This project studies a network model for the transportation of consumable resources.  As in most models, a supply of the resource (in this case, "pebbles") is placed on various vertices of the graph, and a demand of pebbles is identified at certain other vertices.  In this model, the movement of a pebble across an edge requires the loss or payment of a partner pebble; that is, while two pebbles leave a vertex together, only one arrives at the other end of the edge, as the other pebble vanishes.  The central question asks whether a given supply can satisfy a fixed demand, but many other significant questions exist.  We study this paradigm on an array of special graph classes, seeking to discover when one can be successful in achieving assorted goals.  This field is new enough to have numerous interesting problems within reach of talented undergraduate students, and there have been many strong contributions from students over the years. Students with some background in graph theory, discrete math, combinatorics, and/or computer programming are particularly well suited, but the essential concepts can be learned as needed.

Contact Professor Hurlbert 

A nematocyst is a specialized organelle within cells of jellyfish and other Cnidarians that sting. They contain a barbed, venomous thread that accelerates faster than almost anything else in the animal kingdom. This is counter-intuitive because the organisms themselves generally move very slowly. Using computational flow simulations, we want to investigate the nature of the fluid flow associated with variations in the geometry of the organisms and the configuration of the nematocyst(s). There is a large variety of sizes and shapes of nematocysts and this may have important fluid dynamic consequences, particularly important in the ability of the organism to actually capture prey.  Students with a background in MATLAB or other coding platforms will be most successful. Some knowledge of odes and/or pdes is also useful but the essential concepts can be learned as needed.

Contact Professor Segal

Chronic disease resulting from a diet high in fat and cholesterol affects over 50% of adults. Although the relationship between atherosclerosis and inflammation is not fully understood, it is known that macrophages and foam cells promote inflammation by secreting pro-inflammatory cytokines following accumulation of cholesterol.  We are developing a computational model that accounts for the conversion of macrophages to foam cells and the formation of arterial plaques in atherosclerosis. Understanding the progression of size/shape/density and markers of disease progression are all important to examine with regard to the diagnosis and long term prognosis of individuals with atherosclerosis.  Students with some background in MATLAB or other coding platforms will be most successful. Some knowledge of odes and/or pdes is also useful but the essential concepts can be learned as needed.

Contact Professor Segal

Many phenomena in the physical and life sciences share surprisingly similar complex structures. Pattern formations in these complex systems, including swarming of animal populations, motion of human crowds, and molecular distribution, are driven driven by competing between long-range attractive and short-range repulsive forces. Various agent-based models (defined via systems of simple ODEs) can be used to describe these phenomena. Current research in the literature is, particularly, interested in understanding fine properties of these pattern formations as the number of agents becomes large. The goal of this project is to develop an efficient computational code (either using MATLAB or another coding platform) which can handle large number of particles as well as domain and density constraints. Beyond good computational and coding skills some basic knowledge of ODEs and linear algebra is expected.

Contact Professor Topaloglu

In quantum computers the fundamental bits of information are states of a quantum system known as qubits. In analogy with classical algorithms, quantum algorithms can be thought of as recipes to transform information stored in qubits. Some algorithms are faster than others and classifying problem based on how fast they can be solved by a quantum computer is of great importance for both theoretical and practical applications, including cryptography and on-line security. This project aims at developing efficient algorithms to efficiently solve specific instances of the quantum satisfiability problem (QSAT), which is expected to be in general intractable even on a quantum computer. Basic knowledge of linear algebra is recommended, some experience with coding (in any language) is desirable but not required.

Contact Professor Aldi

Lie algebras provide an algebraic framework for encoding infinitesimal symmetries. Besides their many applications in geometry and in the theory of differential equations, Lie algebras are of fundamental importance in high energy physics as they are required in order to formulate the Standard Model of particle physics. The goal of this project is to study a class of geometric symmetries of Lie algebras known as generalized CRF structures, which have applications to the mathematical study of string theory.  No prior knowledge of physics or Lie algebras is required. Familiarity with linear algebra is recommended.

Contact Professor Aldi

This project studies a model of breathing in extremely premature (EP) infants as it relates to the naturally compliant (floppy) chest wall typical of this population. EP infants that present with respiratory distress are currently treated with surfactant replacement therapy which aids in lung development but mechanical ventilation is almost always required as well. However, invasive ventilation is a source of trauma leading to irregular lung development in 1/3 of infants, and noninvasive ventilation has been observed to ineffective in this population. It is hypothesized that this is due to the highly compliant (floppy) chest wall leading to insufficient inhalation and progressive lung collapse. Our short term goal is to develop a full model with which to investigate this hypothesis and potentially uncover biomarkers that can distinguish infants at risk for ventilation failure, with a long term goal of developing a supportive treatment to stiffen the chest wall and facilitate ventilation. We are currently working with a base model parameterized for an idealized preterm infant, and several potential directions exist for further model development and analysis. Experience coding in MATLAB or related languages, familiarity with differential equations, and interest in human physiology are desired.

Contact Professor Fix