Students' Research and Internship Experiences in Summer 2013

Becki Beadling, University of Texas at Austin, Austin, TX

Participation in this program has allowed me to experience a new field of research in which I had no prior exposure; network and form bonds with faculty, graduate students, and peers in the same field as me; and experience what it is truly like to be a graduate student. Along with the research, this program provided me with the opportunity to explore Austin, Texas, and all it has to offer, which made my experience very exciting and enjoyable. This experience has greatly influenced my choice to continue my education at the graduate level.

Abstract:

Many bacterial infections are not caused by a single bacterial type, but rather interactions between multiple bacterial species. Bacterial microcolonies are known to interact with one another synergistically resulting in attributes such as an increased resistance to certain antibiotics and enhanced colonization of the infection site. Spatial arrangement of microcolonies within a bacterial infection site may have a significant impact on these interactions. The ability to conduct studies to assess the effects of bacterial spatial organization has been limited by the availability of techniques for organizing cells in three dimensions. Here, we present preliminary work on a strategy for confining low numbers of bacteria within gelatin-based microstructures for delivery to physiologically relevant environments. This technique will be capable of creating microcolonies of defined density, size and shape.

Ashleigh Craig, Goldman, Sachs & Co, Salt Lake City, UT

Goldman Sachs is one of the leading investment banking firms in its industry. "Operations is at the core of Goldman Sachs. For every trade agreed, every new product launched or market entered, every transaction completed, it is Operations that enables business to flow." The Risk and Control Subdivision of Operations is responsible for ensuring the integrity of books and records as well as maintaining accordance with regulator requirements. As a member of the Global Control-Ledger Integrity team, it was our job to ensure the integrity of books and records. This was accomplished by performing trade transaction reconciliations and reporting breaks (discrepancies) to the Operations teams for them to correct. I addition to performing these daily duties on the team, I also worked on two projects. One project required me to gather all of the procedure documents for my team and upload the corresponding information into a new database. The other project involved analyzing manual reconciliations in an attempt to automate the functions. I documented controls to be put in place in order to minimize the risk involved in performing manual functions. In addition, I gave recommendations for automating the manual functions. This internship was a truly rewarding experience. It gave me insight into what it would be like to work for a leading firm in the investment banking industry.

Charles Culbertson, Texas A&M University, College Station, TX

This summer, I spent 10 weeks at Texas A&M University working with Dr. Kim Dunbar and PhD candidate Xuan Zhang. My project was to synthesize novel magnetic compounds. Magnetic molecular compounds have the potential for applications in data storage and as quantum bits (or qubits) in quantum computing. However, many problems face the use of molecule-based magnets in practical applications, namely spin relaxation. For a molecule-based magnet to be practical it must be able to retain the magnetization at a reasonable operating temperature. A possible solution to prevent spin relaxation is to inhibit spin randomization. My research focused on preparing "single molecule magnets" (SMMs), using -conjugated organocyanide ligands to inhibit spin randomization. We were following several theoretical studies that predicted strong magnetic coupling between trivalent metal ions through organocyanide anions. Lastly, we had to use pentadentate capping ligands to restrict the reactivity of the metal centers, leaving only one open site to be occupied by the bridging ligand. I was successful in creating the mono substituted species, but due to time constraints the research remains incomplete.

Theresa Ann Scarnati, North Carolina State University, Raleigh, NC

In the summer, I was a member of the North Carolina State Summer Research Experience for Undergraduates in Mathematics: Modeling and Industrial Applied Mathematics. During the REU we developed a patient specific model of cerebral blood flow velocity in attempts to better understand cerebral autoregulation (CA). Numerous studies in the past have been dedicated to developing computational models in order to understand the physiological mechanisms associated with CA. However, the majority of previous models only supply us with a qualitative understanding of CA. This summer, we aimed to create a simple, quantitative, and patient specific CA model. Because of the limited data available and the complex physiological processes involved in CA, we predict cerebral blood flow given only systemic arterial blood pressure. The model was made patient specific by using sensitivity and covariance analyses to obtain a subset of parameters that can be estimated given available, quantitative data.

Matthew Sulkosky, SAIC/Leidos Internship

From June 10 until August 23, I interned at the company formerly known as SAIC and that is now called Leidos. During my time at Leidos, I worked with a geographically diverse team on an unclassified Internal Research and Development project called the PL-5. The goal of the project was to develop a piece of technology, called a cross-domain solution, that would permit an unclassified security network to connect and send bidirectional traffic with a higher security level, such as secret. Central to the project was the use of a CS-4000, a highly secure blade server equipped with anti-tampering technology produced by the Cloudshield Company in our cross-domain solution design. There are very few viable PL-5 cross-domain solutions on the market, and by using the CS-4000 it permits speed not available in the market.

The team consisted of two interns at the Columbia (Maryland location), two full-time Leidos employees located at remote locations, and one employee of the Cloudshield Company that worked in Reston (Virginia). Most communication was in the form of e-mails and phone conversations. My role on the team included:

  • Researching message transfer software and test direct transfer of messages
  • Researching SSL encryption using PKI certificates
  • Researching and documentation of a comparison of Linux guest operating systems, including PacketLinux to run on CPAs
  • Researching JMS messages over Tibco EMS framework
  • Evaluating the capabilities of the Linux VMs, and its development tools
  • Developing testing procedures for each stage of the PL-5 project.

During my time at Leidos we were able to work through the initial design phase, and were beginning the steps towards implementation as my internship ended. Overall, a lot of my time was spent researching different technologies and software packages in order to help design the cross-domain system. The internship position required a lot of systems and software engineering experience. A strong knowledge of the OSI model and protocols related to each level of the model was helpful for each stage of the project.

Scott Rega, Lancaster General Health, Lancaster, PA

My summer internship was located at Lancaster General Health, in Lancaster, Pa. My job title was business intelligence graduate intern within the Information Services Department.

While at Lancaster General Health, I performed predictive modeling for the Business Intelligence Team and Information Services Department. This consisted of conducting research and building models from scratch to determine the likelihoods of their corresponding given scenarios in the healthcare field using logistic regression and statistical analysis. This also provided clinical and quality decision support.

My main project focus was on predicting days to procedure, total length of stay, and readmission rates for cardiac patient procedure stents. Other business outcomes result in net profit/loss, total cost, total direct cost, total direct variable cost, and capital of risk.

The processes were carried out through intensive use of the statistical-based software program R, Excel, and database systems such as Midas.

Jon Wayland, Phoenix Benefits Groups, Pittsburgh, PA

During my summer as an actuary intern at Phoenix Benefits Group, Inc. in Pittsburgh, Pa., I did a multitude of different projects. Many of these involved different types of mathematics ranging from statistics and probability to simple arithmetic and algebra. When filling out mandatory reconciliation forms for single-employer benefit pension plans, I needed to calculate the asset reconciliation where sometimes only specific asset amounts were known. Since the total amount of the plan's assets was known, I needed to do simple algebra to find the unknown amounts. Other calculations within these forms involved finding the percentage of how well a balance meets its target amount. I also used statistics and probability when finding the expected amount of total deaths for a given year of a specific plan and comparing them to the actual amount of deaths for that plan. I found these using the software R that I learned here at IUP.

Jeremy Yagle, NASA Independent Verification and Validation Facility, Fairmont, WV

In my cooperative, I worked in the Office of the Director of the NASA Independent Verification and Validation Facility as an engineering trainee. My main responsibility is assisting software engineers and systems engineers in the areas of advanced concepts and business development related to solving specific NASA mission challenges. My daily work involves researching the technical challenges of our customers, and identifying new technologies that the IV&V Program should or could adopt.