Astronomy

Data Compilation and Analysis of Stars in the Orion Nebula Cluster

Sara M. K. Dwyer

I took nine surveys of the Orion Nebula cluster and combined them into one master catalog. Using this catalog, I examined I-magnitude data to see how the cluster has changed over time in terms of magnitude variability. I found that large and small amplitude variable stars have different positional distributions in the cluster.

Biology

Molecular Strategies for Augmenting Hydrogen Production by Synechocystis 6803

Michael J. Abrams

The cyanobacterium Synechocystis PCC 6803 is an excellent model organism for the production of biologically-produced hydrogen gas (H2) for use as a fuel source. In order to maximize H2 production, we have sought to culture Synechocystis under conditions in which it is actively photosynthesizing − and consequently evolving O2 − in order to stimulate the production of NADPH, the reducing power of which is used by the bidirectional hydrogenase enzyme of Synechocystis to produce H2. However, the hydrogenase enzyme is acutely inhibited by O2. To overcome this problem, we identified two genes involved in the natural symbiotic relationship between legumes and rhizobium bacteria, which function to protect nitrogenase from O2 inactivation in a manner similar to that encountered in hydrogenase. Using these genes, we engineered two genetic constructs with which to transform Synechocystis. One gene encodes the O2-binding protein leghemoglobin (Lb), which has an extremely fast O2 association rate and supplies O2 to oxidative respiration or exports it from the cell. The other gene, encoding the protein ferric leghemoglobin reductase (FLbR), catalyzes the NADH-dependent reduction of ferric to ferrous Lb. We made one construct to express Lb alone and a second, using PCR SOEing, to express both Lb and FLbR. We have also designed a reaction chamber that allows us to test the two transformed strains of Synechocystis for H2 accumulation. We have yet to test our transformed strains, but we have determined basal levels of H2 and O2 evolution in wild type (untransformed) Synechocystis.

Small Heat Shock Proteins and Zebrafish Left-Right Asymmetry: Exploring the Functional Roles of hspb7 and hspb12 in Cardiogenesis

Jonah Zuflacht

The small heat shock proteins (sHsps) hspb7 and hspb12 have unique. unprecedented roles in the establishment of visceral left-right (LR) asymmetry. Morpholino (MO) knockdown of these genes randomizes cardiac laterality and perturbs normal southpaw, lefty-1, and lefty-2 expression (Lahvic, 2010). The mechanism through which hspb7 and hspb12 affect asymmetry, however, has yet to be determined. Here we present insight into the functional roles of these sHsps on a cellular level. Through transmission electron microscopy, we demonstrate that hspb7­ morphants exhibit perturbed ciliary ultrastructure and propose that these microtubule defects could underlie the randomization of visceral situs. Previous work has demonstrated that hspb7 and hspb12 are expressed specifically in cardiac tissue (Lahvic, 2010). We have repeated in situ hybridization for these genes and observe far earlier expression in the yolk syncytial layer, a transient extra-embryonic tissue with important roles in developmental patterning and heart migration. The expression patterns of southpaw, lefty-1, and lefty-2 in double morphants suggests that hspb7 and hspb12 work together to influence laterality. We also show that hspb7 is important for normal cardiac morphogenesis and heart function in addition to laterality.

Regenerative Effects of Exercise Following Global Ischemia

Kylie Huckleberry

Studies from animal models such as rats and gerbils have provided evidence that post-injury exercise can be neuroprotective and reduce damage resulting from ischemia. Exercise alone has also been shown to improve performance on the same tasks that ischemia impairs. The present study sought to explore the effects of exercise following global ischemia. Adult male mice were subjected to either global ischemia via a two-vessel occlusion (2VO) or a sham surgery. Twenty-four hours post-injury, animals were given either a running or stationary wheel for two weeks, during which period they were injected with 5-bromo-2-deoxyuridine (BrdU) three and four days post-injury. At the end of the two-week period, mice were behaviorally tested on the rotorod, the Morris Water Maze (MWM), and Contextual and Cue Fear Conditioning (CCFC) to assess motoric function as well as spatial and contextual learning and memory. Hippocampal slices were stained with Cresyl Violet (CV), Fluorojade (FJ), and BrdU. There were some minor motoric differences between the conditions, and all groups performed comparably on the tests of learning and memory. However, there is reason to suspect that some of the conditions used alternative strategies during the tasks that were not hippocampally-dependent such as knowing the distance from the platform to wall and swimming in a circle. Only the CV stain reliably worked, but no significant results were found. It is possible that this is due to a measure that was not sensitive to the cell death. In summary, I found no detectable differences resulting from surgery and therefore also found no evidence supporting the idea that running is protective.

An acute dose of estradiol immediately after transient global ischemia does not protect against hippocampal structural and functional damage

Marijke J. Devos

Both focal and global occlusion of blood flow to the forebrain leads to a variety of negative behavioral effects including declines in executive function, memory, language, speed of processing, and visuospatial/constructional abilities. The hippocampus, a structure in the forebrain implicated in learning and memory, is severely affected by transient global ischemia, and infarct is associated with neurodegeneration in the CA1 and neurogenesis in the dentate gyrus in rodent models. Pretreatment with estradiol in rodent models protects against cell death in the CA1, increases neurogenesis in the dentate gyrus, and improves deficits in hippocampally dependent learning and memory. However, controversy surrounds the use of estrogen therapy in women as a method of decreasing the negative effects of stroke, as estrogen therapy may actually increase the incidence of stroke in women. If an acute dose of estradiol could be administered immediately after an ischemic event, in a therapeutic manner, the negative consequences of both stroke and long term treatment with estradiol could be avoided.

Here, an acute dose of estradiol (4 mg/Kg) was administered to C57BL/6J mice immediately after 15 minutes of transient global ischemia and the structural and functional effects on the hippocampus were examined.

Histological sections were stained with cresyl violet for healthy cells and fluoro- jade B (FJB) for degenerating neurons, and labeled with bromodeoxyuridine (BrdU) for dividing cells. Motor coordination, spontaneous activity and anxiety, and learning and memory, were assessed with the roto-rod task, the open field task, and the context cue fear conditioning task.

Estradiol had no effect on hippocampally-dependent learning and memory, but marginally improved hippocampally-independent learning and memory after stroke. Estradiol decreased spontaneous activity, increased anxiety behavior, and a trend was observed where estradiol seems to decrease mortality immediately after stroke as compared to oil-treated animals.

Although no differences in FJB staining were evident between groups in the CA1, 2VO marginally increased FJB+ cells in the dentate gyrus and estradiol marginally decreased neurogenesis in the dentate gyrus.

To conclude, an acute dose of estradiol immediately after transient global ischemia does not affect hippocampally-dependent learning and memory, does not protect against neurodegeneration in the dentate gyrus, and decreases neurogenesis in the dentate gyrus. However, mortality after stroke was marginally decreased by estradiol treatment, suggesting that estradiol’s known immediate effects on neural plasticity may provide protection to the brain overall following occlusion of blood flow. Further, the marginal increase in anxiety in the estradiol group may have been caused by withdrawal from estradiol, which is known to increase anxiety and depression in both humans and rats. In the future, a shorter duration of occlusion, namely 10 minutes, should be used to prevent selective effects of the occlusion from affecting the results, and chronic estradiol treatment, for a week after ischemia, should be used to rule out withdrawal effects from estradiol in the behavioral result.

 

The Influence of Land Management Practices on the Abundance of Aster, Goldenrod, and their Pollinators in Williamstown, MA, USA

Jillian Hancock

Asters, goldenrods, and their pollinators form an important and delicate ecological relationship. Asters and goldenrods rely on pollinators to visit and cross-pollinate between plants so that they can reproduce. Pollinators need the pollen and nectar resources provided by these late blooming flowers in order to survive the cold New England winter. This relationship is in jeopardy due to declines in abundance and diversity of both plants and their pollinators. Numerous factors including loss of habitat, competition with nonnative species, disease, and maladaptive land management practices contribute to these declines. My study examines the influence of mowing schedules on abundance and diversity.

A field last mowed in October 2008 had significantly greater flower visitation than a field mowed in August 2010 (Paired t-test, t = 4.317, d.f. = 9, N=10, p = .001942). The unmowed field had a significantly more diverse pollinator species sample with Shannon Diversity Index of H = 1.09 +/- 0.12 SE compared to the mowed field’s index of H = 0.29 +/- 0.15 SE. The unmowed field had significantly greater number and diversity of blooming stems of asters and goldenrods than the mowed field. The unmowed field had 11,718 stems including 7 different species of asters and goldenrods and the mowed field only had 2,154 stems including 5 different species of asters and goldenrods.

Our local survey of Williamstown, MA, USA sampled 43 properties that included 1,066.5 acres of open land. Results showed that most (62% or 681.5 acres) open land in Williamstown is committed to agriculture that requires two to three mows per year.  Eight-hundred thirteen (813) acres have mowing schedules dedicated for specific purposes such as haying or ground-nesting bird conservation. Of the lands without committed mowing schedules (253.5 acres), caretakers and

property owners of 180 acres indicated they were willing to delay mowing until October or later to promote the growth of aster, goldenrod, and pollinator populations. Thus there is potential to increase the success of asters, goldenrods, and their pollinators in Berkshire County, MA.

The Role of the Agrobacterium tumefaciens T6SS in Modulating Host Defense Responses

Helen Cha

Agrobacterium mutants lacking the recently discovered Type VI Secretion System (T6SS) exhibit attenuated virulence; we hypothesized that this defect may be attributable to a diminished ability to dampen host defenses. In the A. thaliana ecotype Landsberg erecta, no transgene expression was found in seedlings infiltrated with T6SS-deficient bacteria, whereas all the seedlings infected with the wild-type bacteria strongly expressed the transgene. In contrast, a low level of expression was found with both wild type and T6SS-mutant infected seedlings of the Col-0 ecotype. Furthermore, infiltration with wild-type but not mutant bacteria inhibited growth of the seedlings, in the same way as exposure to known bacterial elicitors. We are currently testing the possibility that T6SS effectors are necessary to dampen siRNA-mediated silencing of transgenes delivered by the bacterium. This anti-silencing function may promote virulence by allowing expression of the oncogenes, but at the expense of triggering primary lines of defense. The apparent reduced susceptibility of the Col-0 ecotype to Agrobacterium transformation, coupled with marked growth inhibition of this line upon exposure to bacteria, may reflect its ability to perceive and respond effectively to pathogen-associated elicitors other than T6SS substrates. In contrast, the lack of the ERECTA kinase in the Landsberg ecotype may make it more prone to transformation as long as the pathogen can dampen other defenses via delivery of T6SS effector(s). Preliminary data suggest the Landsberg ecotype of A. thaliana is more prone to bacterial transformation than Col-0, which is consistent with the greater susceptibility of Landsberg to other pathogens.

Characterization of Chicken Ovalbumin Upstream Promoter-Transcription Factor in Helobdella sp. (Austin)

Mari M. Lliguicota

Our lab investigates segmental pattern formation in annelid, the segmented worm. Developmental mechanisms are poorly understood in this group; therefore the goal of my research project was to identify potential regulators that function during early annelid development. An unbiased search for gene products expressed in annelid segmentation has been conducted and a number of genes have been identified as potential candidate segmentation genes. One of the gene products was HauCOUP, which was identified as a member of the chicken ovalbumin upstream promoter transcription factor family due to high-shared amino acid sequence identity found through sequence analysis. In this thesis, I used in situ hybridization to characterize the spatial and temporal expression pattern of HauCOUP in the leech Helobdella sp. (Austin). Although the function of HauCOUP was not investigated, it is apparent HauCOUP is expressed all through out development, primarily enriched in mesoderm precursors. The expression pattern of HauCOUP suggests it is an early mesoderm marker. The understanding of developmental mechanisms of annelids still remains poorly understood but the expression profile of HauCOUP suggests HauCOUP is present for mesoderm segmentation.

Neuroscience Thesis: Mirror Neurons and Emotions in Motion

Fhatarah A. Zinnamon

When we observe someone performing an action, mirror neurons enable our brains simulate that same motion. One way to study this system of mirror neurons is to compare people who have and haven’t learned a certain repertoire of movement. The mirror neuron system has also been proposed as the neural basis of empathy. In this study, we examined the responses to emotional modern dance pieces in modern dancers and non-dancers. Participants were Williams College students, who completed measures of empathy and questionnaires about their responses to Martha Graham performance pieces. Results indicated that dancers had increased Empathy Quotient (EQ) scores and stronger connections to the piece rated as most emotional compared to non-dancers. All participants with higher EQ scores rated the most emotional piece as more emotional than participants with lower EQ scores. Attempts to quantify mirror neuron activation through electroencephalography failed due to high impedances and results were not recorded. This study extended previous action observation studies to modern dance stimuli and showed that modern dancers have higher empathy scores than non-dancers. Clinical implications of these findings are discussed, as are directions for future research.

Are We Singing The Same Song? Culture-Based Behavioral Differences in Savannah Sparrows

Clint Robins

The importance of dialect differences in songbirds and the origin of those differences was investigated by studying beak gape patterns and behavioral responses to playbacks of the buzz section of song in two populations of Savannah sparrows (Passerculus sandwichensis). Males from both populations exhibited more variability in beak gape patterns in the buzz section of song than in other song sections. However, beak gape patterns did not vary significantly across populations. This result suggests that either beak gape patterns were difficult to maintain during buzzes or that beak gape was adjusted to compensate for variability in other vocal parameters. Such song-to-song variability in acoustic parameters in beak gape patterns may allow the buzz to carry information about a male’s motivational or physiological state. Playback responses demonstrated that males from both populations discriminated between the buzzes of local and foreign dialects. Males in each population responded longer when the buzz stimulus was drawn from the local dialect. The stronger responses to playbacks of local buzz stimuli indicate that males perceive the local dialect to be more threatening. Taken together, these results suggest that a) the buzz carries information about dialect, and b) differences in the buzz portion of the song are not due to differences in the way the vocal instrument is used, but rather are subject to cultural influences mediated by learning.

Searching for Candidate Barrier Genes Related to Wing Coloration in Heliconius Butterflies

Leanne Lin

Speciation is a fundamental process in evolutionary biology. It is the main source of biodiversity, a remarkable claim considering the countless forms of life on Earth. However, as an evolutionary process occurring over long periods of time, speciation is difficult to study by direct observation. Therefore, to determine how new species evolve, a deeper understanding of barriers to gene exchange is crucial. These barriers lead to reproductive isolation, genetically and physiologically preventing new species from mating with one another. Sequence comparisons of the Cr/Yb locus were performed for three F2 broods of Heliconius butterflies derived from pure parental populations. Restriction enzymes were designed based on nucleotide sites that differed consistently for black, shadow, and yellow individuals. This approach of correlating specific restriction digest patterns with hindwing phenotypes allowed for 5^th instar larvae, early pupae, and ommochrome only pupae individuals to be genotyped. Based on these results, further analysis can be done using methods such as qPCR and next-generation sequencing to determine the RNA expression patterns in different stages of development. These subsequent steps will reveal how gene expression determines wing color, and hence which gene in the Cr/Yb locus may function as a barrier gene.

Characterization of an Orthopedia Homolog in Helobdella sp (Austin)

Alexandra M. Peruta

What molecular mechanisms operate in segmentation of the annelid phylum are important questions within developmental biology. Candidate gene products expressed during annelid segment formation in the leech Helobdella sp (Austin) were identified through a novel, unbiased approach. One candidate EST possessed high sequence similarity to the insect Orthopedia (Otp) gene, termed Hau-Otp. The comparative sequence data support the theory that leech Otp and fly Otp belong to a subfamily of homeobox genes. Multiple amino acid sequence alignments and a molecular phylogeny independently confirmed Hau-Otp’s identity. Whole mount in situ hybridization was carried out in Helobdella sp (Austin) embryos and juveniles to determine the spatiotemporal expression pattern of Hau-Otp. The differential and restricted spatiotemporal expression patterns implicate Hau-Otp in ectoderm, mesoderm, and cell fate specification. This is the first reported role of Orthopedia in early developmental patterning of the mesoderm. The phylogenetic and immunohistochemical data indicate that Orthopedia is present in multiple members of the annelid phylum as well as other protostomes and even deuterostomes. Therefore, the presence of Orthopedia in the molecular mechanism of development in the early Bilateria Helobdella sp (Austin) continues through evolution.

Chemistry

Exploring the Binding Mechanism of the LexA and RecA Proteins Through the Use of Cross-Linking Studies

Lucas Bruton

The mechanism underlying the regulation of the bacterial SOS response to DNA damage, the RecA-mediated autocleavage of LexA, is not well understood, due, in part, to the lack of knowledge regarding the interaction between the RecA and LexA proteins. Recent work in the Lovett laboratory has produced fourteen Bacillus subtilis LexA mutants in which conserved amino acid residues in the putative RecA binding domain were replaced with cysteine. The goal of this research was to use these LexA mutants to identify the corresponding binding sites on RecA using the bifunctional cross-linker 2-{N2-[N6-(4-Azido-2,3,5,6-tetrafluorobenzoyl-6-amino-caproyl)-N6-(6-biotinamidocaproyl)-L-lysinylamido]}ethyl methanethiosultonate (Mts-Atf-LC-Biotin), which contains a sulfhydryl-specific group and a non-specific azido group.

After purifying the fourteen LexA cysteine mutants, I successfully attached Mts-Atf-LC-Biotin to thirteen of the mutants as evidenced by Western analysis with streptavidin. Appropriate LexA mutant conjugates were chosen for cross-linking trials with RecA based on their ability to bind RecA in LexA cleavage assays. Unfortunately, no RecA attachment was observed with any mutant. After determining that RecA cross-linking conditions were optimal in LexA dimer cross-linking studies, I hypothesized that LexA autocleavage may have interfered with the cross-linking reaction. As a result, I used site-specific mutagenesis to change the catalytic serine residue on each cysteine mutant to an alanine to block the cleavage of these mutants. These mutants are now ready for future cross-linking studies.

Toward the Asymmetric Total Synthesis of Jerangolid D

Mary Beth Daub

Progress towards the asymmetric total synthesis of the natural product jerangolid D is described. Jerangolid D is a secondary metabolite produced by the myxobacterium Sorangium cellulosum and a potential antifungal agent. Jerangolid D bears a strong structural similarity to the better‑known metabolite ambruticin; the two molecules have identical right halves. The key structural features of jerangolid D include a cis‑dihydropyran, an a,b‑unsaturated d‑lactone, and a doubly allylic isolated stereogenic center. The cis‑dihydropyran has been constructed in eight steps from commercially available materials using the method for the elaboration of thiazolidinethione aldol adducts developed in our lab and a reductive carbon‑Ferrier rearrangement, which intercepts a key intermediate in Jacobsen’s ambruticin synthesis. Although the reductive carbon‑Ferrier rearrangement still requires optimization, the product structure has been verified by comparison to published spectral data. The doubly allylic stereogenic center will be established using the exact asymmetric hydroformylation reaction utilized by Jacobsen. Our route will then diverge toward the asymmetric total synthesis of jerangolid D where the a,b‑unsaturated d‑lactone will be prepared using chemistry previously developed in our lab for the synthesis of the kavalactones.

Toward the Total Synthesis of Enigmazole A:
Model Studies Directed to C₃ Hydroxyl Deoxygenation

Marian Deuker

Enigmazole A is a phosphorylated 18-membered macrolide natural product with seven chiral centers, isolated from the marine sponge Cinchyrella enigmatica. The molecule has emerged as a captivating synthetic target because enigmazole A exhibits novel architecture, and also because although enigmazole A possesses anticancer activity, its biological target remains enigmatic. We envision inducing asymmetry into the synthesis with a chiral auxiliary, and then using the tin-promoted Evans b-keto imide aldol reaction to increase molecular complexity through the modular installation of three stereogenic centers. However, the highly efficient Evans b-keto imide aldol method will introduce an unnecessary hydroxyl at the C₃ position. Herein we present research efforts to build a model system with a sterically crowded C₃ hydroxyl that mimics the actual system, in order to test the radical Barton-McCombie deoxygenation as a tactic to reduce the C₃ hydroxyl to a methylene.

The Iron-Catalysis of the Epoxidation of Plant-Based Fatty Acid Derivatives
and Other Acid-Containing Alkene Substrates

Matthew Everhart

[Fe(BPMEN)](OTf)₂ (2, BPMEN = N,N’-dimethyl-N,N’-bis(2-pyridylmethyl)-ethane-1,2-diamine), was used to catalytically oxidize a number of plant-based fatty acid derivates and other acid-containing alkene substrates. Reaction conditions were explored and optimized with respect to reaction time, temperature, atmosphere, solvent, reagent ratios, and oxidant using oleic acid as a model substrate. A 6 minute reaction time, an inert Ar atmosphere, 25mL of solvent MeCN, 600 eq. of oxidant H₂O₂, and 1 eq. of 2 led to 76% substrate consumption and 82% epoxide selectivity with 200 eq. of oleic acid. The addition of acetic acid, an additive crucial for achieving favorable selectivities and yields in previously-reported systems, was not required. Additionally, consumption appeared to peak between 25°C and 50°C and to drop at 60°C. Choice of solvent significantly impacted consumption but did not affect selectivity. With ethyl oleate, diminished consumption and selectivity were observed, confirming the importance of the carboxylic acid functionality in oleic acid. Catalyst composition and the role of acetic acid were also investigated. Specifically, [Fe(bipy)₃](OTf)₂, [Fe(phen)₃](SO₄), and [Fe(iso-BPMEN)](OTf)₂ were inferior catalysts relative to 2. While increasing oleic acid consumption to 100%, acetic acid did not improve selectivity. Epoxidations of 4-pentenoic acid and 5-hexenoic acid produced two useful lactone synthons with 100% and 93% selectivity, respectively. Finally, promising results of 65% consumption and 100% selectivity were achieved for the epoxidation of acrylic acid.

Undecylprodigiosin Production in Streptomyces coelicolor A3(2)

Moyukh Ghosh

The genus Streptomyces is a group of soil dwelling, gram positive, bacteria that are prolific producers of secondary metabolites including many of the clinically significant antibiotics used today. As such, understanding and maximizing antibiotic production in Streptomyces is essential to the medical field. The model organism Streptomces coelicolor produces 4 well characterized antibiotics, and essential to the biosynthesis of these antibiotics are phosphopantetheinyl transfereases (PPTases) which catalyze a necessary posttranslational modification of their biosynthetic enzyme. Previous studies have attempted to characterize the three PPTase encoded in the S. coelicolor genome. In characterizing the SCO6673 PPTase, researchers found that a SCO6673 PPTase knockout mutant had an unexpected 5 fold increase in undecylprodigiosin production relative to wild type.

The intent of this study was to first, elucidate the molecular mechanism behind the increase in undecylprodigiosin production in the SCO6673 PPTase mutant, and second, to further maximize the biosynthesis of this antibiotic in this strain. Towards the first goal, studies into activity of glycolytic and pentose phosphate pathway enzymes in the mutant, as well as quantitative real time PCR (qPCR) assessment of the expression of antibiotic regulatory and biosynthetic genes were performed. While the metabolic enzyme activity data remains preliminary, it suggested perhaps minor alterations in activity of glycolytic enzymes in the SCO6673 mutant. qPCR revealed that the increased undecylprodigiosin production in the SCO6673 mutant is, at least in part, due to the up-regulation of expression of undecylprodigiosin regulatory and biosynthetic genes. The expression of the redD regulatory gene was observed to be elevated as much as 8 fold in the SCO6673 PPTase mutant as compared to the WT and corresponded to an approximate 10 fold elevated expression of the biosynthetic genes redF and redN. Additionally, the SCO6673 mutant, which hyper-produces undecylprodigiosin, was further engineered in order to maximize undecylprodigiosin biosynthesis. Overexpression of redD in the SCO6673 mutant was successful in increasing undecylprodigiosin production by 25-50%. However, addition of a pfKA2 mutation to the strain actually decreased undecylprodigiosin production.

Efficient Asymmetric Synthesis of á,â-Unsaturated ä-Lactones
and Direct Refunctionalizations of N-acyl Thiazolidinethiones

Mark Johnson

A versatile and efficient asymmetric synthesis of á,â-unsaturated ä-lactones, a common structural moiety to many biologically active natural products, is described using a single-flask reaction involving three stages: 1) base catalyzed lactonization, 2) triflation, and 3) hydride reduction via conjugate addition and â-elimination. Ultimately, a range of á,â-unsaturated ä‑lactones were produced in excellent enantiomeric excesses (>95% ee) and good yields over the three steps (54–58%) from aldehyde starting materials. Additional work is reported on direct refunctionalizations of N-acyl thiazolidinethiones through acylation reactions involving benzotriazole and mildely-generated enolates of ketones and thioesters.

The Characterization of Arabigopsis Thaliana Mutants
Deficient in Ceramide Kinase

Elizabeth Kalb

Sphingolipids are a ubiquitous class of lipids that function as integral structural components of eukaryotic endomembrane systems and, more interestingly, serve as signaling molecules in diverse pathways. Sphingolipids and their complex derivatives mediate plant responses to drought stress and pathogen infection. Moreover, it has been demonstrated that ceramide, one sphingolipid species, initiates programmed cell death in Arabidopsis thaliana. Ceramide kinase appears to be particularly important for regulating the abundance of free ceramide and, consequently, moderates programmed cell death in plants.

Through characterizations of A. thaliana mutants deficient in ceramide kinase, we may better understand how ceramide accumulation affects plant fate. Although the naturally occurring ceramide kinase missense mutant accelerated cell death 5 (acd5) has been previously studied, we must corroborate these findings in other mutant lines to demonstrate that the results are not unique to acd5 plants. Here we study four mutant plants containing insertions in the CERK1 locus in addition to acd5. We quantify the ceramide kinase activity of these mutants relative to wild-type. More importantly, we provide confirmatory evidence that mutants deficient in ceramide kinase are more susceptible to cell death following exposure to abiotic stresses, including heat shock and dark-induced senescence. The objective of this is not only to confirm that ceramide kinase deficient mutants are predisposed to precocious cell death, but also to provide avenues for future research that might elucidate the pathways linking ceramides to programmed cell death.

Synthesis of the C₁–C₁₂ Fragment of Tedanolide C
and Initial Efforts Towards the C₁₃–C₁₇ Fragment

Zebulon Levine

The marine natural product tedanolide C has some ambiguity associated with its stereochemical assignment. To resolve this ambiguity, we propose to build four macrolide models without the epoxide side chain that could encompass all stereochemical configurations that fit the available data. Spectral comparison between these models and the natural product should allow the elucidation of the true structure of tedanolide C. Herein, we report the completion of both methyl ketone fragments that form the C₁–C₁₂ portion of the macrolide models. The (R) C10 methyl epimer is available in 16.0% overall yield over nine steps from N‑propionyl-(S)-isopropyl thiazolidinethione; the (S) C10 methyl epimer is available in 9.0% unoptimized yield in nine steps from the same precursor. The initial efforts to synthesize the C₁₃–C₁₇ fragment are also reported; modifications to improve the synthetic strategy towards this fragment are discussed in detail.

Mathematical Modeling of Glycolysis Based on the Higgins Model

Ang Li

Glycolysis is a fundamental and ubiquitous metabolic pathway, and glycolytic oscillations have been observed in a range of organisms and cell types both in vivo and in vitro. The Higgins Model, proposed in 1964, describes glycolytic oscillations with two differential equations that focus on the positive allosteric control on the enzyme, phosphofructokinase-1. To incorporate more molecular mechanisms of glycolysis into the model and build a more accurate model that exhibits richer dynamic behaviors, we modify the original Higgins Model by adding the glucose transporters on the cell membrane and the allosteric control of hexokinase into the model. For the modified model including the glucose transporters, we can find oscillations that are similar to those found in the original Higgins Model but without the point at infinity with physical parameter values. This model also exhibits Canard explosion. For the model that further includes the regulation on hexokinase in addition to the glucose transporters, we find a range of dynamic behaviors including simple oscillations, complex oscillations, chaos, bursting oscillations, and mixed mode oscillations. Our work has expanded the repertoire of existing glycolytic models and may suggest new directions for experimental research to discover new modes of glycolytic oscillations.

Isolation and Identification of Bioactive Materials
Present in H2-22 (Sagaraea sp.)

Christina Meade

Over the course of the 2010-2011 academic year, research was conducted in collaboration with Dr. Andria Agusta and the Indonesian Institute for Science in order to explore antibiotic activity in native Indonesian plant extracts, specifically Sageraea sp. (“H2-22”). The two main goals of this thesis were to develop High Performance Liquid Chromatography/Mass Spectrometry (HPLC-MS) analysis methods, for general analytical use, and to isolate and identify the source of antibiotic activity of the aqueous methanol fraction of the crude H2-22 plant extract. Conditions were determined for effective HPLC gradient analysis. Samples were subjected to a solvent gradient of water and acetonitrile in order to elute components that could be detected by the UV-Vis detector (190-350 nm) in the HPLC system. MS analysis was conducted under positive ionization conditions with 0.1% formic acid as a solvent additive in order to protonate the samples and produce ions that could be detected by the MS detection.

The extraction and subsequent fractionation of H2-22 resulted in the isolation of bioactive material at an R_f value of 0.35, I-CMM-IndoAnti-56a and 56b. HPLC-MS analysis did not provide definitive information regarding the identity of these samples. The absence of a UV-Vis chromaphore in the most pure bioactive H2-22 samples produced very weak HPLC peaks, and experimentation with MS analysis was inconclusive, suggesting that the solvent conditions should be developed to convert the neutral components to their corresponding anionic forms. ¹H-NMR analysis indicated that the biologically active components isolated were likely long chain unsaturated fatty acids. The samples isolated were likely the same as those isolated from the hexane layer during the preceeding thesis research by Jacob Kravetz’10, which were tentatively identified as oleic and palmitic acid.

The Free Energy of a Stretched RNA Chain, with Applications to Small RNA-mRNA Binding

Yuzhong Meng

Small RNA-mRNA binding is involved in an important cellular regulatory process called RNA interference (RNAi). Previous calculations of the binding free energy, used in binding site prediction, neglect the cost of mRNA stretching by the rigid small RNA-mRNA duplex. Here, using both polymer physics theory and random-walk simulations, we estimate the free energy of a stretched RNA chain. We then use this estimate to calculate the free energy of small RNA binding to mRNA hairpin loops. Our calculation shows that chain stretching significantly increases the free energy of seed+3’ miRNA binding and siRNA binding to mRNA hairpins. We also make the important observation that 3’ supplementary miRNA has two binding modes: seed-only and seed+3’, and that the former is often more favorable in mRNA hairpins. The stretching effect is insignificant in seed-only miRNA binding and when binding to mRNA bulges, internal loops, or multibranch loops.

Towards Folate-Targeted Polymer-Enzyme Conjugates for Chemotherapy

Colin Platt

The advent and development of chemotherapy has given a new lease on life to thousands of cancer patients, delaying progression or even inducing full remission in many cancer types. However, most chemotherapeutic agents are plagued by severe systemic toxicity due to their untargeted killing of all rapidly-dividing cells, both healthy and neoplastic. In an effort to direct the delivery of these highly toxic agents specifically to neoplastic cells, researchers have taken advantage of the leaky vasculature of solid tumors by use of polymer-conjugated enzymes and prodrugs, which due to their macromolecular size extravasate and accumulate preferentially in tumors. This strategy is known as polymer directed enzyme prodrug therapy (PDEPT).

In this work, we sought to enhance the passive accumulation of current PDEPT conjugates within tumors through the addition of active targeting. As a first step towards the development of a novel, actively-targeted PDEPT system, we synthesized polymers incorporating folic acid. Incorporation of folic acid moieties provides targeting to folic acid receptors known to overexpressed in several cancer types, particularly ovarian carcinomas. These polymers were also fluorescently labeled so that they could be tracked during in vitro cell studies and their receptor-specific targeting assayed.

Our polymers were found to display minimal cytotoxicity, even at concentrations as high as 3 mg/mL, and the results of initial in vitro binding studies are suggestive of folate-receptor-specific targeting. Conditions for the synthesis of our polymers using reversible addition-fragmentation chain transfer (RAFT) polymerizations were also evaluated. A moderately successful RAFT technique was developed which will allow future synthesis of our polymers with high control of molecular weight and introduction of functional end-groups that will allow our polymers to be conjugated to prodrug activating enzymes such as â-lactamas.

Towards the Synthesis of Amino Acid Based Thermoresponsive Polymers

Charles Seipp

The development of new thermoresponsive and self-assembling copolymers has the potential to profoundly affect modern medicine due to their usage in selective targeting of therapeutic agents. To this aim, the amphiphilic diblock copolymer poly((N-acryloyl-valine-N-isopropylamine)-co-poly(acrylic acid)) was synthesized at three different molecular weights (26000, 13000, 6800 Da) with nearly identical block length ratios. While none of the copolymers demonstrated thermoresponsive behavior all three showed self- assembling properties. These copolymers exhibited low critical micelle concentrations of 3.4, 2.6, 2.5 ìM respectively. DLS confirmed the presence of micelles ranging in diameter from 16 to 33 nm. Due to their self-assembling nature, these polymers have promise for drug delivery applications.

The Isolation and Structural Characterization of a Bioactive Compound in Crude Irvingia malayana Extract

Mara Shapero

The Indonesian medicinal plant, Irvingia malayana, possesses bioactive components that have the potential to be of great use in our race against growing antibacterial resistance. Dr. Andria Augusta from the Indonesian Institute of Sciences (LIPI) generously provided our lab with crude Irvingia extract, and from this extract we have worked to isolate and identify the structure of its bioactive components. The bioactive compound we are pursuing and the bioassay-guided fractionation procedure we followed were established by Andrew Yoo in 2009-2010. This year was aimed at isolating more of this bioactive compound, AEY-IndoAnti-94a, in a pure form, and to use a wide variety of analytical instruments to investigate its structure.

From a cumulative total of 34 g of crude extract, we were able to isolate 7.2 mg of this bioactive material, MNS-IndoAnti-90a. 1D- and 2D-NMR spectroscopic analysis was used, in addition to UV-Vis and IR spectroscopy, to further our knowledge of this component’s structure. Because of the enhanced level of purity in our sample relative to Yoo’s, we were able to obtain cleaner, less ambiguous spectra that led us to several more refined structural conclusions. We now conclude that MNS-IndoAnti-90a has 30 carbons, 44-48 hydrogens, and 2-3 oxygens, and includes a terminal methylene group, a carboxylic acid or ester carbonyl, and a functional group containing a carbon-oxygen single bond. Further spectral analysis is necessary before the structure of MNS-IndoAnti-90a can be completely assigned.

The Synthesis and Analysis of Copper(I) Complexes of Tridentate Pyridine-Imine Ligands Used as Catalysts for Atom Transfer Radical Polymerization

Sara Turner

Atom transfer radical polymerization is mediated by a transition metal catalyst, which must provide for an accessible one electron redox couple, readily reversible halogen atom transfer, and full catalyst solubility. Catalyst structure heavily influences the efficiency of ATRP; even subtle changes in catalyst structure can have a large effect on the control of the radical polymerization. Here we investigate copper(I) complexes of a family of closely related tridentate ligands based on a pyridine-imine framework with a third N or O electron donor. We have generated 19 such ligands using a combinatorial building block approach to synthesis, and complexed them to copper. Structural characterizations of the complexes are presented here, and include single crystal XRD, NMR, IR, and ESI-MS analyses. In addition, polymerization data are presented and discussed with respect to catalyst structure. Multiple lines of evidence indicate the facile chelation of one ligand per copper. This chelation causes a change in the bonding of that ligand, as evidenced by NMR and IR analyses. In the catalysts studied here, copper(I) prefers a pseudotetrahedral coordination environment in the solid state. Three of the structurally characterized complexes are dinuclear, and one is mononuclear. A catalyst’s preference for a dinuclear structure may be caused by the strain associated with two chelate rings to the same copper atom.

Computer Science

Support Vector Machines: Efficient Training Algorithms and Applications to Feature Selection

Nicholas A. Arnosti’11

This analyzed several feature selection algorithms that score features according to their impact on estimated class probabilities. In it, we presented a theoretical approach that clarifies the quantities estimated by each scoring system and allowed us to make several predictions about their performance. We validated our predictions with tests on synthetic data.

Developing Mobile Applications with Shims

Yuxing Danny Huang’11

By preserving the programming model of the operating system’s networking interface, we developed a framework through which applications can be adapted for mobile networks without changing the source code, in a way that is automatic, dynamic and flexible.

Visualizing Large Communication Graphs

Steven S. Rubin’11

Communication graphs from parallel programs, which are useful for debugging, can be difficult to visualize. We first showed methods of creating and describing the symmetric structure of these graphs using graph grammars and procedural methods. Finally, we presented several techniques for visualizing these graphs, included a generalized algorithm for finding contextual subgraphs around important nodes.

Minimum Spanning Tree Algorithms in MapReduce

Erdem Sahin’11

This thesis investigates MST algorithms for the MapReduce framework of distributed computation. We consider MST algorithms designed for MapReduce, CRCW PRAM, and RAM models of computation. We implement each algorithm on the Hadoop MapReduce platform, and analyze our implementation’s run-time and space complexity with respect to MapReduce. We run our algorithms on massive graphs consisting of millions of nodes and billions of edges. The performance of our algorithms follow theory, but all except our adaptation of Boruvka’s MST algorithm fail to process very large graph instances due to unrealistic reducer memory assumptions. We conclude that the streaming model used by our adaptation of Boruvka’s is preferable for MST computations on the MapReduce framework, as it makes no assumptions about what can be held in memory, harnesses the power of the distributed framework well on massive graphs, and features desirable scalability.

Geosciences

Examining Knickpints in Middle Boulder Creek, Colorado

Evan N. Dethier

The apparent stasis of our current landscape belies the constant change it has undergone for millions of years before the present. The earth’s surface continues to transform as denudation balances slow uplift of rock material removed from slopes and by channels, which transmit climatic and tectonic signals through the landscape. In postorogenic terrains such as the Colorado Front Range, knickpoints—steep channels bounded by relatively shallower reaches—may reflect the influence of rock strength or slow, complex response to external forcing.

Most knickpoint research has focused on regions with high uplift rates, weak rock, and rapid landscape evolution. In contrast, the Front Range in Colorado—in the interior of the North American continent—is a region with low rock uplift and low precipitation, relatively strong rock, and thus comparatively low rates of incision. Knickpoints are found in all channel orders in the Middle Boulder Creek catchment, likely the result of increased precipitation and channel mouth lowering initiated between 7 and 3 Ma.

This study focused on the Middle Boulder Creek catchment west of Boulder, CO. Middle Boulder Creek has a drainage area of roughly 350 km², an area that includes the other two study areas: Gordon Gulch (~4 km²) and Betasso Gulch (~0.45 km²). In the field, we surveyed channel profiles and recorded width, depth, and grain size measurements. We described the character of the hillslopes, surveying transects perpendicular to the channel and measuring rock strength using a Schmidt hammer. We supplemented these field data with LIDAR DEM observations with a pixel size of 1 m². We calculated relationships between spatial data, comparing downstream distance, channel slope, mean hillslope angle, rock strength, and stream power. In these channels, knickpoints are roughly twice as steep as the mean channel slope, producing local convexity in channel profiles that would be concave-up in steady state. In Middle Boulder Creek, the knickpoint slopes average 7.4%, higher than the 4.2% channel average. A second, minor knickpoint is preserved upstream in a reach where average stream slope is only 4%. Rock type is uniform along Middle Boulder Creek, but rock strength is highest in the middle of the lower knickpoint and at the second knickpoint. In those zones, Schmidt Hammer values range between 50 and 65, considerably higher than the values of 35-45 that characterize the rest of the basin. Gordon Gulch has two knickpoints, with mean slopes of 15.3%, compared to the 10.8% channel average. These knickpoints are located in Silver Plume Granite that is harder than the metasedimentary basement rock that underlies the rest of the basin. Schmidt hammer values average between 45-50 in the knickpoints, compared to values between 0 and 40 upstream. Betasso has three small knickpoints that do not dramatically disrupt the channel concavity, but instead appear as bedrock steps in the channel. These steps correspond to areas of hard rock in the channel and on the adjacent hillslopes and have Schmidt hammer values of >45. Schmidt hammer values elsewhere in the catchment are between 5 and 40, and are <5 above the third knickpoint, where a thick layer of colluvium and saprolite covers the bedrock.

In each catchment, knickpoints mark the boundary between steady state and adjusting landscapes. Below and perpendicular to knickpoints, hillslopes are steep and rough relative to smoother and flatter hillslopes above. Steepening is a result of knickpoint-driven baselevel lowering. Where the hillslopes are steepened past the angle of repose, mass-wasting events hinder the adjustment process, overwhelming the ability of the channel to transport material downstream. Less dramatic baselevel lowering can accelerate hillslope processes, but rock strength, stream power, and continued channel disruption inhibit this adjustment and prevent a return to steady-state processes.

Bedrock strength plays a major role in knickpoint dynamics. In the Middle Boulder Creek and Betasso Gulch catchments, basins without local change in lithology nonetheless contain knickpoints that mark the transition between resistant and nonresistant bedrock. The strongest rock is at the knickpoint, and the lower margin of weaker rock lies immediately above. This localized hardness likely results from the removal of weak rock by highly erosive flows within the knickpoint, exposing the fresh unweathered rock beneath. Weakness above the knickpoint—where stream power is low—is produced by the absence of effective removal and transport in those reaches. The stasis of this untransported rock makes it susceptible to weathering processes, and low rock strength could be the result of longer exposure. Where bedrock in a catchment is non-uniform, knickpoints form in the zones of higher rock strength. In Gordon Gulch, incision has been less effective in resistant rock, and the resulting disparity in baselevel lowering has localized the steepest reaches at the contact between less- and more-resistant rock.

Knickpoint migration in post-orogenic landscapes is difficult. Incision is not accelerated by rapid uplift, and must be driven by other forcing. The presence of resistant rock, low stream power, and the aggradation of sediment in the channel further hinder incision in the Front Range. Evidence for continued knickpoint migration is absent from this landscape and tentative evidence shows that the knickpoints may have stalled in each catchment. We must conclude that incision in the Middle Boulder Creek requires a synergy of many erosive processes in order to effectively incise the channels. This synergy is currently missing several key components, and knickpoint migration is likely to remain stalled until conditions change in the Front Range. Various climatic forcings or the renewal of tectonic action in the Rocky Mountains could reinitiate knickpoint migration in the Middle Boulder Creek catchment.

Reconstructing Pinedale Ice in the Green Lakes Valley and Adjacent Areas, Colorado

Keith M. Kantack

During the Pinedale Glaciation (~32-15 kya) alpine glaciers filled cirques and valleys extending from the crest of the Colorado Front Range. These glaciers are small remnants today, but their erosion over time dominates the landscape. Using field evidence, modeling techniques based on glacial flow rules, and Lidar imagery, I determined glacier size and local ice-flow direction and estimated volumes of sediment deposited during late Pinedale time. The volume estimates help constrain how rapidly those glaciers cut their cirques and valleys.

At glacial maximum, ice flowing out of the Green Lakes and Arapaho valleys joined and covered 22.3 km², but was only 83 m thick on average. Between 21 and about 15 kya, the glacier deposited morainal debris with a volume between 67 and 90 x10⁶ m³, which equates to a bed-lowering rate between 0.5 and 1.35 mm yr^-1. Moraine volumes suggest that cutting the Arapaho and Green Lakes Valleys required as few as 11 glacial events of size similar to the late Pinedale. Results from the Horseshoe and Rainbow Cirques to the south suggest respective bed-lowering rates of 0.25 to 2.98 mm yr^-1 and 0.15 to 1.37 mm yr^-1. Calculated erosion rates are similar to those measured for glaciers in the Swiss Alps. Late Pinedale cirque glaciers in the Front Range eroded at a moderate rate – far greater than polar ice sheets (0.01 mm yr^-1), but well below rates of glaciers in southeastern Alaska that lower their beds on the order of 1to 10 cm yr^-1.

Petrogenesis of Precambrian Igneous and Metaigneous Rocks South of the Madison Mylonite Zone, Henrys Lake Mountains, SW Montana and Idaho

Caleb O. Lucy

In their geologic map of the Hebgen Lake Quadrangle, O’Neill and Christiansen (2004) show units of Middle Archean tonalite and diorite gneiss along with Late Archean amphibolite in the middle of the Henrys Lake Mountains, SW Montana and Idaho. These bodies appear to be plutonic components along the western edge of the Wyoming province and broadly trend southwest to northeast, but they have not been studied in detail. After ascertaining the nature of the protoliths of these rocks, my project seeks to determine their metamorphic history, establish their original tectonic setting, and relate their history to that of southwest Montana, the northwestern Wyoming province and potentially the Big Sky orogeny.

Upon close inspection, the igneous rocks found in the Henrys Lake Mountains mapped by O’Neill and Christiansen as tonalite and diorite are petrographically and chemically indistinguishable. Geochemical analysis confirms their original mafic to intermediate igneous character; field relationships and petrography suggest an intrusive origin, although petrography also indicates that since crystallization, the rocks have experienced extensive hydrothermal alteration or low greenschist facies metamorphism. Because of their similarity and subsequent history, the “tonalite” and “diorite” of the Henrys Lake Mountains should be considered one unit, ranging in composition from metagabbro to metadiorite. The amphibolite is more mafic, with a protolith composition of subalkaline basalt.

Geochemical evidence suggests the metagabbro, metadiorite and amphibolite are associated with arc volcanism and may have been part of a continental margin arc or a mature island arc that docked on the western edge of the Wyoming province. A cryptic, low greenschist facies metamorphic or extensive hydrothermal alteration event affected the rocks after their crystallization. The crystallization age for the metaigneous rocks of the Henrys Lake Mountains remains undecided.

Assessing Eolian Contributions to Soils in the Boulder Creek Catchment, Colorado

James A. McCarthy

In high-relief environments, soil geochemistry and morphology reflect the weathering of both parent material and materials added to weathering profiles by downslope transport and dustfall. In the Colorado Front Range, transport is limited in alpine soils developed on stable surfaces. In the montane zone, soils and regolith on hillslopes are mobile and mix during downslope transport. In this study, measurement of soil texture, citrate-buffered dithionite-extractable iron (Fe_d), and bulk geochemistry permitted evaluation of weathering, downslope transport, and eolian deposition in the Critical Zone of the Boulder Creek catchment. The accumulated mass of Fe_d and clay are positively correlated with deposit age in the catchment. Stable alpine soils form from Pinedale age till and the cool, moist climate generates sufficient acidity to develop strong horizonation. Soil morphology in the upper montane Gordon Gulch is controlled mainly by downslope transport; soils on the north-facing slope thicken downslope and have complex morphology. Soils on the south-facing slope are thin and overlie saprolite at a maximum depth of 53 cm. Soils on the north-facing slope have higher clay and Fe_d contents than those that face south, indicating more rapid erosion of the south-facing slope and greater weathering on the north-facing slope. On older surfaces of low relief, soils with thick Bt horizons develop from deeply weathered saprolite and regolith, and locally contain buried sequences and features that suggest periglacial mass movement and slope instability in pre-Holocene time.

In soils throughout the catchment, enrichment of fine particles and low concentrations of Fe_d in surface horizons suggests eolian sedimentation. Enrichment of fines is most apparent at stable sites, but soils on lower positions on slopes are also enriched. The amount of clay and fine silt produced in situ and the amount added from dustfall is poorly constrained, but the dustfall rate is less than 60 g cm^-2 100 kyr^-1. Immobile element geochemistry indicates that surface enrichment in high field strength elements (HFSE) is not uniform throughout the study area, suggesting that the dust deposition may vary spatially and/or temporally. Ratios of the immobile elements Ti, Zr, and Nb, suggest that surface fines are geochemically distinct from the dominant parent materials in the catchment; however, elemental ratios may also reflect increased fines in the surface horizons released by weathering, because Zr and Nb are preferentially enriched in fine fractions of parent rocks in the study area. The composition of fine sediments in surface horizons are similar to silt mantles in other basins in the Indian Peaks Wilderness Area and to silt-sized alluvium in North Park and Middle Park to the west of the study area. Surface enrichment of fines with low Fe_d and distinct immobile element ratios of low compositional variability suggest that a substantial portion of the fine fraction of soils examined in this study originate as dustfall, potentially derived from North Park and Middle Park.

Pre-Tectonic Fabric in Unstrained Conglomerates and Implications for the Rf-phi Strain Analysis of Their Deformed Counterparts

Lisa M. Merkhofer

The Rf-φ method of strain analysis is based on the assumption that sedimentary rocks will have initially random fabrics before the onset of strain (Ramsay, 1967). In random fabrics, clasts are unaligned, such that their long axes are uniformly distributed across all orientations. Alternatively, the clasts of aligned fabrics are non-random, with a preferred orientation of clast long axes.

While assuming an initially random state is necessary to quantify strain, clasts of sedimentary rocks may be aligned and frequently such alignments are strong. It is therefore difficult to disentangle tectonic alignment from that of prior depositional or diagenetic alignment. When such pre-tectonic alignment exists and is not recognized in a strained rock, Rf-φ analysis can result in significant strain errors (Patterson & Yu, 1994).

The large size of conglomerate clasts makes them common subjects for strain analysis. However, although the effects of pre-tectonic fabric on strain have been studied for many sedimentary rocks, including sandstone, tuff, and oolitic limestone, little work has been done for conglomerate. In this study, I conduct analysis by digitizing clasts from unstrained alluvial-fan conglomerates using high-resolution GigaPan photography. From the analysis of both Cartesian and polar plots, I characterize how strongly conglomerate initial fabrics deviate from random fabrics. Secondly, by simulating strain on undeformed conglomerate, I calculate the error aligned initial fabrics impose on strain estimates and assess prospects for detecting the traces of initial fabric after strain.

The conglomerates studied in Deerfield Basin, Massachusetts, show random fabrics on bedding-parallel faces and non-random fabrics on bedding perpendicular faces. Non-random fabrics are aligned consistently imbricate to bedding trace. The magnitude of clast alignment in non-random fabrics can be described by an average apparent strain aspect ratio of 2.0 and consistency ratios above 0.40, indicating strongly preferred fabrics. Three-dimensional fabric is described by an oblate apparent strain ellipsoid with maximum dimensions of 2.5:1.9:1.0, far greater than those found in other sedimentary rocks (Holst, 1982). Within fabrics, quartz vein or quartzite clasts are the least likely to be preferred.

Strain simulations performed on conglomerate fabrics assuming a random fabric demonstrate that strain magnitudes can be overestimated by 50%, when stretching is parallel to the initial preferred orientation. Errors of this scale occur even at small strains of an Rs=1.5.

Basement Faults of the Edelman Lineament, Bighorn Mountains, Wyoming

David O. Oakley

The Bighorn Mountains are a basement-cored arch in north-central Wyoming formed during the Laramide orogeny (70-45 Ma). The range and orogeny are exemplary of mid-continent deformation distant from a plate boundary, but many unanswered questions remain about the structures involved. The Archaean crystalline basement rocks at the core of the Bighorn arch and the series of topographic lineaments of likely structural significance within them (Hoppin, 1974) are particularly understudied parts of the range. I conducted stereographic analysis of faults and fractures, calculated paleostresses, and did petrographic study of fault materials from within the Edelman Creek lineament, a narrow valley trending approximately 040°. My goal was to determine whether the structures were active during Laramide arch formation, to assess the significance of fault mineralization, and to determine the structural role of the lineament.

My results suggest that the Archean basement rocks of the Bighorn Mountains record a long history of deformation. Field observations and paleostress analysis indicate some Laramide deformation, but the majority of faults appear to be non-Laramide. Many faults have been reactivated under changed stress conditions subsequent to their formation. Chlorite and epidote fault mineralization correlate roughly with fault orientation and more closely with different microscopic characteristics, indicating different conditions and thus ages of formation. Epidote mineralization appears to be pre-Laramide and to predate chlorite, but otherwise the timing of mineralization and fault movement remains mostly uncertain. Evidence suggests that a regional scale strike-slip fault underlies the Edelman lineament. Absolute, and in many cases relative, ages remain as crucial unknowns and a potential topic for future work.

Paleogeography of a Devonian Island-Reef Complex, Canning Basin, Western Australia

Daniel R. Walsh

The Devonian island-reef complex of the Canning Basin in Western Australia provides a unique opportunity for paleoecologists and geomorphologists to explore an intact biophysical system preserved in three dimensions. The burial of the currently exposed Devonian reefs by 2,500 m of carbonates after the Frasnian-Famminian boundary and subsequent geological quiescence allowed structures such as skerrys, islands, lagoons, and clastic deposits, as well as the associated invertebrate faunas, to be preserved in detail despite the 375 million years separating us from the living reef. This study aims to reconstruct the paleogeography of the island reef system in a holistic sense.

The island-reef complex consists of two parts: the highly folded quartzite-phyllite Oscar Range, which formed a chain of islands in Devonian times, and the stromatoporoid reef structure, which includes a classic reef-front to reef-flat sequence as well as an inner and outer lagoon, which are separated by the Oscar Range. Fieldwork focused on the far eastern end of the range, where the Tunnel Creek Road provides access to a wide variety of facies including siliciclastic deposits sourced from the Oscar Range. Studies in three different phases of the various aspects of the island-reef system provide insights into the paleoecology, geomorphology, and stratigraphic history of the region.

The 2005 discovery of a fan-delta deposit composed of siliciclastics apparently sourced from the Oscar Range (Johnson and Webb, 2007) spurred a return trip to the region to map in the deposit and surrounding bio-facies. The first stage of this study involved the creation of a detailed map and stratigraphic section for the fan-delta and overlying carbonates, which are located in the distal outer lagoon. The stratigraphy reveals a sequence of episodic siliciclastic sedimentation events with provenance in the Oscar Range that decrease in magnitude with time. A transition to active reef growth indicated by a massive stromatoporoid fauna and the stabilization of oncoids is diagnostic of a rise in sea level near the close of Frasnian time.

The second stage of the study focuses on the paleogeographic relationship between the fan-delta deposit and the paleoislands of the Oscar Range. This relationship suggests that the stream channels that now converge and pass immediately alongside the fossil fan-delta are inherited from those which originally delivered sediments from the Oscar Range to the fan-delta in Frasnian times. Quantitative assessment of this possibility reveals that differential erosion rates between quartzite and limestone, derived from studies of landscape denudation in arid Australia, would not result in enough erosion within the Oscar Range to override structural controls on stream erosion. Channel quantification shows that stream flow is controlled almost entirely by folds in Proterozoic quartzite and phyllite, therefore demonstrating that modern stream channels are likely to be very similar in location to those active during the Devonian. Study of the ancient landforms reveals that geomorphological techniques can be extended into the past under very specific circumstances.

The third stage of the project focuses on a small paleoisland located in the inner lagoon just north of the Oscar Range. Detailed reconstruction of an invertebrate community rich in gastropods, stromatoporoids, and bivalves is possible due to incredible in situ preservation. Determination of species zones and regions of silisiclastic deposition allows for reconstruction of paleoenvironmental conditions such as wind and water circulation. Analogs in modern reefs show that species with similar morphologies have filled specific niches since Devonian times. A diverse gastropod fauna in the lagoon near the island allows for a similar analysis in a different setting.

The spectacularly preserved biophysical systems of the island-reef tract allow for reconstruction of an ancient geography with an incredible level of detail. This location is therefore ideal for the creation of a geopark – a conservation framework that would ensure protection of this geologic resource and education of the public about its value. The last phase of the project was to make recommendations as to how a geopark could be implemented in the field area from a practical and educational standpoint.

The division of paleogeographic reconstruction into several small-scale projects allows for a well-constrained and detail-oriented interpretation. Much fruitful research remains to be done in the eastern Oscar Range alone; the analysis contained herein provides a firm basis for such work.

Mathematics and Statistics

A Study of Hitting Times for Random Walks on Finite, Undirected Graphs

Ariel Joseph Binder

This thesis applies algebraic graph theory to random walks. Using the concept of a graph’s fundamental matrix and the method of spectral decomposition, we derive a formula that calculates expected hitting times for discrete-time random walks on finite, undirected, strongly connected graphs. We arrive at this formula independently of existing literature, and do so in a clearer and more explicit manner than previous works. Additionally we apply primitive roots of unity to the calculation of expected hitting times for random walks on circulant graphs. The thesis ends by discussing the difficulty of generalizing these results to higher moments of hitting time distributions, and using a different approach that makes use of the Catalan numbers to investigate hitting time probabilities for random walks on the integer number line.

n-Level Densities of the Low-Lying Zeroes of Quadratic Dirichlet L-Functions

Jake Levinson

The statistical distributions of zeros of L-functions can be used to study prime numbers, elliptic curves and even the ideal class groups of number fields. L-functions have been studied in connection with random matrix theory, which provides easier methods of computing these distributions. One statistic, the n-level density of low-lying zeros for a family of L-functions, measures the distribution of zeros near the central point s = 1/2. The Density Conjecture of Katz and Sarnak states that the n-level density for an L-function family depends on a classical compact group associated to the family. We extend previous work by Gao on the n-level densities of quadratic Dirichlet L-functions. Our main result is to confirm up to n = 6 that, for test functions of suitable support, the density is as predicted by random matrix theory. We also consider a (conjectural) combinatorial identity for certain Fourier transforms of the test functions which, if true, would help in extending the result to all n.

Chains of Rings with Local Formal Fibers

Sean Carlos Pegado

Let $R$ be a local (Noetherian) commutative ring with unity. If $R$ is complete, its structure is understood; however, less is known if $R$ is not complete, and thus the relationship between a ring and its completion is a subject of current research. To this end, previous work has begun to investigate the relationship between prime ideals of a ring and the prime ideals of its completion. We generalize these results to chains of rings that share the same completion.

Optimal Control of the Generalized Moving Point Mass Dynamic

Thuy Vinh Pham

We study the generalized time-optimal control problem where the underlying dynamic is a moving point mass under Newtonian mechanics with acceleration and velocity constraints. The optimal control of this control problem coincides with the viscosity solution of a specific partial differential equation of Hamilton-Jacobi type. Using the dynamic programming approach, we derive the associated Hamilton-Jacobi-Bellman equation and obtain its numerical solution with a semi-Lagrangian discretization scheme.

Robust Regression Boosting

Ville Satopaa

In 2010 Long and Servedio suggested that boosting algorithms that are based on convex loss functions are flawed in a sense that they cannot tolerate outliers. Inspired by Long and Servedio’s observation, this undergraduate thesis introduces a novel regression boosting algorithm that is based on a non-convex loss function. First, several properties of this algorithm are stated and proven. Second, experimental evidence showing that this algorithm is highly robust in the presence of outliers is given.

Generic Formal Fibers

Philip Vu

Let T be a complete local ring. We present necessary and sufficient conditions for which there exists a local integral domain A, a subring of T, whose completion is T with a generic formal fiber that has countably many maximal elements. We also present results on the elements we can adjoin to this integral domain A.

Geometric Degree of 2-Bridge Knots

Jacob Wagner

In 1987, Kuiper introduced geometric degree alongside superbridge index, but degree has been studied far less than superbridge. In this thesis, we calculate degree for all 2-bridge, 4-superbridge knots. Then, we modify the definitions of degree and a related invariant, thin position, to generate new measures.

Sturm-Liouville Oscillation Theory for Differential Equations and Applications to Functional Analysis

Zhaoning Wang

We study the connection between second-order differential equations and their corresponding difference equations. With this connection in mind, we investigate quantitative and qualitative properties of the zeros of the solutions of differential/difference equations and of the eigenvalues of the associated Jacobi matrices. In particular, we study various applications of the Sturm-Liouville Oscillation Theory to differential equations and spectral theory.

The Limiting Spectral Measure for the Ensemble of Generalized Real Symmetric Block m-Circulant Matrices

Wentao Xiong

Given an ensemble of N x N random matrices with independent entries chosen from a nice probability distribution, a natural question is whether the empirical spectral measures of typical matrices converge to some limiting measure as N tends to infinity. It has been shown that the limiting spectral distribution for the ensemble of real symmetric matrices is a semi-circle, and that the distribution for real symmetric circulant matrices is a Gaussian. As a transition from the general real symmetric matrices to the highly structured circulant matrices, the ensemble of block m-circulant matrices with toroidal diagonals of period m exhibits an eigenvalue density as the product of a Gaussian and a certain even polynomial of degree 2m-2. This paper generalizes the m-circulant pattern and shows that the limiting spectral distribution is determined by the pattern of the i.i.d.r.v. elements within an m-period, depending on not only the frequency at which each element appears, but also the way the elements are arranged. For an arbitrary pattern, the empirical spectral measures converge to some nice probability distribution as N tends to infinity.

Physics

The Ubit: An Exploration of Real-Vector-Space Quantum Mechanics

Antoniya Aleksandrova

In 1960, Stueckelberg showed that if the real-vector-space version of quantum theory is supplemented by a specific “super selection rule” limiting the set of allowed observables, then the theory is mathematically equivalent to ordinary quantum mechanics. We give an information-theoretic interpretation of this rule by proposing a model in which all real quantum objects can interact with a universal binary object called the ubit. In addition, we use our model to investigate whether the rule can be achieved through a generic dynamics rather than being assumed.

Experimental Characterization of Algorithms for Holographic Optical Trapping

Peter K. Gottlieb

Tightly focused light can be used to manipulate objects on a very small scale. My thesis looks at the merits and difficulties of using a phase-modulated laser beam to simultaneously manipulate an arbitrary number of objects.

Observation of Dynamic Processes in Weakly Segregated Diblock Copolymer Films

Leah L. Hurwich

This thesis studied the dynamic processes of weakly segregated diblock copolymer films to better understand the behavior of smectic systems. Thin films samples of polystyrene-block-polymethylmethacrylate (PS-b-PMMA) were cast atop silicon substrates and imaged under an atomic force microscope (AFM) while heating to determine the annihilation processes prevalent in the system. Preliminary findings show these annihilations differ greatly from those of strongly segregated diblock copolymers, thus suggesting not all smectic systems evolve through similar processes.