Seniors
Emily Bartlett
Chemistry
Mentor: Jón T. Njarðarson
Category of Research: Physical Sciences
Total Synthesis of Platensimycin
Platensimycin is an important natural product isolated by a Merck research group in 2006 from a soil sample from Africa. It is of great interest as a synthetic target because it shows potent antibiotic activity against gram-positive bacteria with multiple antibiotic resistance. Especially important is the fact that it operates by blocking bacterial fatty acid biosynthesis, a mode of action distinct from any previously known antibiotic. This project centers around efforts towards the total synthesis of platensimycin in order to provide an efficient way to access the molecule and to provide a means to systematically alter its structure if necessary to improve biological and pharmacological properties. Our retrosynthetic analysis of this molecule divided it into two parts: an aromatic side chain joined by an amide linkage to a pentacyclic core, both of which have been successfully synthesized to complete a formal synthesis of the molecule. In addition, our synthesis of this molecule uses methodology developed in our group involving copper-mediated rearrangement of vinyl epoxides, thus showing the utility and generality of this reaction type.
SueYeon Chung
Physics and Mathematics
Mentor: Professor Julia Thom
Category of Research:
Activating Backside Junctions in Thin Silicon Pixel Detectors through Laser Anneal
This project is part of a larger effort to develop thinned silicon pixel detectors for particle detection in High Energy Physics experiments. These detector arrays, containing active CMOS pixel circuitry on the top of the wafer, will require formation of an extremely low leakage backside junction at the end of processing. This requires a high dose implant followed by activation anneal. Due to the already existing circuitry, static temperatures must remain below 450oC during this anneal. Pulsed laser annealing is currently the most promising method for achieving suitable activation of this backside implant. We have focused specifically on understanding and optimizing a process for activating these backside junctions. We have fabricated a range of planar diodes with guard rings on conventional silicon wafers and implanted and laser annealed them with varying parameters. We have studied the diode reverse leakage properties as a function of diode area as well as implant and laser anneal parameters.
Stephanie Dusaban
Biological Sciences
Mentors: Richard Cerione and Marc Antonyak
Category of Research: Life Sciences
Tissue Transglutaminase Secreted by MDAMB231 Breast Cancer Cells Promotes Cell Survival
Tissue transglutaminase (TGase) has been implicated in the progression of certain types of human cancer, including breast cancer. TGase is a dual function protein that couples an enzymatic transamidation activity with the ability to bind and hydrolyze GTP. The transamidation activity of TGase has been well studied and involves a calcium dependent reaction, which results in the formation of a covalent bond between two proteins or a protein and a polyamine. Previous research in our laboratory has shown that TGase plays a role in conferring metastatic and survival properties to breast cancer cells. My past research has been aimed at determining how TGase can promote chemoresistance. To gain some insight into this question, I began by characterizing the localization of TGase in these cells. Immunofluorescence studies and cell fractionation studies have shown that TGase is primarily in the cytoplasm of cells; however, it can also be externalized from cells where it can interact with additional proteins like integrin and fibronectin. I have been able to show that TGase is getting excreted from the cell in a non-classical manner. Moreover, I have determined that TGase lacking either its GTP-binding activity, cross-linking ability, or both activities can still be externalized. Though we do not know exactly how TGase is being secreted from cells, we have determined that is transamidationally active and confers a survival advantage. Currently, we have determined that a possible mechanism by which TGase is being secreted is through microvesicles and may provide a novel way that TGase can promote chemoresistance.
Asad Haider
College Scholar Program
Mentor: Jonathan Culler
Category of Research: Other
The Politics of Popular Culture
Popular culture is a phenomenon that is often reduced to mere entertainment, but this overlooks the important fact that the new forms of culture that saturate our reality are a fundamental aesthetic, technological, economic and political reality. Culture has always had an important social role, from the Ancient Greeks’ use of theater as a means of political contestation to the way presidential candidates today represent themselves on television. This research project explores three notions: first, popular culture is a serious form of cultural production that must be studied with all the careful attention that is devoted to high culture like novels and paintings; second, popular culture represents political positions in complex and often contradictory ways, as it must communicate the message of those who own the technologies of information while simultaneously appealing to an audience which may have different desires; third, culture and information have become the primary economic sector in the developed world, especially in American society, where our everyday lives are often part of the production of forms of knowledge that make our symbol-driven economy possible. The work has been guided methodologically by two traditions in critical theory: cultural studies, which first demonstrated that the rich concepts and methods of literary theory could be applied to the mass media and everyday life; and the recent sociological methods of “biopolitics,” which simply means the politics of life, the ways in which power is exerted today not primarily by violence but by the production of life and its regulation.
Jenna Holloway
Information Science
Mentor: Dr. Geri Gay
Category of Research: Social Science
ArtLinks: Fostering Social Awareness and Reflection in Museums
Technologies in museums often support learning goals, providing information about exhibits. However, museum visitors also desire meaningful experiences and enjoy the social aspects of museum going, values ignored by most museum technologies. We present ArtLinks, a visualization with three goals: making connections between a visitor and exhibits and other visitors by highlighting those visitors who share their thoughts; encouraging visitors’ reflection on the social and sacred aspects museum going and their expectations of technology in museums; and doing this with transparency, aligning aesthetically pleasing elements of the design with the goals of connection and reflection. Deploying ArtLinks revealed that people have strong expectations of technology as an information appliance. Despite these expectations, people valued connections to other people, both for their own sake and as a way to support meaningful experience. We also discovered several of our design choices in the name of transparency led to unforeseen tradeoffs between the social and the sacred.
Daniel Katz
Neurobiology/Economics/International Relations
Mentor: Professor Tim Devoogd
Category of Research:
Tracing hippocampal memory formation in the European Starling
In humans, the hippocampus has been implicated in learning and spatial memory, and produces outputs to multiple cortical regions. Because of the complexity of the human brain and difficulty studying it in vivo, these outputs have not yet been located. The recent technology of diolistics provides a rapid and efficient way to transfect neuronal cells with fluorescent dyes, enabling mapping of complicated neuronal networks. By using a gene gun and multiple diolistic dyes, the neural networks found in the hippocampus of the European starling (Sturnus vulgaris) can be traced to determine inputs and outputs to this region. It is hypothesized that this will identify avian homologs for cortical regions involved in the limbic system, and memory storage
Rachel E. Kulick
Archaeology
Mentor: Professor Sturt Manning
Category of Research: Social Science
Cypriot Dendrochronology: Historical Dating and Climatological Applications
By comparing tree-ring growth patterns of closely related species which have grown under the same climatic conditions within the same region, it is possible to ascertain a specific tree-ring pattern which is shared by all members of one species and is unique to that species, environment, and area. The aim of this study has been three-fold: to extend the chronologies for Pinus nigra and Pinus brutia trees in Cyprus, and then to utilize these chronologies both to establish dates for historical structures and to reconstruct past climatological records. Results from this study have extended the Pinus nigra chronology (AD 1142-2007) from the Chionistra Area and the Pinus brutia chronology (AD 1637-2007) from the Doxa Soi o Theos Trail Area. Further indications have been that 20th century renovations were made to Agios Sozomenos, a 16th century church located in the Troodos Mountains, and that historical samples with longer ring-counts must be obtained in order to successfully cross-date other Cypriot historical structures and strengthen the existing chronologies. Preliminary results from dendroclimatological analysis have indicated that, as determined by previous studies in the region, the monthly May, June, and July regional precipitation records demonstrate significant correlations with tree-ring growth patterns. Thus, if employed in conjunction with other environmental predictors, Cypriot tree-ring chronologies may be utilized to accurately reconstruct the past precipitation conditions for Cyprus.
Young Lu
Biological Sciences
Mentor: Professor Robert Weiss
Category of Research: Life Sciences
Effects of the Deregulation of Ribonucleotide Reductase
Ribonucleotide Reductase (RNR) catalyzes the synthesis of deoxyribonucleoside diphosphates from the corresponding ribonucleoside diphosphates, which is the rate-limiting step in the synthesis of deoxyribonucleotides (dNTPs). RNR is composed of two subunits, which must combine for catalytic function. The larger subunit R1 contains the active and allosteric sites while the smaller subunit R2 harbors a diiron center that generates a radical needed for enzymatic activity. In mammals, RNR activity is regulated in part by levels of the small subunit and by negative feedback. We are investigating the effects of deregulation of RNR activity. In order to reduce RNR feedback inhibition, we engineered a mutant large subunit (Rrm1-D57N) that has a non-functioning allosteric site. We also deregulated RNR activity by elevating levels of R2. Although elevating levels of the mutant R1-D57N in mice had little effect, elevation of levels of the small subunit induced lung tumorgenesis through a mutagenic mechanism. It is known that elevated levels of dNTPs may lead to incorrect base pairing. It is also known that radical-generating pathways may induce oxidative DNA damage and thus mutagenesis. Thus, we are trying to determine whether R2-induced mutagenesis is a result of RNR enzymatic activity or the consequence of the intrinsic properties of R2. The second aspect of the project focuses on measuring RNR activity in tissues and cell lysates over expressing theses subunits. In order to validate the activity assay, we will purify RNR subunits from bacterial cells overproducing these subunits, combine known levels of purified RNR subunits and assay their activities. We expect that combining the R1-D57N mutant with the small subunits will result in higher activity than the combination of wild type subunits.
Ezekiel Rediker
College Scholar, History, Africana Studies
Mentors: Ray Craib, RCPRS advisor, and Sandra Greene, thesis advisor
Category of Research: Other
Sierra Leone’s Diamond Rush and the Colonial Response
The diamond is a tiny, easily transportable, and immensely valuable resource. It was, at one and the same time, utterly unnecessary to human survival, and yet the cause of tremendous violence and bloodshed. In essence, the diamond is the quintessential luxury good. Possession of this resource has been a source of both human misery and happiness throughout the world, and nowhere more so than in Sierra Leone. Between 1952 and 1956, a period known as the diamond rush, lawlessness and violence characterized the Sierra Leonean hinterland. The diamond rush was the culmination of many years of colonial neglect and exploitation. Sierra Leone was formally a British colony between 1896 and 1962. The discovery of diamonds in 1930 drastically altered the economy of Sierra Leone, and created sharp changes in British colonial policy. Between 1932 and 1951, the government and a British mining company known as the Sierra Leone Selection Trust (SLST) extracted huge quantities of diamonds and faced little resistance in their operations.
The compounding effects of underdevelopment in Sierra Leone contributed to a rising consciousness about the value of diamonds among its people. By 1951, people were aware that extremely valuable gems were strewn all over the Sierra Leone countryside, and a mad scramble ensued. Thousands of people from all over the world descended on this tiny undeveloped nation in the hopes of finding, transporting, or buying diamonds. Massive slums proliferated in the diamond towns. Sierra Leonean farmers neglected their harvests. The rising price of food led to hunger and starvation, setting off strikes by the trade unions, and food riots throughout the country.
My CPRS project will analyze colonial response to the diamond rush. Special emphasis will be placed on the ways in which the British pursued a strategy of underdevelopment in Sierra Leone, as well as the police operations that resulted from the proliferation of illegal mining.
Tamar Weinstock
History and English
Mentor: Mary Beth Norton
Category of Research: Other
The Solitude of Self: Transcendentalism and Elizabeth Cady Stanton’s Feminist Thought
Although most historical research on Elizabeth Cady Stanton has focused on her work as an activist and leader in the early woman’s suffrage movement, recently more attention has been paid to her as an important feminist and political thinker. My study aims to contribute to this growing project, by focusing on an overlooked element of Stanton’s thought: the transcendental influence. I trace the biographical links between Stanton and the circle of transcendentalists she encountered in Boston in 1840s. I also look at the persistence of transcendental ideas of self-culture, self-reliance, and self-development in her writing up until the end of the century. I argue that Margaret Fuller, the first to apply these transcendental ideas to the position of women, was a much larger influence on Stanton than has previously been acknowledged. Finally, I identify the overlooked language of romantic individualism in Stanton’s 1892 speech, “The Solitude of Self,” and analyze how Stanton used this to provide a rationale for women’s rights. Tracing the transcendental thread in Stanton’s thought allows us to develop a deeper understanding of Stanton and a more nuanced view of the women’s movement in nineteenth century American. The ways in which Stanton’s views on selfhood and subjectivity, which are revealed in this context, prefigure aspects of twentieth century feminist thought, bears further scrutiny.
Angie Wolfgang
Physics
Mentors: Jason Wright and James Lloyd
Category of Research: Physical Sciences
Ruprecht 147: A Close, Old Open Cluster as a New Benchmark in Stellar Evolution
At an estimated age of two billion years and an estimated distance of 200 pc, the star cluster Ruprecht 147 was recently identified by Kharchenko, Piskunov, Roser, Schilbach, & Scholz (2005) as the closest old open cluster. Here we identify > 100 probable stellar members determined through position, photometric, proper motion, and radial velocity cuts, and we calculate an improved estimate for the cluster’s mean radial velocity. With this expanded membership list, we present the cluster’s new age and distance estimates based on Padova isochrone fitting and rudimentary extinction adjustments; we find that the cluster is older and more distant than is reported in Kharchenko, et al. (2005). SME analysis further indicates that the cluster is roughly solar metallicity. Future work will include a robust determination of the age and distance to the cluster by the assignment of membership probabilities to the members identified here. Once its age and distance are calculated with high confidence, Ruprecht 147 will become an important benchmark in stellar evolution: its relative proximity will allow us to observe its member M dwarfs and thus observationally calibrate the age-activity relation found for low-temperature stars by Hawley, West, Bochanski, & Covey (2006).
Juniors
Alex Ainslie
Information Science
Mentor: Dr. Geri Gay
Category of Research: Other
Modeling Community: Mobile Sensing and Network Analysis
Commercially available devices with embedded sensors and networking hardware are becoming increasingly pervasive. This preliminary work examines the sensed digital traces shared by such devices to inform a richer understanding of social behavior and network structure. Two sensing platforms are being developed to continuously scan and log information about discoverable, collocated Bluetooth-enabled devices. Analysis of collected data will focus on community structure and will incorporate theory from the social sciences.
Dorian Bandy
Music, College Scholar
Mentor: Neal Zaslaw
Category of Research: Humanities
The Romantic Lied: Text and Music
At the height of Germany’s so-called Sturm und Drang movement in the late 18th and early 19th centuries, composers became fascinated with the expressive possibilities of poetry set to music, and the Lied, both as individual songs and as semi narrative cycles, became a form central to the development of music over the next 75 years. My research and analysis focuses on the works of three poets and three composers, and examines both the linguistic content of the poetry (metrical issues, syntax, and poetic form) as well as musical construction (harmonic arcs, through-composition the cycles as cyclical.) At the heart of my analytic interest are the songs of “As die ferne Geliebet” (poetry by Willhelm Muller, music by Franz Schubert), and “Dichterliebe” (poetry by Henrich Heine, music by Robert Schumann).
Grace Chen
English & Biology
Mentor: Matthew Belmonte
Category of Research: Other
Efficacy of a Prompted-Pointing Therapy in Improving Learning Behavior in Non Speaking Children with Autism
Non-speaking people with autism are not typically included in neuro scientific studies. Analysis of DVD-recorded sessions of non-speaking children with autism undergoing a communication therapy provides insight into the specific interactions between autistic behaviors and learning-related behaviors over time. Exploring learning-related behavior in children with autism will help to clarify the cognitive processes involved in how non-speaking individuals with autism learn. The effect of communication therapy on learning-related behaviors and behaviors inhibiting the learning process in non-speaking children with autism were examined through a longitudinal analysis of video recordings. Leaving aside the question of whether prompted pointing reliably evokes valid and independent communications, analysis of video-recorded sessions of non-speaking children with autism undergoing this therapy suggests that exposure to the therapy has a positive effect on joint attention and decreases time off task. These data can provide valuable insight into the specific relationships between autistic behaviors and learning-related behaviors over time.
Michael Grundler
Biological Sciences, Ecology & Evolutionary Biology
Mentor: Kelly Zamudio
Category of Research: Other
Population and Conservation Genetics of Brazilian Frogs and North American Snakes
I use molecular DNA markers to infer patterns of gene flow and relatedness among populations of amphibians and reptiles. I have been working on a Brazilian species of frog trying to understand the consequences of habitat specialization for structuring genetic diversity across the landscape. Additionally, I am developing a molecular phylogeny of Eastern Northern American Pinesnake populations in an effort to identify potential areas and populations for use in conservation relocation strategies. The Eastern Pinesnake is threatened or endangered throughout much of its range making this work particularly relevant. Finally, I have also been working with collaborations in the Zamundio Lab on understanding the amphibian disease chytridiomycosis and how interactions between host ecology and environment affect disease prevalence in natural communities.
Elena Kwan
Biology
Mentor: Daniel Barbash
Category of Research: Life Sciences
Untitled
Last semester I worked with Dr. Daniel Barbash to study the role of the HMR protein, which is believed to play a role in oogenesis in female drosophila.
Qingchen Liang
Biology
Mentor: Monica A. Geber
Category of Research: Life Sciences
Analysis on Soil Seed Bank and Demographics Data in Clarkia Xentiana
Over the last summer, I developed a computer program that extracts Aggregations of Raw data on the Demography of Clarkia Xentiana Inputted in A specific format, organizes the data, and randomly generates a large amount of (# of seeds)/plot data based on the original data. The consistently skewed histogram of the output immediately suggests the inappropriateness of using certain statistics, such as the mean, to describe or model the population. The code can be easily modified to read different data or to extend on its present capabilities. I also worked on Professor Geber's team for field experiments and data collection in California at the beginning of last summer. The tasks we did include labeling and operating on flowers for a reproductive assurance experiment and planting plastic grids in preparation for an upcoming experiment.
Over the past academic year, I worked on conceptualizing a group of models that would allow us to project the population of Clarkia xentiana based on various demographics and Life History Parameters, some of which we do not yet have. The significance of such a model is that it would be able to put limits on the values of unknown parameters based on the known persistence of Clarkia xentiana at the sites we studied. It will also suggest the significance of Processes that we have had difficulty observing directly, such as the soil seed bank.
Kim-Yen Nguyen
Astronomy
Mentor: Prof. Martha Haynes
Category of Research: Physical Sciences
A New Population: Gas-Rich Massive Galaxies
I have been spending the past academic year abroad at Durham University in the UK and have tried to get involved in some of the research in optical astronomy that Durham University excels in. While department members' availability did not allow me to embark on a full-fledged project during my time abroad, I was able to set up occasional meetings with one of the Astronomy department's professors, Prof. Martin Ward, to explore and discuss the optical characteristics of high mass galaxies belonging to a small sample (37 objects) I began investigating the previous year at Cornell. This sample was compiled using data taken in the radio frequency from the Arecibo telescope in Puerto Rico. The radio data allows us to measure the gaseous mass of galaxies; the sample was therefore generated by selecting the most massive gas-rich galaxies from all of the data taken. This sample was further refined by calculating the stellar mass of the galaxies from optical data taken in other surveys (Sloan Digital Sky Survey) and retaining the galaxies with the highest gas-to-stellar mass ratios.
While at Durham, I have been looking at these galaxies' optical properties, such as their morphologies, colors, and spectra. These optical properties can reveal whether the galaxies are currently forming stars, where they lie in the galactic evolutionary scheme, and what kind of environment they exist in. Once these are determined for all galaxies in the data set, we can determine what kinds of trends (if any) exist and if we can make any conclusions about this unusually massive population as a whole.
Aman Prasad
Chemistry & Chemical biology
Mentor: Frank Schroeder
Category of Research: Life Sciences
Novel Sulfated Nucleosides from Spider Venoms
Natural products from terrestrial animals are generally underrepresented in screening collections. Our own research on the chemical ecology of arthropods has resulted in the identification of structurally unique compounds like sulfated nucleosides, most of which have not yet been subjected to broad biological screening because the amounts available from the respective arthropod sources are very limited. The goal of this project is to synthesize four small libraries of 20-30 compounds each, based upon four arthropod-derived lead structures. For the initial stages of this project, two groups of arthropod compounds have been selected, for which limited testing has indicated interesting activities: the sulfated nucleosides, which show specific affinity to kainate receptors, and the azamacrolides, which bind to imidazoline-1 receptors. The arthropod-inspired libraries will be formatted for broad biological screening at the NIH and the Investigator-Initiated Screening Program at the Broad Institute, Cambridge, MA
Sophomores
Alexander Gorovits
Biological Sciences
Mentor: Holger Sondermann
Category of Research: Life Sciences
Protein Components of Cell Signals in Biofilm Formation
Certain bacteria are capable of entering a sort of communal state referred to as a biofilm. Formation of these biofilms results in much more environmentally resistant bacteria, and has been implicated in the pathogenesis of some diseases. One of the signals responsible for initiating this state (as well as certain other pathways) is a molecule called cyclic di-GMP. This compound is especially interesting from a drug development standpoint because it exists only in bacteria, and is thus more easily targeted. A number of proteins are responsible for the synthesis and breakdown of cyclic di-gmp. They all share certain structural features, specifically in the domains that catalyze the above reactions. One protein we are working with (fimx) contains both domains, which is itself interesting, and seems to serve a regulatory function - we have identified a potential interaction partner, which we are currently investigating. The lab is largely involved in elucidating the nature of these processes through examining the structures and interactions of relevant proteins. To this end, we use x-ray crystallography along with other experimental methods. We are also investigating possible inhibitors for an enzyme that catalyzes the synthesis of c-di-gmp. Work involves expressing the protein, starting from inserting the relevant gene into bacteria (molecular cloning). The protein is then purified and crystallized for x-ray analysis, or otherwise analyzed.
John Hinnefeld
Physics
Mentor: Matthias Liepe
Category of Research: Physical Sciences
Development of an RF Sample Host Cavity
One technique used in particle accelerators is to shoot the particles through cavities in which there is a strong electric field. These strong fields can be produced by applying a time varying input at a resonant frequency of the cavity. By exciting the cavity at a resonant frequency one can produce a large field with a (relatively) small driving input. The same principle is in effect on a swing set-if you pump your legs at just the right frequency you get a large result from a small input. These resonant frequencies are determined by the shape of the cavity, and usually lie in the radio frequency range, hence RF. One detail that must be considered in designing these cavities is the effect the field will have on the cavity itself. Different materials are able to withstand different magnitude fields before bad things happen. The point of my research is to design a cavity that will be able to test different materials for the maximum field they are able to withstand. This requires a cavity that has very large fields on the sample being tested and very small fields on the rest of the cavity. This is necessary to ensure that when a failure occurs (I.E., when the maximum field is reached) it is the sample that is failing and not the host cavity.
The majority of my research is done by computers. There are a couple of different programs that will calculate the resonant frequencies of a given shape, as well as the magnitude of the fields at various points in the shape. I model different shapes, trying to find the one that has the largest ration of field on the sample to field on the host cavity.
Aaron Hubbard
Mathematics
Mentor: Ann Forsyth
Category of Research: Social Sciences
Reliability of Neighborhood Mental Maps
For the last year I have continued my investigation of reliability in the sketches drawn by residents of Minneapolis when they were asked to draw their “neighborhood.” Reliability is whether or not people are consistent over time, meaning if you ask them a question and then the same question a month later they will give you a reliable similar answer. Reliability can be useful for cognitive sciences and psychology, but also to City and Regional Planning as a way of better defining the terminology used, in this case “neighborhood.” My physical activities started with finishing the measure of the maps, and since then my mentor and I have worked on an early version of the paper associated with the study. We intend to do some simple descriptive statistics on the numbers, but also more advanced analysis involving the nonnumeric parts of the maps. They are often very interesting, with circles and branches like webs or very descriptive labeling of many parts of the so called neighborhood. These factors will be linked to a demographic survey associated with each map, and we hope to glean relationships between various social patterns and reliability and map characteristics.
Allen Miller
College Scholar
Mentor: Deborah Streeter
Category of Research: Social Sciences
The Fading Newspaper (An Analysis of Newspapers: Past, Present, and Futures)
Throughout our nation’s history, newspapers have always been a staple of American media. From the Early Colonial Days to the civil war to the progressive era, newspapers have always impacted society as a whole. Yet in recent decades new technological advances have enabled news to be presented to the public in many different shapes and forms. As a result many major print publications, such as the New York Times, are struggling to remain profitable. This financial problem poses a significant problem for the newspaper industry. This Semester my research focused on a number of interesting areas related to this topic. I began with an examination of the evolution of the modern newspaper. I then studied the rising sources of media--sources that appear to be replacing the Newspapers. I then studied the growing trends in modern journalism. Of particular importance were the revenue problems newspapers are facing. Ultimately, I strove to develop enough knowledge to provide insight on the future of newspapers. This project cumulated with a research paper that was submitted for review with the Visible Hand.
Ryan Musa
Computer Science
Mentor: Michel Louge
Category of Research: Physical Sciences
Air Flow Across Desert Sand Ripples
One of the defining traits of a desert such as the Sahara is its ergs, or large fields of sand dunes. In the ergs, dunes can arise at lengths and breadths in excess of 500m. These mountains of sand are transient, moving on the order of a few meters per year. The surfaces of the dunes are covered in small, periodic ripples, formed by saltation. The desert winds blow more or less continuously, and often dislodge grains of sand (i.e., the saltating particles). After being raised up, the grains are flung down by wind/gravity into the dune. The impacting particle also dislodges additional grains of sand, albeit at lower velocities. The pattern of crests and troughs that forms the ripples occurs on the windward face of the dune. Over a small area, this pattern is relatively regular, with parameters dependent on environmental conditions. After a certain point, the sand at the trough of a ripple is much more densely packed than that at the crest. My project focuses on the effect of this packing on the saltation process, as well as on the pressure systems within the ripple itself (where, in effect, the ripple "breathes" through its crest). Over the course of this past year, I machined a model from fluidizing plastic. The pores in the plastic are on the order of those in a sand ripple (~50 microns). The structure of the model is similar to that of a sand ripple as well: the periodic surface has a wavelength of 10 cm and amplitude 3 mm. Using holes spread along the length of the ripple and drilled to precise depths from the bottom of the model, we were able to obtain measurements of the static pressure within the ripple, and to compare those measurements to Professor Louge's proposed model.
Benjamin Strauss
Computer Science
Mentor: Mats Rooth
Category of Research: Other
Reranking of Parse Trees and Using Adverb Classification
My research has centered around automatic parsing of English sentences using probablistic context free grammar (PCFG). This model of language that encodes the syntax of a sentence as a tree and learns rules to describe the structure of trees; E.G. S->NP VP means that the sentence (s) is made up of a noun phrase (np) and a verb phrase (vp). The PCGF also learns the probability of each rule occurring by training on a large corpus of several thousand sentences that have been annotated by linguists. In the first part of the year, we worked on developing a classification system for adverbs that could provide the parser a way of re-ranking the 1000 best candidate trees it predicts for the test sentence. The idea is that if a reasonable set of classes of adverbs is defined that partitions adverbs based on certain criteria such as the parent or siblings of the adverb in the syntax tree, the probability of a certain word appearing in a certain class can be estimated. Then, after the parser has produced 1000 candidate trees, it can re-rank them based on the probabilities of its adverbs being in the classes which they were placed in those host trees. To be able to do this, it was necessary to produce a set of reasonable classes, which were evaluated using a metric known as cross entropy, a measure of how costly it is to identity all the different words given the classes in which they are placed. Some sets of classes were created by hand; an automatic method was also used where one class was created for each possible parent of an adverb, and then these classes were merged together until peak cross entropy was attained. However, the hand crafted sets still gave better results.
Finally, after a reasonable set of classes was decided upon, a system developed to link all the parts of the parser together to see if this re-ranking method gave better results. Data is still being collected at this time.
Freshman
Swarn Arya
Chemistry
Mentor: David Usher
Category of Research: Life Sciences
Untitled
Most are familiar with the Darwinian Theory of Evolution; the ideas of natural selection and survival of the fittest are essentially common knowledge in modern society. However, few are aware of the profound implications Darwin's theory holds for those studying the origin of life on earth. Many scientists agree that the ancestor to all life may have been a collection of molecules that is simple enough to self-replicate, but complex enough to take on the essential properties of a living organism. The search for the simplest possible form of life lead to the development of the 'RNA world' hypothesis, in which primordial cells lacking the ability to perform protein synthesis use RNA both as an enzyme and as a repository for heritable information. These ancient cells then evolved, via the processes of natural selection and survival of the fittest, to give rise to the many complex forms of life that exist today.
The first step to assessing the validity of this hypothesis is the successful assembly of an RNA molecule that can act both as a template for the storage of genetic information and as an enzyme that can replicate its own sequence and catalyze the formation of peptide bonds (the bonds that hold amino acids together in proteins). Dr. Usher is actively pursuing the last of these three goals: the design of a short strand of RNA that can catalyze the formation of peptide bonds. My work this semester focused on studying current literature in the field of RNA chemistry and origin of life research, and developing a strong background in theoretical and experimental organic and biochemistry. I plan to continue my studies this summer, and commence an original research project in the fall.
Jing Yee Chee
Physics
Mentor: Georg Hoffstaetter
Category of Research: Physical Sciences
Dependence of Optics of Light Source on Electron Beam Parameters
When a charged particle follows the circular path around an accelerator, it experiences acceleration and thus emits radiation. Such radiation is known as synchrotron radiation. For accelerators that aim for higher particle energies such as the LHC, synchrotron radiation is actually a nuisance as it leads to energy losses and limitations on accelerator design.
However, the properties of synchrotron radiation also lend itself to numerous applications. Synchrotron radiation is one of the brightest x-ray sources known to man, and many accelerators have been built for the sole purpose of producing x-rays for applications such as x-ray crystallography. These often use insertion devices such as undulators, which are a series of magnets which bend the electron beam in a way such that cause the radiation to be emitted quasi-monochromatically, with extremely high spectral brightness. In this project, I will be investigating how optics from an undulator is dependent on parameters describing the electron beam. I will be verifying the results obtained from SPECTRA, a code that does such calculations. The results from this project will be useful for the construction of the Energy Recovery Linac which is a planned extension to the Cornell Electron Storage Ring aiming to produce photon beams that are even brighter than currently possible. To this end, I will be writing a program that does the same calculations to check that SPECTRA can indeed simulate the optics that we are investigating.
Sudesh Kalyanswamy
Undeclared Major
Mentor: John Hubbard
Category of Research: Other
Solutions of Loewner Differential Equation
The Loewner differential equation is a first-order complex differential equation. One parameter of the equation is a driving function, say L(t), and we are analyzing solutions of the equation with a given L(T). Our solutions can be graphed on the x-y-t coordinate axes, and we are trying to analyze the points where if you start there, you will eventually hit your function L(T), which lies only in the x-t plane. A lot of work has already been done in the area, and our analysis (So far) has reproved everything in a much simpler manner, and the proofs we have constructed are much easier to understand. We are using the method of funnels in order to accomplish this. A funnel is simply a "FencE" where any solution entering the funnel will stay inside the funnel.
In addition, we are modeling solutions on the computer in order to fully understand the nature of these points which eventually crash into L(T). Again, those models have confirmed previously proven ideas.
Ivana Thng
Mathematics
Mentor: John Hubbard
Category of Research: Other
Stochastic Loewner Evolution - an Elementary Approach
We worked on a more elementary proof of the loewner differential equation for k < 4 for a specific equation, showing from the differential equations perspective of fences and funnels that it is not a simple arc by bounding the map of solutions from the upper half-plane to the reals with a 3-dimensional funnel that catches the solutions, showing that for periodic spaces on the upper half plane where brownian motion has K > 4, the solutions crash into a single place on the real plane. We also did a programme in java that helped create graphics of the map of the loewner equation for various driving functions using the runge-kutta numerical method of solving differential equations. This programme helped us to visualize the non-simple trace of the loewner equation for when k > 4. Currently we are still trying to generalize our proof to brownian motion, which is holder continuous for C < 1/2.