  Simulating this process in MATLAB yields insights into the competitive determinants of sex ratio in these organisms inhabiting tropical rainforests.  It emerges that between-sex overgrowth competition eventually leads to elimination of one sex or the other (often males) from the patch in a fashion resembling unstable Volterra (logistic) dynamics, as anticipated in simpler models of this system.  But the role of within-patch reproduction and disturbance in determining the outcome seems to be reduced in this more realistic representation.

 

 

Optimal Control of Therapy for HIV Based on Maximising Immune Response

 

Culshaw, Rebecca Veronica and Ruan, Shigui (research advisor)

 

DEPARTMENT OF MATHEMATICS AND STATISTICS, DALHOUSIE UNIVERSITY, HALIFAX, NOVA SCOTIA,  CANADA

Rebecca@mscs.dal.ca

 

The role of the specific immune response to HIV infection has received much attention in recent years. In this paper, we attempt to determine the qualitative nature of optimal chemotherapies for HIV when immune response is specifically considered. Our model is a three-dimensional system of ordinary differential equations based upon a model presented by Wodarz and Nowak (). We approach the problem from an optimal control perspective. We seek to maximise levels of healthy CD4+ cells and immune response cells while minimising drug cost. Existence of an optimal control is proven, and it is characterised in terms of the state and the adjoint variables. Numerical simulations of the optimality system reveal the optimal treatments are linearly decreasing from their maximum value, and that at the final time, we retain a portion of the initial pool of immune response cells, while reducing infection to very low levels.

 

 

Non-equilibrium Competition

 

Cushing, J. M.

 

DEPARTMENT OF MATHEMATICS, UNIVERSITY OF ARIZONA, 617 N. SANTA RITA, TUCSON, AZ 85721

cushing@math.arizona.edu

 

The fundamentals of competition theory in ecology (ecological niche, competitive exclusion, limiting similarity, etc.) are founded on the well-known Lotka/Volterra two species competition model. The Lotka/Volterra model permits only a limited number of dynamic scenarios, all of which involve only equilibrium asymptotics. While it is true that many other models also support such a dynamically restricted, equilibrium theory of competition, there are exceptions -- exceptions that often contradict some of the basic tenets of classical theory. In this talk I will discuss such a competition model. The model is based on a single species model (the LPA model) that has an outstanding track record of successful application to real population data. The LPA model has been extensively used during the last decade to study the dynamics of an insect species (flour beetles) that, as it turns out, played a significant role in the early history and development of competition theory. Indeed, several famous laboratory experiments involving flour beetles carried out by Thomas Park during the 1940-1960's helped to establish the fundamental principles of competitive exclusion and ecological niche. Park's experiments were designed explicitly to test Lotka/Volterra theory and, to this day, are still cited in most ecology textbooks. The experimental results, however, contain anomalies that are not mentioned in these citations (although Park and his collaborator P. H. Leslie wrote about them in several papers). These anomalies seemingly do not support Lotka/Volterra dynamics or the basic principle of competitive exclusion (namely, that two species cannot share a single limiting resource). I will show how the LPA model can provide a possible explanation of these anomalies. The explanation involves, however, non-equilibrium dynamics and non-Lotka/Volterra competitive scenarios that "contradict", or are not encompassed by, the classical theory.

 

 

Aggregation and centering in fish melanophore cells - a quantitative exploration of cytoskeletal dynamics

 

Cytrynbaum, Eric1, Alex Mogilner1, Vladimir Rodinov2

 

1INSTITUTE OF THEORETICAL DYNAMICS, UNIVERSITY OF CALIFORNIA, DAVIS, 2201 ACADEMIC SURGE, ONE SHIELDS AVE., DAVIS, CALIFORNIA 95616

2UNIVERSITY OF CONNECTICUT

cytryn@itd.ucdavis.edu

 

Fish melanophore cells demonstrate a self-organizing behaviour that depends on the same cytoskeletal components that are at work in the centering of chromosomes during mitosis and therefore provide a good "warmup" problem for that more complicated and vital process.  When a fragment of the melanophore cell is excised, eliminating the centrosome (the regular cytoskeletal organizer) and therefore the cytoskeletal structure, stimulating the cell with adrenaline somehow reintroduces cytoskeletal organization, leading to the formation of a microtubule aster and the aggregation of the cell's pigment particles at the center of the fragment.  It is this centering behaviour that is analogous to the mitotic process of chromosome alignment.

 

We derive a system of seven non-linear PDEs (1D) that describes the biological system.  Numerical simulations of the equations demonstrate certain observed features (aggregation) but not others (centering).  The system can be reduced so as to facilitate analysis which allow for an understanding of the successes and failures of the original model.  Finally, we generalize the reduced model to 2D, incorporating a stochatic element, and present numerical results.

 

 

Modeling Biological Control of Scentless Chamomile Using a Optimization on a

Coupled Map Lattice

 

de-Camino-Beck, Tomas*1 , Lewis, Mark1, McClay, Alec2

 

1CANADIAN CENTER FOR MATHEMATICAL BIOLOGY, DEPARTMENT OF BIOLOGICAL SCIENCES, UNIVERSITY OF ALBERTA, EDMONTON AB, T6G 2E9, CANADA

tomasd@ualberta.ca

2ALBERTA RESEARCH COUNCIL

 

Scentless Chamomile (Matricaria perforata Mérat) is an invasive annual or short-lived perennial weed that is becoming a pest problem in Canada. It is a patchily distributed plant which can rapidly increase to outbreak densities on disturbed sites. Several potential biological control agents have been tested empirically, but little is known about effect that these agents have on the persistence and stability of scentless chamomile populations, and what biological control optimal release strategy would work best to reduce scentless chamomile populations.

Our goal is to determine optimal strategies for biocontrol of scentless chamomile. Our first step was to develop a multi-stage coupled map lattice model with long distance dispersal, to study the plant-herbivore interactions in a spatial context. This model will allow us to study the interaction between scentless chamomile and the biological control agents (the gall midge Rhopalomyia tripleurospermi and the seed weevil Omphalapion hookeri), and provide a solution for the best release strategies for biological control. To consider spatial heterogeneity we link our model to a GIS landscape information layer. Future research includes linking the model to a genetic algorithm to find optimal release strategies.

 

MHC polymorphism by heterozygote advantage

 

De Boer, Rob J.*1, Jose A. M. Borghans2, Michiel van Boven3, Can Kesmir1 & Franz J. Weissing4

 

1THEORETICAL BIOLOGY UU, PADUALAAN 8, 3584 CH  UTRECHT, THE NETHERLANDS

2INSTITUT PASTEUR, PARIS

3INSTITUTE FOR ANIMAL SCIENCE AND HEALTH, LELYSTAD

4 THEORETICAL BIOLOGY, UNIVERSITY OF GRONINGEN

R.J.DeBoer@bio.uu.nl

 

Molecules from the major histocompatibility complex (MHC) play a crucial role in vertebrate immunity and are encoded by the most polymorphic genes known for vertebrates. Since hosts heterozygous at the MHC express a higher diversity of MHC molecules to present peptides from infectious pathogens, there should be selection for heterozygous hosts.  We have developed a mathematical model to show that under realistic assumptions heterozygote advantage yields a small degree of MHC polymorphism.  A large degree of polymorphism can only be accounted for if the different MHC alleles confer unrealistically similar fitnesses.  Predicting the immunodominant peptides from various common viruses we find that different MHC alleles are expected to provide quite different levels of protection.  Heterozygote advantage is therefore insufficient to explain the high observed polymorphism of the MHC.

 

 

Determining the mobility of actin proteins in the cell nucleus with a compartmental model

 

Gustavo Carrero, Ellen Crawford, Michael Hendzel, and Gerda de Vries*

 

DEPARTMENT OF MATHEMATICAL AND STATISTICAL SCIENCES, UNIVERSITY OF ALBERTA, EDMONTON, AB, T6G 2G1, CANADA

devries@math.ualberta.ca

 

Fluorescence Recovery After Photobleaching (FRAP) is an imaging technique used to study the mobility of proteins in the cell nucleus. In FRAP experiments, the protein being studied is tagged with Green Fluorescent Protein (GFP). An intense laser beam is used to bleach the fluorophore of the tagged proteins within a small region of the cell nucleus. Due to diffusional exchange between the bleached and unbleached proteins, fluorescence in the targeted area recovers. The fluorescence recovery data can be used to quantify the mobility of the proteins. We propose the use of compartmental models to interpret FRAP data and estimate dynamical parameters. In particular, we develop a compartmental model that describes the dynamics of nuclear GFP-actin.  The fluorescence recovery data for this protein cannot be explained by simple diffusion alone. We can obtain a good fit of the data if we include in our model terms that describe association and dissociation of actin monomers with filaments. The model supports the hypothesis that actin is present in both monomeric and filamentous forms. We predict the average residency time of GFP-actin within a filament, the average wandering time of monomers between bindings, and estimate the proportions of GFP-actin in monomeric and filamentous forms.

 

 

A model of individual exposure to contaminants using multidimensional structure arrays

 

Dixon, Kenneth*1, and Lorinda L. Sheeler2

 

1THE INSTITUTE OF ENVIRONMENTAL AND HUMAN HEALTH, TEXAS TECH UNIVERSITY, BOX 41163, LUBBOCK, TX 79409-1163

ken.dixon@tiehh.ttu.edu

2TENNESSEE DEPARTMENT OF HEALTH

 

Individual animals are exposed to contaminants at different environmental concentrations depending upon their location in time and space.  We developed an individual based, spatial, stochastic model that simulates exposure to spatially varying environmental concentrations of contaminants.  As the animal moves around the landscape, it receives a contaminant dose through ingestion of food items with varying concentrations.  The body burden of each individual then is tracked over time.  Many sub-lethal toxic effects of exposure to contaminants affect the health status of the individual.  To model these effects, we need to follow the status of the animal over time.  Population effects are estimated by simulating many individuals.  Modeling health status, and the probability of a sub-lethal effect, require a modeling approach that maintains a history of the health status and not just a distribution of the status in the population.  Our modeling approach is to use multidimensional structure arrays to track health status, geographic location, and other variables for each individual in the population.  Mathematical functions, such as arithmetic means and variances, can be calculated for each variable operating on fields in the arrays using indexing.  A current application of the model is the simulation of the effects of perchlorate on a raccoon population at a superfund site.

 

 

Optimal spatial control of antibiotic resistance in Tuberculosis

 

Duke-Sylvester, Scott M., Gross, Louis J., Salinas, Rene A., Purucker, Thomas, Joshi, Hem R.,Harrell, Susan

 

DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY, 569 DABNEY HALL, KNOXVILLE, TN 37919-1610

sylv@tiem.utk.edu

 

The development of resistance by infectious micro-organisms to antibiotics is a matter of global concern. Resistant pathogens present a significant health risk to humans in both developed and developing nations. Among the pathogens of concern is Mycobacterium tuberculosis (TB). A number of different techniques have been proposed to limit the buildup of resistance including drug cycling. In drug cycling patients are grouped based on when they became infected and all are treated with a single anti-microbial agent. In this way a single drug is used for all new infections that occur over some period of time. Individuals who become infected before or after the given interval fall into other groups which receive different single drug treatments. Several studies have reported the successful use of this approach to limit the development of resistance. We have developed a model of resistance dynamics for TB that incorporates drug cycling in a number of discrete populations connected through migration. This model is used in an optimal control framework that minimizes the total resistance through time and across all populations. This approach allows us to address two important questions. First, it is known from empirical work that drug cycling can limit the build up of resistance. Given this observation, what sequence of drug application provides the smallest buildup of resistance. Second, how does spatial structure effect the treatment plan, and is a spatially-structured treatment plan substantially better than a plan in which all regions receive the same sequence of treatment.

 

 

Fitting population dynamic models to doubly-noisy data: methods and insects

 

Ellner, Stephen

 

DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY, CORNELL UNIVERSITY, E145 CORSON HALL, ITHACA, NY 14853-2701

Spe2@cornell.edu

 

Population data, especially from outside the laboratory, are typically noisy in two ways that complicate model fitting: process noise and measurement noise. Until recently, methods for fitting mechanistic models to population time series data could only cope with these one at a time: deterministic models with inaccurate data, or stochastic models with accurate data. I will describe two methods that can handle both at once: gradient matching, and indirect inference using nonlinear forecasting. Gradient matching is more efficient but makes stringent assumptions about the data; I will illustrate its use on some laboratory insect populations, including some non-Nicholson blowflies. Indirect inference is very general but computationally demanding. I will describe an analysis of long-term data on outbreak cycles of a forest insect pest (Bupalus piniarius), in which comparative model-fitting is used to evaluate 3 longstanding hypotheses for the mechanism underlying the cycles: interactions with food supply, parasitoids, and delayed self-regulation through maternal effects.

 

 

Mathematical Study of Some Immune System Response Models

 

Elsady, Zeinab

 

25 HAMOUDA ST. HELWAN GARDENS, CAIRO, EGYPT

zelsady@hotmail.com

 

Nonlinear systems associated with: Normal T cells model and HIV and cancer immune interaction models adopted with biological theory will be given.  Despite of there are several studies associated with the solution of these models but there are no mathematical details about their solution.  In addition, examination of long-term behavior of the dynamical system by using small time step requires extra computing costs. So the need to use numerical methods, allows largest possible time step and in the same time consistent with stability and accuracy criteria.

 

 

Stability and Chaotic Dynamics in Deterministic and Stochastic Structured Epidemic Models

 

Emmert, K. * and Allen, L. J. S

 

DEPARTMENT OF MATHEMATICS AND STATISTICS, TEXAS TECH UNIVERSITY, LUBBOCK, TEXAS

kemmert@math.ttu.edu

 

A discrete-time structured model for the spread of infection in a host population is developed and analyzed. The host population is subdivided into larvae, juvenile and adult stages. In the absence of infection, conditions are derived for existence and stability of a positive equilibrium. In the case of a Ricker type density-dependence in the larval stage, the model is shown to exhibit chaotic dynamics.  Conditions for persistence of the infection in the host population are studied.  In addition, an analogous stochastic model, a discrete-time Markov chain model, is formulated. The dynamics of the deterministic and stochastic models are compared and illustrated with several numerical simulations.

 

 

LIFE AS A QUANTUM STATE: SOME IMPLICATIONS FOR QUANTITATIVE BIOLOGY

 

Finkel, Robert

 

PHYSICS DEPARTMENT, ST. JOHN’S UNIVERSITY, 8000 UTOPIA PKWY., JAMAICA,  NY   11439

finkelr@stjohns.edu

 

Quantum physics determines the properties of atoms and molecules—properties not evidenced by separated electrons and nuclei. The whole exceeds the isolated parts because mutual interactions are included. Similarly, the integrity and synchronization of living cells is only partly explained by detailing biochemical pathways and nucleic acid sequences. The interacting processes of such elements constitute life. A simple physical model of the living state can be based on a cooperative quantum effect.1 Here autocatalyic chemicals in a solution of substrate interact to form a coherent aggregate with a negative energy spectrum. Two parameters of the theory, presumed to be universal, are determined from E. Coli data. We show that the aggregate is stable against thermal assaults and can grow by accretion in sufficient substrate. The model demonstrates quantitatively how cells can convey solitary short-range molecules to appropriate locations against any classical statistical expectations. We derive the venerable Kleiber law relating metabolic rates to mass as an immediate consequence of the theory.

[1] R.W. Finkel, Nuovo Cimento, 221B, 41 (1978).

 

 

Optimal Control of an Immunotherapy System

 

Fister, K. Renee and Jon Ernstberger* (student)

 

6C FACULTY HALL, DEPARTMENT OF MATHEMATICS AND STATISTICS, MURRAY STATE UNIVERSITY, MURRAY, KY  42071

Renee.Fister@murraystate.edu

 

We investigate a mathematical model that studies the dynamics between tumor cells, immune-effector cells, and cytokine interleukin -2. In order to determine under what circumstances the tumor can be eliminated, we implement optimal control theory. Via an objective functional, we maximize the effector cells and the interleukin-2 cells while minimizing the tumor burden and the cost of the control. We show that an optimal control for this problem exists and is unique through the use of an optimality system. Then we analyze the system graphically using Matlab.

 

Computational Challenges in Modeling Blood Clotting

 

Fogelson, Aaron

 

DEPARTMENT OF MATHEMATICS, UNIVERSITY OF UTAH, SALT LAKE CITY, UT 84112

fogelson@math.utah.edu

 

Fluid dynamics, biochemistry, and biophysics, in particular cellular adhesion processes, interact in extremely complex ways to determine whether, where, to what size, and how stably blood clots form within the vascular system.  Modeling these processes poses substantial computational challenges.  We will describe efforts to build

computational models of clotting at the cellular level that include immersed-boundary-based fluid-cell interactions and both fluid-phase and cell-surface phase-biochemistry.  We will also describe novel computational methods for studying continuum models of clotting that have important features in common with classical viscoelastic polymer

flow models.

 

 

Back to Basics:  <