Student: Walter Ingram

The goal of this project is to determine the density and fraction of Active Galactic Nuclei (AGN) in different redshift bins: we chose .01-.03 (close distance); .4-.5 (medium distance); and .7-.9 (far distance). By looking at random parts of the sky using the NASA Extragalactic Database (NED) we were able to obtain a survey of about 500 close objects, 250 medium objects and 150 far objects. By entering right ascension and declination coordinates into NED the search returns all objects within a 20-arc minute circle that have been found by various satellites and telescopes around the world. Each search can return anywhere from 0-10,000 results. However, most of these objects have not been thoroughly researched and do not have redshift determinations. By adding redshift to the search criteria we get results that are almost all previously studied objects that have been defined as Galaxies, Quasars, UV Excess, or Radio Excess.  We found that at the smallest redshift it is easiest to find objects; however, the odds of finding an AGN are very low.  At the medium and large redshifts finding objects takes more trials, but we find greater fractions of AGN. There are two main reasons for this: (1) there are more AGN’s in the past and the bigger the redshift the farther back in time we are looking;  (2) AGN’s are much more luminous and outshine other galaxies around them, so at farther distances we cannot see the dimmer galaxies but we can still see Quasars and Seyfert Galaxies. Our survey of randomly selected patches on the sky included less then 1% of the entire sky. To do this project more thoroughly one would expand the survey and look at more patches of the sky and make corrections for the selection effects that favor the detection of Quasars.

Student: Megan Hartline

This summer Melanie Crampton, Walter Ingram, and I worked with Impress-Ed, a program funded by NASA that gives future science educators the opportunity to learn more about earth science by participating in focused research in either astrophysics, geophysics, or atmospheric physics. During the first two weeks of the program, we were engaged in the “common module”, where we learned content involved with those fields of earth science from Dr. Magee, Dr. Benoit, Dr. Wiita, and Dr. Kavic. In addition to content, we learned science pedagogy by discovering helpful resources available in the science community.

Following the two weeks, we were each paired with a mentor based on the field in earth science we were interested in. I worked with Dr. Magee in atmospheric physics. More specifically, I focused on tornadogenesis, which is our current understanding of how tornadoes form. Using NASA’s A-Train satellites, particularly CloudSat, as well as records of tornados from the past six years available through the National Weather Service, I was able to find times when the satellite passed over a thunderstorm that produced a tornado. A computer program, originally written in MatLab by Rachel Goldberg with Dr. Magee in the spring, was designed to scan every day since 2006 and formulate a list of these matches with respect to close latitude, longitude, and time. The CloudSat satellite gives us more information about the reflectivity and altitude of the clouds in its path, which helps to understand the formation of tornados. After the lists were devised, I narrowed down the matches in more detail by using the images from Calispo and MODIS, two more satellites in the A-Train, to find thunderstorms with not only a close match between the time of the tornado touchdown and the satellite overpass, but also significant reflectivity and altitude. I focused in on one F3 tornado from May 2^nd, 2010, which touched down on the border of Mississippi and Tennessee. Using the National Weather Service’s Weather and Climate Toolkit, I was able to look at the base reflectivity, storm relative velocity, base velocity, and enhanced echo tops of the section in the thunderstorm where the tornado was formed. From there, I compared the current understandings of tornadogenesis with the results I found through these satellite images.

Student: Melanie Crampton

The lithosphere of east coast of the United States has been modified dramatically by two different supercontinent cycles over the past 1.2 billion years.  A supercontinent cycle involves continental collision and subduction as tectonic plates converge, as well as rifting as the continental plates break apart and diverge.  The Northeast of the United States has been a site of numerous collisions and rifts during these supercontinent cycles.  In order to learn more about the structure of the Earth in the past as well as the process of rifting, we can examine the crust of the Earth in the Northeast to determine exactly where rifting occurred and why it occurred.  This summer, we analyzed gravity data collected by the United States Geologic Survey to help understand the crustal structure.  Gravity can enact a different amount of force on certain areas on the surface of the Earth based on the density of the materials under the Earth’s surface.  The thinner the crust, the shallower the high-density mantle is to the surface, creating a stronger gravity force in that particular area.  In order to analyze the gravity data, we used two different techniques: inverse modeling to estimate crustal thickness values and analysis of the gravity gradient, which shows finer scale variations in the gravity data.  To inverse model the gravitational data, we used a Matlab program, 3DINVER.m.  This program inverse models the data in the frequency domain and then Fourier transforms the results to calculate crustal thickness values.  The inverse model results correlated very well with crustal thickness values found using seismic analysis, except in the area of the Scranton gravity high.  Here, 3DINVER.m estimated a thin crustal thickness based on the high gravity values, while seismic analysis predicted a much thicker crust.  This discrepancy confirmed our hypothesis that the Scranton Gravity High is the location of a failed rift basin from ~ 1 billion years ago.  Analysis of the gradient of the gravity data along with known fault lines suggests that there are likely small scale variations in crustal thickness that are unexplored along the margin.  More detailed seismic studies are needed to confirm these results, but these studies would help elucidate the location and scale of known and unknown rift basins along the eastern margin of the US.

Student: Kayla Spector

The focus of this project was cancer cell invasion, the process by which cancerous cells leave the primary tumor site and enter healthy tissue.  Invasion is the first step of metastasis, the formation of secondary cancer colonies.  My goal was to build a mathematical model which accurately describes how the tumor microenvironment impacts cancer cell proliferation and invasion.

Using MATLAB, I developed a computational algorithm based on biological knowledge of how cancer cells grow and invade.  It involves several components: proliferating, invasive, hypoxic (low oxygen), and necrotic (dead) cancer cells as well as the extracellular matrix (ECM) which surrounds them. The algorithm is initialized by introducing a small tumor in a simulated tissue region containing an ECM of uniform density.  A novel feature of our biologically-based algorithm is how cancer cells interact with the microenvironment: the probability of division and invasion depend on pressure imposed by the ECM, and if a cell divides, the location of the daughter cell is determined using physical and geometric constraints of the microenvironment.  Further, physically-motivated functions are introduced for “pushing” the ECM as a tumor grows.

Using these rules we can study the dependence of tumor growth on ECM density. Tumors that develop in higher density environments are characterized by a small necrotic and hypoxic core with large amounts of branching and a higher probability of invasion, while low-density environments foster tumors that are more circular and isotropic with less invasion.  These findings are consistent with biological studies.  While we currently determine the probability that a daughter cell becomes invasive, our simulation of invasion is incomplete.  Once this aspect of the algorithm is developed, we would compare our output with experimental data and possibly calibrate parameters.  Eventually, a more complex microenvironment could be implemented to include blood vessels, non-uniform ECM density, and drug therapy.

Student: Jacob Levene

valuable for use in catalysis. One place nickel is utilized in nature is in the hydrogenase enzyme, a biologically relevant catalyst. As organometallic chemists, we are interested in identifying new nickel complexes for use in catalytic applications. Our work focuses on the rational design of nickel complexes for use in different avenues of catalysis. Nickel complexes catalyze a number of different reactions, including the formation of carbon-carbon bonds. However, an area in which nickel complexes have not been utilized effectively is in the hydrogenation of alkenes. This poster highlights our progress towards the synthesis and reactivity of new nickel complexes containing hemilabile groups. A hemilabile group has the ability to coordinate to a metal to stabilize open sites but is readily dissociated to allow for substrate coordination and further reaction. One focus of this work is towards the synthesis of nickel complexes containing a hemilabile pendent phosphonate arm and weakly coordinating ligands. Ideally, the compound will be stable, yet reactive enough to act as a catalyst. Another area of study is the use of nickel and palladium complexes to polymerize alkenes and dienes. This work also highlights our progress towards studying the reactivity of nickel complexes containing phosphine ligands with hemilabile arene groups to polymerize norbornene.

Student: Katrina Wunderlich

In our lab we wish to study the Brook Rearrangement on a series of aromatic silyl ketones. However, aromatic silyl ketones are not readily available commercially. Therefore, they must be synthesized. The current literature for the synthesis of silyl ketones tends to follow three basic steps:

1. Protection of aldehyde
2. Addition of TMS
3. Deprotection of ketone

Though the first two steps are largely straightforward, the deprotection in the third has proven more challenging. To date, literature suggests that the strategy most useful for the deprotection step largely relies on the use of mercury chloride which is toxic and expensive.  In our lab, we’ve developed a synthesis to deprotect the ketone using an inexpensive and safe reagent known as oxone. The third step of this synthesis, the optimization of the conversion of the protected ketone to the deprotected ketone has been the focus of this study. Results indicate that heat hinders this conversion and increasing the molar ratio of oxone to starting material to 4:1 decreases the time necessary for deprotection. To date in our lab, we have been able to increase the GCMS percent yield of starting material to desired product from 20% to >99%. We have analyzed this synthesis strategy on F, Cl, Br, and OMe substituted benzaldehydes as well as trans-cinnamaldehyde and a heterocyclic derivative. Further characterization and purification of these compounds are necessary before they are used to study the Brook Rearrangement.

Student: Tyler Higgins

Through previous research, it has become known that 1,2-cyclohexanedione can be used in Michael addition reactions. The use of 1,2-cyclohexanedione in Michael additions was previously restricted to reactions involving β-nitrostyrenes; this current project sought to expand the amount of Michael additions that 1,2-cyclohexanedione could undergo by examining reactions with different α,β-unsaturated ketones and alkynes.

After determining whether or not there exists a Michael addition reaction between 1,2-cyclohexanedione and these substrates, the next step is to condense the Michael adduct into a new heterocyclic ring structure. These heterocyclic ring structures are known to possess pharmacological activity as anti-cancers and anti-tumors.

Student: Daniel Ferrer

This project’s goal is to determine the function of two genes that are believed to function in the early stages of development of zebrafish embryos. The two genes of interest, hnrnpab and zgc:77052-201, are the two genes that are most homologous to the squid gene in Drosophila melanogaster. In D. melanogaster, squid codes for an RNA-binding protein that localizes the product of another gene to specific areas of the egg. The presence of the localized gene product influences the dorsal/ventral patterning of the egg. The functions of the genes in zebrafish were studied this summer through a series of microinjections. The embryos were injected with morpholinos specific to those genes between the one to two cell stages and the eight cell stage. Morpholinos are small antisense molecules that are designed to block the translation of a specific mRNA. The morpholinos were injected to prevent expression of the two genes under investigation. The embryos were fixed in methanol:DMSO and stained using immunohistochemistry to determine the effects of the injection on the expression of muscle-specific myosin, which is specifically expressed in somites in the embryo. Somites are embryonic structures that, in vertebrates, eventually develop into dermis, skeletal muscle, and vertebrae, and therefore are indicators of correct dorsal patterning of the embryo. The results of the injections suggest that blocking the expression of hnrnpab and zgc:77052-201 disrupts the sharp, chevron patterning of the somites and the somite size. Therefore, these initial results indicate that the genes of interest in zebrafish may be involved in dorsal/ventral patterning and possibly homologous in function, as well as structure, to the gene in D. melanogaster.

Student: Peter Swetits

Protein sequence alignments allow researchers to quickly determine regions of similarity between different proteins. They also provide important clues about the nature of the proteins that may be important to their study. While working with sequence alignments, researchers often find that they need to quickly shade or edit their alignments. However, the most widely used shading program, BoxShade, is difficult to use, does not allow editing, and has a limited number of output options. OpenShade is being developed as an open source software application that solves these issues. It allows the user to input multiple sequence alignments in all popular formats, including FASTA, ALN, MSF, and Phylip. Once imported, the alignment can then be dynamically shaded for identities and similarities, with the consensus being either automatically calculated or defined by the user. The user is able to specify the criteria to form a consensus, change the scoring matrix, and set the minimum score required for shading of either identical or conserved residues. After shading, the user has the ability to edit individual amino acids, entire columns of amino acids, or select and edit a single section of the entire alignment. The shaded sequences can be exported as a document in PDF, PNG, or RTF formats.  OpenShade also contains the ability to conduct pattern matching using regular expressions. The user can input a string of amino acids and then the program will highlight all occurrences of that string in each of the protein sequences independent of position.  The basic graphical interface and shading algorithms have been completed.  We anticipate the completion of the project within the next year.

Student: John Fang

The objective of this summer’s project was to the study the regulation of GLD-1 (defective in Germ Line Development), a RNA binding protein that is important for normal germ-line development in the model organism Caenorhabditis elegans (C. elegans). In mutant strains where GLD-1 is not expressed, hermaphroditic oogenesis ceases and the germ-line tumors form. Therefore, correct expression of GLD-1 in the germ-line of C. elegans is critical and is highly regulated by various cellular mechanisms. One suspected mechanism is the phosphorylation of the GLD-1 protein for the modulation of its functionality. In normal animals, GLD-1 protein levels gradually increase from the mitotic zone through the transition zone, and reach the highest levels of expression during the pachytene phase of meiosis. Prior to oocyte development, GLD-1 levels drop abruptly. By analyzing the phosphorylation levels of GLD-1 protein in germ-line, the relationship between phosphorylation and the levels of expression of GLD-1 may be explored. Western Blotting is common method used detect changes proteins according to mass, and may be used to analyze whether or not the phosphorylation of GLD-1 is actually utilized as a regulatory mechanism. This summer was focused on optimizing the detection protocol for phosphorylated GLD-1 versus non-phosphorylated GLD-1 isoforms. I plan on continuing this research in the future with more focus on the specific stage GLD-1 is phosphorylated as well as the specific effects of phosphorylation.