Eclipsing binaries are systems in which two stars orbit and pass in front of each other. They are important astrophysical tools we utilize to measure fundamental properties of stars, such as their masses, radii, and temperatures. Here, we analyzed photometric data acquired from the Kepler satellite in order to model the light as a function of time of specific eclipsing binary systems. We selected these targets from the Villanova Kepler Eclipsing Binary Catalog. For each system, we used a least-squares optimization method to determine the best fit solution between the model and the observed light curve. Our light curve model has a total of 13 parameters. They include stellar elements such as sum of radii and flux ratio, as well as orbital elements such as period and eccentricity. In addition to providing orbital solutions, our models can be used to search for the presence of additional bodies in the system, such as a third star or a circumbinary planet.

When a star undergoes the transition to the white dwarf stage of the stellar life cycle, an outburst of ionized gas is expelled in a planetary nebula. It is currently unknown if planets orbiting a white dwarf are able to survive this transition. However, between one fourth and one half of white dwarfs have elements in their atmospheres that are heavier than helium and will sink into the white dwarf quickly unless they are somehow being replenished. The recent detection of a disintegrating rocky planet orbiting a polluted white dwarf suggests that the source of metal pollution may be debris from these bodies. As the Sun will become a white dwarf in the distant future, our team began the search for an intact planet, which could provide a model for examining the future of our own planet and solar system. Using observations from the ARCSAT 0.5 m telescope at Apache Point Observatory, we examined fluctuations in brightness of a small sample of white dwarfs, over a period of several nights. We visually inspected the white dwarf light curves for strong dips in brightness that could indicate the transit of a planet, and promising candidates were further analyzed via Lomb-Scargle Periodograms. Most of the white dwarfs displayed periodicities in flux, but none were indicative of the presence of a planetary body. Included in our data was an eclipsing binary white dwarf system with a known periodicity that our function could not detect correctly. We have improved our search by using the Box-Least Squares method to detect weak transits instead of periodograms, but have yet to detect an intact planet orbiting a white dwarf.

Transformations over assembly code are common in many compilers. Such transformations are useful in applications ranging from peephole optimizations to software fault isolation. These transformations are also some of the most bug-dense compiler components. Such bugs could be eliminated by formally verifying the compiler, but state of-the-art formally verified compilers like CompCert do not support assembly-level program transformations. This paper presents Peek, a framework for expressing, verifying, and running assembly-level program transformations in CompCert. Peek contributes four new components: a lower level semantics for CompCert x86 syntax, a liveness analysis, a library for expressing and verifying peephole optimizations, and a verified peephole optimization pass built into CompCert. Each of these is accompanied by a correctness proof in Coq against realistic assumptions about the calling convention and the system memory allocator. Verifying peephole optimizations in Peek requires proving only a set of local properties, which we have proved are sufficient to ensure global transformation correctness. We have proven these local properties for 26 peephole transformations from the literature. We discuss the development of our new assembly semantics, our liveness analysis, representation of program transformations, and execution engine, describe the verification challenges of each component, and detail techniques we applied to mitigate the proof burden.

One third of adults from ages 20 to 44 have untreated dental caries, typically treated through irreversible fillings and root canals (Centers for Disease Control and Prevention, 2012). Current technologies require professional consultation for caries detection – an inconvenient and costly process. Our objective is creating an optical system of detecting early caries formation to be performed at home on a weekly basis via spectral imaging of the plaque on teeth. The bacterial growth in plaque and demineralization of underlying enamel is detected by optically monitoring the fluorescent byproduct of bacterial metabolism, porphyrin. By imaging teeth in blue reflectance light and red fluorescence signal, concentration and growth of plaque deposits are measured using target versus background (T/B) pixel peak values (PPV) and T/B pixel count (PC) of PPV. Repeated measurements are plotted for plaque deposits and trends. Statistical analysis and comparison of T/B ratios and PCs trending can identify the most efficient measurement for locating plaque deposits most susceptible to caries development, prompting more efficient preventative care and triggering a dental examination. This new concept was simulated in vitro using a dental training model of teeth, measured applications of red stain, image stitching software (123D, Autodesk), image analysis software (ImageJ), and a mobile phone. Red stain was applied in increasing concentration and volume to both occlusal and interproximal crevasses that would most likely be missed by brushing and flossing. At each time period, a series of 2D images were acquired to reconstruct 3D models of simulated bacterial growth and were processed using ImageJ for T/B ratios of PPVs and PCs, which were plotted against each time period. T/B ratios of PPVs produced the most consistent measure of trending of the simulated growth, R^2 values 0.77, with statistically significant differences in variance with PC ratios (f-test).

During restoration of strip mines, a one meter soil cap is added to prevent metal contamination from buried coal spoil. Research has shown that recolonization by invasive grasses and soil compaction often prevents the return of native forests. To encourage recovery of forests on reclaimed mine sites, hybrids of American and Asian chestnut species (Castanea dentata × C. mollissima) are being used to accelerate succession on reclamation sites. In this study, we are testing whether soil treatments that remediate compaction and remove invasive grasses facilitate the movement of metals from buried spoil to the soil and/or tree foliage or flowers. Soil treatments included 1) plow disking, 2) deep soil ripping, 3) combination of plow disking plus deep-ripping, and 4) an undisturbed control. Metal concentrations were measured from soil, foliage, and flowers of chestnut trees by Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Our results indicate no significant differences in metal uptake in chestnut trees due to soil treatment after 8 growing seasons; however, the treatments that involved deep soil ripping resulted in a trend of elevated metal concentrations in foliar tissue for lead, copper, zinc, selenium and manganese. As these trees mature, these trends warrant further investigation with regard to the effect of deep soil ripping on metal uptake.

Due to anthropogenic pollution, toxic heavy metals such as arsenic, silver, cadmium, copper, lead, zinc and selenium have been deposited into the environment along the west coast of North America. Fungi have been shown to uptake or accumulate heavy metals at concentrations that can exceed the environmental background levels. Foraging for wild edible mushrooms has been gaining popularity in recent years, but many edible mushrooms can take up toxic heavy metals, causing concern for public health. Previous work from our lab has shown that heavy metals uptake varies dramatically in our region, but we are now expanding the geographic range of the study. Samples of numerous species of edible fungi, as well as the surrounding soil and duff were collected along the west coast from California to northern Vancouver Island by researchers at Oregon State University. Metals analyses are currently in progress: samples are digested in Nitric acid, diluted and will be analyzed for heavy metal concentration using inductively coupled plasma - mass spectrometry. We predict that different fungal species will preferentially uptake different metals, and therefore some species of fungi will concentrate certain metals within their tissue at greater levels than others. The results of this study will be used to determine if certain fungal species that are now considered edible may be a health risk due to heavy metal concentrations that exceed the EPA recommended annual intake levels. Certain locations that have recieved excess pollution may also be pinpointed and the results of this study will be reported to mycological societies to educate the public about unsafe species and locations to pick mushrooms. Additionally, if any of these fungi can accumulate heavy metals at levels that greatly exceed levels present in soil or duff, we can assert that these species have potential for use in bioremediation of polluted lands. 

This project seeks to identify root colonization of ectomycorrhizal fungi (EMF) on Abies grandis (grand fir) seedlings that were planted at the former Lake Mills restoration site along the Elwha river. Ectomycorrhizal fungi are a root-dwelling fungus that form mutualistic symbioses with many tree species and fulfill an important role in nutrient cycling within forest ecosystems. EMF species from early successional river bed sites may contain more suitable types of EMF compared to EMF that inhabit late successional mature forest soil. We hypothesize that the two soil inoculant types will produce a difference in EMF root seedling colonization and potentially have an impact on the survival rate of conifer seedlings at the restoration site. In fall of 2014, A. grandis seedlings were planted using two treatments of soil inoculant. Mature forest soil (late successional), autoclaved forest soil (control), river bed soil (early successional), or autoclaved river bed soil (control) was mixed with the excavated soil and then used to replant each seedling at the field site. Laboratory extractions of 260 A. grandis root tip samples have been completed and we are currently focused on completing PCR and sequence analysis on the extracted root tips. My laboratory work includes DNA extraction, gel electrophoresis, PCR using ITS1F and ITS4 primers, and sequence analysis of fungal DNA using the Basic Local Alignment Search Tool (BLAST). The ectomycorrhizal taxa present will be identified using BLAST and then percent colonization and taxa diversity will be determined within each treatment. This genetic information will contribute to the greater body of knowledge on the diversity and range of ectomycorrhizal fungi in the Pacific Northwest and be potentially useful in restoration ecology work as we gain a better understanding of the role of ectomycorrhizal fungi on conifer seedling survival within the context of early successional conditions.

During pregnancy, women and the developing fetus face the risk of invasive bacterial infection, which can lead to adverse birth outcomes such as preterm birth or fetal death. Inflammatory cytokines play a key role during infection-related preterm birth. Previous studies have shown that lipopolysaccharide (LPS), a gram-negative cell wall component, causes increases in inflammatory cytokines and preterm birth in mice. However, limitations exist when comparing the pregnant mouse model to humans. Recently, we conducted a non-human primate (NHP) experiment to identify putative targets of inflammation caused by LPS. However, contrary to expectation, no signs of preterm labor were observed. To determine if the LPS batch used in the NHP model induced inflammation in immune cells, I used an in vitro human macrophage cell culture model.  A human immortalized monocyte cell line, THP-1 cells, was differentiated to macrophages using phorbol myristate acetate (PMA) and treated with a range of concentrations of LPS, similar to those used for the NHP experiment. Cytokines in the medium were quantified by the Enzyme-Linked Immunosorbent Assay (ELISA). LPS increased IL-1B in a dose-dependent manner. Additional inflammatory cytokines associated with preterm birth such as, IL-6, IL-8, and TNF-a, will also be measured. The results will be compared to the ELISA results of the same cytokines released in the NHP samples. This work will guide insight into addressing the inflammatory response associated with preterm birth.

Cholangiocarcinoma, better known as bile duct cancer, is a rare and aggressive form of cancer in the biliary system. Bile duct cancer has the third lowest 5-year survival rate among all forms of cancer in the US. Due to its location, access to the biliary system for detection or diagnosis is difficult without invasive methods. Early detection of bile duct cancer is crucial and could be achieved by using the multimodal scanning fiber endoscope (SFE) developed in our lab. The SFE is able to noninvasively enter the biliary tree through the digestive system and present realistic information about the potential malignancies or disease state. Typically, a phantom is used to mimic the in-vivo presentation of disease state in order to validate the use of a new medical device. To prepare the SFE for this kind of cancer detection, a bile duct phantom with three dimensional (3D) cell culture was developed to test the sensitivity and specificity of the real-time in-vivo imaging. A biocompatible and optically clear polydimethylsiloxane (PDMS) device was first made for the cast of the hydrogel phantom. The device was utilized for structural stability of the phantom like a cast. The phantom was constructed by folding a flat sheet of hydrogel into a cylindrical tube with a lumen that reflects the true diameter of bile duct. The cancer cells were suspended at specific locations in order for spatial validation of SFE. The end result was the development of a 3D cell culture phantom with fluorescent probes targeted to cancer biomarkers. The phantom was imaged with SFE to evaluate its ability to accurately locate malignancies, distinguishing cancer cells from normal cells. Future work will extend to validation of fluorescent probes and ultimately lead to development of a new cancer imaging method with SFE to reduce late diagnosis.

Pancreatic cancer is ranked fourth amongst the most lethal cancers in US and first in mortality due to its low five-year survival rate. Accurate, definitive diagnoses are determined by pathologists who visually inspect and evaluate tissue slices from patient-procured biopsies traditionally processed into 2D slides. 3D microscopy makes 3D pathology, consisting of the imaging, visualization, and evaluation of unsectioned tissue biopsies in 3D, become a reality. Similar to the advantages of CT over traditional X-ray, 3D pathology may enhance pancreatic cancer detection and further streamline the diagnostic process in pathology. However, similarities with traditional pathology in optical information needs to first be achieved in 3D pathology in order to pair with current skilled pathologists. Enough color contrast and resolution of intracellular and extracellular structures can provide desirable features of biopsies to pathologists. Hence 3D pathology project in Human Photonic Laboratory aims to accomplish this goal by performing staining, imaging, reconstruction, and 3D visualization for tissue biopsies. Preliminary result illustrates that the reconstruction algorithm with appropriate parameters based on frequency spectrum of data determines the resolution of reconstructed images. Specific 3D manipulation software Amira helps 3D visualization of reconstructed images and preserve color contrast using its colormapping feature. The investigation of proper colormapping protocol starts from colormapping in-silico image of processed 2D microscope slides. An imaging systems with designed image stitching program creates grayscale image of microscope slides. Spectrometer is used to take spectral reflectance measurement. A developing algorithm will take the data and calculate coordinate values of the specimen in color space of human vision. Those values will be used in Amira in colormapping for in-silico image of microscope slides to pair with full color image of slides. This colormapping technique has potentials to extend into unsectioned sample and hence combine with reconstruction algorithm to achieve colormapping for 3D visualization.

Systemic immune activation (IA) is a proven key factor in human immunodeficiency virus type 1 (HIV) infection pathogenesis and progression to acquired immunodeficiency syndrome. The reproductive hormone progesterone plays a role in regulating the immune system, and most cells contain progesterone receptors. Previous studies have linked high-progestogen states in women, such as pregnancy and progestin-based contraception use, to increased infectivity in HIV-seropositive women and increased HIV acquisition in seronegative women. Progestin-based contraceptives such as depot-medroxyprogesterone acetate (DMPA) can create high plasma levels of progesterone analogs. Therefore, studying DMPA use by HIV-seropositive and seronegative women may clarify DMPA’s role in HIV-related IA and genital shedding, and immune changes related to HIV acquisition. We hypothesize that DMPA increases systemic IA. Increased IA could be clinically significant by contributing to increased HIV acquisition in high-risk HIV-seronegative women, and by increasing genital shedding and infectivity among HIV-seropositive women. To answer this question, peripheral blood mononuclear cells (PBMC) were collected at three postpartum time points over six months from 120 Kenyan women with both HIV-seropositive and seronegative status, including a sub-population in each category that elected to receive DMPA for contraception. Two immunophenotyping panels were developed to measure T-cell IA and cellular subsets of interest using flow cytometry (CD38, HLA-DR, Ki-67, Bcl-2 for T-cell activation; FOXP3, CD25, CD127 for regulatory T-cells; CCR7, CD45RA, CD27, CD28, CD57 for memory subsets; CCR5 for the HIV co-receptor). With preliminary analysis completed for most samples, specific antibody populations have been gated with FlowJo software. After completion, we will compare different population percentages in the four arms of the study, observing how these change with DMPA administration. These calculations will help elucidate expected differences in inflammatory and regulatory changes in both HIV-seropositive and seronegative women with respect to the relationships between DMPA, immune activation, and HIV susceptibility and infectivity.

Although roughly two thousand exoplanets have been discovered and confirmed to date, exomoons have yet to be detected orbiting these planets. The detection of an exomoon would give insight into planetary formation and possibly increase the habitable real estate in a planetary system. Current telescopes are not capable of spatially resolving an exoplanet and its exomoon, or of separating the two blended spectra in a combined light measurement. However, previous work has shown that there is a wavelength dependent photometric centroid shift between a planet and its moon due to the weighted nature of the center of light (Agol et al. 2015). This spectroastrometric shift is highest in bands where the planet is dim and the moon is relatively bright, which can happen if it differs compositionally from its planet. As part of the “Finding the Needles in the Haystacks” project, we generated a realistic spatial / spectral model of an Earth-like exomoon orbiting a warm Jupiter in the habitable zone of a Sun-like star, including plausible exozodiacal dust structure. Preliminary results show that the presence of an Earth-like exomoon can produce centroid shifts greater than a milliarcsecond at some wavelengths, enabling the detection of the Earth-like exomoon even in the presence of dust. However, extracting the spectrum of the Earth-like exomoon has proved to be challenging, even when employing a simple telescope simulation devoid of coronagraphic effects, and further work will be needed to determine if it is possible even with 12-meter-class space telescopes.