Poster Session #1 (Saturday 10:45am 11:45am)
Hill Ballroom North
Easel #1: Vector-Targeted Subunit Vaccines for the Prevention of Vector-Transmissible Diseases
Tarab Ajjan (Georgia Institute of Technology)
Currently, there are over 500 Arboviruses on file, over a quarter of which can be transmitted to humans and animal reservoirs through vectors such as mosquitos¹. Given the rapid reproduction cycle of vectors, new vector control strategies must be developed to help combat the spread of these viruses. The goal of this study was to develop a subunit vaccine geared towards the control of these vectors such as ticks. Peptides previously identified in interfering in the reproduction of the Rhipicephalus sanguineus tick were genetically cloned into virus-like particles (VLPs) through bacterial expression. Leviviridae-derived PP7 VLP was used for antigen insertion. The most successful particles were injected into mice subcutaneously to assess their ability to elicit an antibody response. Serum collection and analysis was conducted to determine IgG titer levels. Results upon analysis through enzyme-linked immunosorbent assay (ELISA) analysis and further characterization through mass spectrometry, fast-protein liquid chromatography, and dynamic light scattering revealed early robust IgG titer levels and confirmation of particle formation. Through collaborations, we are currently working to improve vaccine design to confirm particle efficacy in their disease setting. We plan to continue to immunize with lower doses and different booster schedules to compare see if similar IgG titer levels are presented.
Easel #3: The Relationship Between Pituitary Volume and IGF-1 Blood Serum Biomarker Levels in Soldiers with Mild Traumatic Brain Injury History
Anna Castellano (University of North Carolina at Chapel Hill)
There is a high mild traumatic brain injury (mTBI) incidence rate in military and sport populations. Moderate and severe TBI is a leading cause of hypopituitarism; however, very few studies have examined the relationship between the pituitary gland volume and peripherally measured Insulin-like Growth Factor-1 (IGF-1) concentrations in those with a mTBI history. Additionally, there is no automated method to measure pituitary gland volume from Magnetic Resonance Image (MRI) scans. The objectives of this study are to create a program to automate measuring pituitary gland volume in military Service Members and student-athletes, to relate pituitary gland volumes to IGF-1 concentrations, and to determine if mTBI history influences this relationship. We will manually measure and segment pituitary gland volumes in our sample using available MITK software. The data will allow us to train imaging software to do the same using NVIDIA Clara AI software. After segmentation, we will relate pituitary gland volumes to IGF-1 concentrations. Then, we’ll assess these relationships with mTBI history. This may automate pituitary gland measurements – which is not well established, and may inform researchers and clinicians of potentially meaningful relationships.
Easel #5: The Role of Carnitine Acetyl Transferase in the Virulence of the Fungus Cryptococcus neoformans
Shaoni Dasgupta (Clemson University)
Developing a more in-depth understanding of the metabolism and physiology of the opportunistic pathogen, Cryptococcus neoformans, the leading cause of fungal meningitis worldwide, may facilitate identification of suitable targets for new anti-fungal therapies. Acetyl-CoA, an intermediate of numerous metabolic pathways, is a critical component of the cellular and biosynthetic pathways of Cryptococcus. This project specifically investigates the role of carnitine acetyl transferase, which is responsible for transporting acetyl-CoA across the mitochondria, as the amphipathic nature of acetyl-CoA hinders it from diffusing across the mitochondria.
Easel #7: Using Forster Resonance Energy Transfer to Understand the Location of Molecular Guests Encapsulated in Metal-Organic Frameworks
Noella D’Souza (Boston College)
The goal of this project was to use Forster-Resonance Energy Transfer (FRET), a spatially dependent fluorescence-based energy transfer, to study previous results suggesting that molecular guests encapsulated in host metal-organic frameworks (MOFs) via aperture opening encapsulation (AOE) are located close to the surface of the MOF particle. Since our encapsulation method is intended for heterogeneous catalysis, demonstrating surface guest distribution means that encapsulated catalysts would still readily interact with reactants in solution, demonstrating the industrial applicability of this method. This was accomplished by encapsulating the FRET donor Coumarin 151 in MOF particles via two ways, MOF synthesis and aperture-opening encapsulation. Subsequently, fluorescence spectra of either sample were taken in a solution of FRET acceptor Rhodamine 6G, since the acceptor emission intensity is correlated with the average proximity of guests to the particle surface. Results from this study suggested that the aperture-opening method did lead to surface distribution but also that MOF energy transfer might have been complicating the FRET phenomenon. This prompted a redesigning of the study for a different MOF and FRET donor, to avoid the MOF’s fluorescence. Recent results suggest that MOF particle size will have to be specified to enhance the observed difference in the spectra.
Easel #9: Probabilistic Enhancements to Bayesian Networks for Arabidopsis thaliana Gene Interactions
John Farrell (Wake Forest University)
The aim of this project is to develop a state of the art network model for molecular signaling. This NSF funded interdisciplinary project is a collaboration of researchers from biology, statistics, computer science and mathematics, and is associated with the WFU Center for Molecular Signaling. This URECA sponsored summer research improved the current gene interaction modeling algorithm (BCHC algorithm). The BCHC algorithm takes gene transcripts as input and outputs a network whose nodes are genes and whose edges are labelled with Bayesian posterior probabilities. Efforts focused on two objectives: implementing time series adjustment (TSA), and incorporating non-uniform prior probabilities. TSA was added to next state and cotemporal analysis paradigms in order to more accurately model Arabidopsis thaliana gene interactions. Interaction models generated with the addition of TSA displayed consistent posterior probabilities. Previously, forbidden edges were implemented exclusively with prior probabilities of 0. This summer, the BCHC algorithm was enhanced to include a range of prior probabilities. This addition allows for flexible gene interaction modeling, allowing the biologists to inform the BCHC algorithm on the probabilities of the relationships between specific gene pairs. Many gene models were constructed on simulated and experimental data. From the Muday lab, experimental data included Arabidopsis thaliana’s gene transcript selections of 12, 26 and 37 genes. Currently, a larger dataset of 65 gene transcripts is being collected in the Muday lab. The algorithm developed this summer will model this larger set. Network models can be computed for other types of data.
Easel #11: Influence of Textile Design on Wearable Biomonitoring Performance
Tashana Flewwellin (North Carolina State University)
Cardiovascular disease is the leading cause of death in the United States. Currently, cardiovascular disease is diagnosed by blood tests or electrocardiogram (ECG) tests. The holter monitor is the most common device used in clinical settings to measure ECG. Although these devices can measure ECG signals, holter monitors are rigid, bulky and require numerous wet electrodes to operate. These factors reduce the monitoring time of patients and thus cannot capture spontaneous cardiac events or ailments hidden deep within the heart. Wearable technologies have shown great promise in addressing these issues with cardiovascular healthcare, as they can continually monitor important vital signs on inconspicuous areas of the human body. This allows doctors to continually monitor patients and receive accurate data to diagnose a myriad of cardiovascular ailments. However, current wearable devices use rigid form factors which reduce user comfort and their sensing ability. Textiles on the other hand can allow for conformable, comfortable, and flexible wearable form factors. Textiles also allow for placement of electrodes and sensors at locations across the body in order to monitor key vital signs such as ECG. This research seeks to integrate ECG technologies into a textile armband form factor, to allow for real-time continuous ECG monitoring while utilizing a comfortable and flexible material that is ubiquitous in our daily lives. This research will specifically explore how the textile design of the armband influences its ECG sensing performance.
Easel #13: Community Gardens: Food Security and Cultural Preservation
April Hill (Syracuse University)
New Americans experience food insecurity at twice the rate of other communities due to barriers such as limited access to food assistance programs and limited social networks. Food insecurity is associated with poor diet quality and higher rates of disease. The purpose of this study is to determine how community gardens contribute to food security and cultural preservation in the New American population. A mixed-methods design was utilized, including environmental audits of food acquisition and interviews with community garden participants. Audit data was analyzed using descriptive and inferential statistics. Interview data was analyzed using a qualitative framework analysis. Environmental data (n=37) included food acquisition sites located within a specified half-mile radius on Syracuse’s Northside. 35% of stores sold 51+ items associated with a specific ethnic group (cultural foods), while 56.8% of stores sold 0-10 cultural foods. Mean total number of fresh vegetables sold in stores with 31+ cultural foods (31.15±18.6) was higher than stores with 0-30 cultural foods (3.83±2.64) (two sample t-test, p < 0.001). 78% of stores sold milk, with 60% of stores with 31+ cultural foods selling milk. Price data was not easily obtained, as only 32% of stores had price data clearly visible for most items. Preliminary findings suggest ethnic stores contribute significantly to a healthy food environment, providing fresh produce and staple items. Food selection differed between cultural and traditional food acquisition sites. Lack of pricing data makes it difficult to determine the impact of cost accessibility.
Easel #15: Tobacco Consumption and Engagement in Preventative Behaviors with Patients Among Chilean Healthcare Workers
Maryam Jawid (University of Miami)
Smoking is responsible for the deaths of 12,000 Chileans each year. Healthcare workers are known to play an integral role in assisting patients with smoking cessation via hospital-based smoking cessation interventions. As part of Chile’s first hospital-based smoking cessation intervention, healthcare workers in Santiago, Chile were surveyed about demographics, smoking habits, and engagement in tobacco preventative behaviors with patients. The survey utilized multiple-choice questions and the Common Practice Scale, a four question Likert Scale (Chronbach’s alpha: 0.856). 65 healthcare workers (82% female) completed the survey. Tobacco use prevalence was 32%, with 48% of users consuming 1-5 cigarettes daily. The average preventative behavior engagement score was 9.5/16. No significant differences in preventative behavior engagement scores between smokers and non-smokers (t=-0.78; p=0.44) or between preventative behavior engagement scores and daily number of cigarettes smoked (rs=-0.169; p=0.53) were found. The small sample size of this study was a limitation, but the average engagement score of 9.5/16 indicates much room for improvement and supports the need for intervention considering the massive global disease burden of tobacco use.
Easel #17: A Supplementary Treatment Targeting the Peripheral Deficits Caused by Spinal Muscular Atrophy
Cassandra Kaufhold (University of Pittsburgh)
Spinal Muscular Atrophy (SMA) is a neurodegenerative disease that causes deterioration of the neuromuscular junction preceding motoneuron death. The only treatment for SMA utilizes gene therapy, which remarkably improves motoneuron death; however, it has yet to be determined how gene therapy modulates neuromuscular pathology. Using ex vivo immunohistochemistry, we characterized the SMA-induced neuromuscular pathology and found a denervation rate of ~10%, a 22% reduction in myofiber endplate size, and 25% increase in neurofilament (a cytoarchitecture protein). SMA mice treated by gene therapy have rescued endplate growth, decreased neurofilament accumulation, and improved innervation. We next tested neuromuscular function using in vivo righting reflex and grip strength assays. SMA mice had an 80% reduction in motor function scores and were 75% weaker than control mice. Treated SMA mice had a 50% reduction in motor function scores and were 50% as weak as control mice. To address persistent deficits in strength, we tested a novel treatment that can supplement gene therapy. We administered an acute, subcutaneous dose of a calcium channel agonist, GV-58, and a potassium channel antagonist (3,4-DAP), which increase neuromuscular synaptic transmission by independent mechanism of actions. In untreated SMA mice, GV-58 increases grip strength by 25%, and by 45% when combined with 3,4-DAP. In treated SMA mice, GV-58 with or without 3,4-DAP increased grip strength by ~42%. This characterization of peripheral dysfunction indicates gene therapy is not sufficient to rescue SMA pathology; however, our data suggests that GV-58 plus 3,4-DAP is an excellent candidate for increasing neuromuscular function in SMA.
Easel #19: How News Media Coverage of Crises Promotes Conspiracy Beliefs
Richard Kornrumpf (University of Louisville)
While scholars of conspiracy theories have recently made great strides in understanding the basic nature and correlates of conspiratorial thinking, we still know little about how conspiracy beliefs are disseminated and communicated, especially when it comes to traditional media. In this instance, we use a unique experiment to investigate whether media coverage of mass shootings – complete with the uncertainty, conﬂicting reports, and dubious oﬃcial narratives that characterize such coverage – provides the raw material for conspiracy theories and promotes conspiracy beliefs among viewers. We ﬁnd that implicit conspiratorial information – that which causes confusion and foments uncertainty – does not enﬂame conspiracy beliefs. However, more explicitly conspiratorial information – that which challenges the oﬃcial narrative and questions the details of investigative ﬁndings – does, fostering conspiracy beliefs about second shooters and governmental false ﬂag operations. These ﬁndings suggest that conspiracy theories may be a natural consequence of the 24/7 news cycle. [Poster]
Easel #21: An Analogue of Gauss Composition for Binary Cubic Forms
Benjamin Nativi (Duke University)
Consider primitive binary quadratic forms, i.e. p(x, y) = ax2 + bxy + cy2 where a, b, c are integers such that gcd(a, b, c) = 1. Gauss showed that one can compose two such binary quadratic forms p1(x1, y1) and p2(x2, y2) of the same discriminant and choose appropriate new variables X and Y as integral bilinear combinations of x1, y1, x2, y2 to get a new form P (X, Y ) of the same discriminant. This composition is well defined on SL2(Z)-equivalence classes and although Gauss’s proof of this composition is tediously computational, more recent proofs such as that by Dirichlet are less cumbersome and more intuitive. One benefit of the study of binary quadratic forms is their relation to the class group. For a ring RD = Z[ D+ D ], its 2 class group is the set of fractional ideals quotiented by the principle fractional ideals. It turns out the elements of the class group correspond to SL2(Z)-equivalence classes of primitive BQF’s and that the composition of ideals gives the same ideal class as that from composing the corresponding BQF’s. This allows the study of class groups by studying BQF’s. In my work, under the supervision of Professor Aaron Pollack, I have expanded on results of Bhargava to show that there is a similar composition of projective binary cubic forms. The SL2(Z)-equivalence classes of BCF’s correspond to elements of the 3-torsion of the class group of certain rings. I was able to prove that, as in the quadratic case, the composition of BCF equivalence classes is equivalent to the composition of ideals. My result is intriguing for two reasons. First, although it was clear from the work of others that some composition law existed on classes of BCF’s, I was able to explicitly demonstrate this law in a form analogous to Gauss composition. Second, my result furthers the connection between classes of forms and the ideal class group and may inspire similar results in related settings.
Easel #23: Haenyeo House
Taylor Olson (Virginia Polytechnic Institute and State University)
The main goal of this project is to identify and reflect the cultural aspects of the South Korean Jeju women divers, or Haenyeo, to the home design by using sustainable features, materials, and finishes with affordability in mind. The purpose of this project is to provide affordable, and sustainable housing for the Haenyeo (who are known go through extreme work conditions to provide for their families until they are physically unable to work) to age in place comfortably. The process of creation includes taking an Implicit Association Test, in-depth research on the topic of Aging-In-Place and Multiculturalism, an in-person interview with Dr. Seung Koh, Director of Jeju Aging Society Research Center, creating an inspiration collage, design concept/goals, schematic plans, choosing materials and features, and lastly, finalizing floor plans, elevations, and 3Ds. In conclusion, I successfully maximized the usage of the small space by featuring flexible, multi-use furniture in an open concept floor plan, all while continuing the main design goals of sustainable coexistence with the natural environment and by balancing the modern and traditional aspects of Jeju culture.
Easel #25: Energy Calibration in the Interface of Tandem Trapped Ion Mobility Spectrometry
Valentina Rangel Angarita (Florida State University)
Significant efforts are being devoted into revealing the molecular mechanisms by which different proteoforms can exhibit different biological activities. Tandem-trapped ion mobility spectrometry (tandem-TIMS) has the potential to determine the structure of a protein and simultaneously its sequence and post-translational modifications. A central aspect is the ability to characterize the unfolding pathways of the protein in the gas-phase. Ions are accelerated by an electric field in the presence of a buffer gas and the gained kinetic energy is converted into internal energy of the ions through collisions with the buffer gas. What remains unknown, however, is how analytes are collisionally-activated in tandem-TIMS. I show using reference molecules that it is possible to relate the dissociation barrier of ions to the activation voltage applied in tandem-TIMS and construct a strong linear model that predicts either the activation energy and the Gibbs energy of activation as a function of the activation voltage needed for dissociation and other parameters that account for the identity of the different analytes. Thus, it is possible to predict the dissociation reaction rate with the model constructed, and can illustrate the different tendencies on reaction dynamics.
Easel #27: Exploration of Emotion Expression in Text Messages Preceding a Suicide Attempt
Maya Stephens (University of Virginia)
Suicide is the second-leading cause of death for teenagers and young adults. With the growing popularity of technology, researchers have begun using social media to research trends in suicidality. However, there has been very little research done with one of the most popular forms of communication: text messaging. Since text messages are more likely to contain personal conversations than public social media posts, research on how suicidality is expressed in text messages is critical for understanding and preventing attempts. Previous research has shown that emotion expression plays a key role in identifying at-risk individuals (Negron et al., 1997). Additionally, machine learning techniques can identify markers of suicidality through text analysis (Desmet & Hoste, 2014). In the present study, we aim to compare how emotions expressed in text messages sent two weeks prior to a suicide attempt correspond with participants’ self-reported, retrospective emotion states from each incident. 46,800 texts will be analyzed from 15 participants using machine learning techniques. All participants gave informed consent to having their text messages extracted from their phones. We expect to see that emotion identified in texts from two-weeks before a suicide attempt will correspond significantly with retrospective emotion self-reported by participants from the same time period. We are examining this from the lens of lexicon validity, participant disclosure rates, and retrospective accuracy.
Easel #29: Machine Learning on Student Learning: Modeling Differences Between Various Student Learning Outcomes in AP Statistics Classrooms
Honoka Suzuki (University of Notre Dame)
In an Advanced Placement (AP) classroom, there are several methods in which a student’s learning can be assessed: a final class grade, AP exam score, or practice AP exam score. These assessment methods hold distinct characteristics, varying from high-stakes to low-stakes, and from capturing a cumulation to a ‘snapshot’ of understanding. Given the contrasting natures of the three assessments, the present study aims to investigate if these assessments are different and what factors explain such difference. For example, if a student receives a high class grade but a low AP exam score, what traits about that student might explain this discrepancy? Using a sample of 381 AP Statistics students, the three learning outcomes and the differences between outcomes are modeled with regression trees and random forests. Models predicting individual learning outcomes revealed that AP exam score and practice AP exam score are predicted by similar factors, whereas class grade is predicted by factors distinct from the other two outcomes. Modeling the difference between the AP exam score and class grade revealed that the discrepancy is explained by a student’s school, age, and behavioral engagement score. The study furthers the understanding of educational and personal factors that differentially impact students’ performances on various assessments, as well as the roles and characteristics of assessments that warrant such differences.