Virtual Scholarship Summit – draft

MWSU CTL Presents the Fall 2020 Virtual Scholarship Summit

College of Business and Professional Studies

College of Liberal Arts

Sathiavanee Veeramoothoo (Economics), “Factors that Affect Bank Capital Structure Decisions and Those that Surprisingly Do Not: A Panel Data Analysis with Error Cross Section Dependence”

Abstract: We use quarterly U.S. bank level data on depository institutions from 2012 through 2017 to study the factors that affect a bank’s capital structure. We test and find cross section dependence (CSD) in the data and confirm that when we ignore CSD, the four factors widely identified the literature as determinants of firm and bank leverage, namely, the market-to-book ratio, collateral, profitability and size, are significant which is consistent with the literature. However, this is no longer true when we account for the CSD since only profitability remains a significant determinant of bank leverage. We thus propose retained earnings, non-interest expenses, sovereign bond holdings and the federal funds bought as new determinants of capital structure in addition to profitability, and estimate our model using five estimation methods that account for CSD, namely, the Common Correlation Effects (CCE) model, the Common Correlation Effects Pooled (CCEP) model, the Dynamic Common Correlation Effects (DCCE) model, the Dynamic Common Correlation Effects Pooled (DCCEP) model and the Cross-Section Distributed Lags (CS-DL) model. Sovereign bond holdings and the federal funds bought are included as proxies for internal and external risk perceptions, respectively. Surprisingly, they are not significant. The three factors that we retain as determinants of bank leverage under CSD are therefore retained earnings, non-interest expenses and profitability. This study shows that the current models of capital structure used in the literature lead to the erroneous conclusion that some factors are significant determinants of bank leverage when they are not. More work in other industries can be done to determine whether this conclusion holds more generally, or if it is only true of financial firms. JEL Codes: G21, G31, G32 and C23. Keywords: capital structure, retained earnings, non-interest expenses, profitability, error cross section dependence.

College of Science and Health

Csengele Barta (Biology),  Lauren Tinoco, Kaitlyn Atkins, Harrison Meers, Brian Jenkins, Devon Lindstrom, Rachael Prawitz (MWSU Undergraduate Co-Authors), “Plants’ Arsenal for Waging Chemical Warfare on Their Unsuspecting Neighbors: Does the Synthesis and Release of Secondary Negative Allelopathic Metabolites into the Environment Provide a Competitive Advantage to Invasive Plant Species in Missouri Habitats?”

Abstract: Invasive species, employing a large array of strategies, through which they negatively impact the germination, growth, survival, or reproduction of native species, are a threat to biodiversity and a relevant concern for species conservation and restoration. One such strategy is the synthesis and release of allelopathic metabolites into the environment. Decomposing leaves of the invasive Amur honeysuckle (Lonicera maackii), rapidly invading the Midwestern US leach a variety of phenolic molecules into the soil, with devastating inhibitory effects on native species in the region. Nevertheless, the action mechanism of inhibition has not yet been studied. In this work we explored the inhibition mechanism triggered by honeysuckle leaf extracts in germinating standard (control, Rbr), gibberellin (GA) synthesis deficient (Rosette-Dwarf, ros) and GA overproducing (Tall, ein) mutants of field mustard (Brassica rapa L. var. rapa), in a bioassay. Honeysuckle leaf extracts (0.01-0.2 g/mL) significantly decreased the germination of mustard seeds and inhibited seedling development in a concentration-dependent manner with most pronounced impacts on the ros mutants, whose germination was arrested even at the lowest tested extract concentration. The ein seeds were able to overcome inhibition, with only a delay in their germination, at the highest extract concentrations. We observed a similar relationship when control seeds were treated in the presence of exogenous GA (up to 100 μM), with the strength of inhibition decreasing with increasing GA amounts. We hypothesize, that the allelochemicals in the honeysuckle leaf extract alter the hormonal balance in germinating mustard seeds, effect which however, can be alleviated by higher than physiological GA concentrations. A better understanding of the inhibition mechanism by allelopathic agents is expected to contribute to developing effective invasive species management approaches in the future.

Dawn Drake (Geography), “Can Suburbanization Be a Savior for Farms in the US’ Dairy Death Spiral?”

Abstract: The expanding provision of urban activities, including housing, employment, and retailing, at the rural-urban fringe generally creates both challenges and opportunities for the agricultural community. For livestock operations, the headaches related to new neighbors that expect livestock to abide by quiet times and be generally odorless and unobtrusive, outweigh the opportunities. This often leads to shifts in operations from livestock to grain or total industry exit.

At the same time that these pressures are occurring, the US dairy industry is facing a death spiral of sorts. Production is at record highs, which is driving prices to record lows. Consumers are choosing alternatives to dairy milk, despite the known nutritional benefits. A series of trade wars are further eroding the market for dairy products. School closures due to the COVID-19 pandemic removed one of the most dependable markets for the sale of milk, resulting in the dumping of hundreds of thousands of gallons of milk. With all of these challenges conspiring against the dairy industry, it is not surprising that farmers are exiting the industry at a never before seen rate, even in rural counties.

The question is, can suburbanization help some farmers resist the dairy death spiral? There are two sides to consider. Suburbanization drives out some dairy farms, which allows others to survive, as supply decreases. Since dairy farms in suburban areas tend to have smaller herd sizes, this is only small salvation. The way in which suburbanization will be a savior for the dairy industry is by providing opportunities for high value niche production, demanded by new consumers in suburban areas. Dairy farmers who provide local organic or raw milk to discerning consumers not only will have higher profit margins, but may be able to escape the dairy death spiral.

This paper will examine the current conditions in the dairy industry, paying special attention to the rural-urban fringe. Through the use of field research and data from the USDA, the study will seek to understand the current challenges and opportunities for dairy farmers in suburban areas as well as hypothesizing the future as urban and suburban land uses continue to encroach on the US dairy industry. Will the outcome be positive or will an increasing percentage of US dairy farms continue to relocate to distant locations, far from the markets they serve.

Michael Ducey (Biology) and John Kosse (MWSU Undergraduate Co-Author), “Development of a Low Cost Raman Spectrometer”

Abstract: A low cost Raman spectrometer was developed using a simple green diode laser and a diode array spectrometer with fiber optic input. This system can be employed in a variety of  educational formats including instruction on instrument design and operation as well as in applications of laser and vibrational spectroscopies. Using this approach Raman system was developed for under $3000 and was employed in a proof of concept in the examination of the Raman spectrum of carbon tetrachloride.

Ashley Elias (Biology), “Using Next-Generation Sequencing to Test the Evolutionary Origins of the Vertebrate Paired Appendage Gene Regulatory Network”

Click Here for Private Link to Video

Abstract: The developmental basis of the transition from paired fins to limbs remains a longstanding question in evolutionary biology. In current models, emphasis on skeletal type led to the hypothesis that fin-folds and autopods (hands/feet) are not homologous, patterned by different developmental modules, despite similar distal positions. However, growing evidence supports an alternative hypothesis: That modularity is conserved across fins and limbs, while skeletal potentials within compartments have evolved within lineages. Functional tests in basal non-teleost finned vertebrates are needed to reconcile the fossil patterns with disparate results from developmental studies in derived tetrapod and teleost systems.

This collaborative project is building a model for the evolutionary origin of the distal paired appendage gene regulatory network (GRN), by expanding on recent discoveries in phylogenetically well-positioned vertebrate the American paddlefish (Polyodon spathula). Preliminary work raises the intriguing alternative hypothesis that a deep regulatory homology is shared by autopods (hands/feet) and fin-folds. This new model predicts that a homologous GRN drives development of the distal compartment in all vertebrates. To test this model, we are using RNA-seq to build a comprehensive map of transcriptional variation in the proximal and distal compartments of the paired fins of paddlefish. These datasets are being compared to existing transcriptomic/genomic resources for tetrapods and zebrafish to identify candidate distal appendage GRN genes.

Similarities between fin-folds and autopods (hands/feet) are underexplored. We hypothesize that the autopod shares a deep regulatory homology with the fin-fold and this project will elucidate key components of the gene regulatory network common to the structures. This project will also generate new research directions in the earliest differentiation events that regulate skeletogenesis and help to answer long-standing questions regarding the evolutionary origin the autopod. Significantly, if our results confirm a deep regulatory homology is shared between fin-folds (which often retain significant regenerative capacity) and autopods (which generally lack this capacity), then this project would reinvigorate interest in the use of fins for the study of congenital limb birth defects in humans and limb regeneration studies.

R. Grey Endres (Social Work), “Child Welfare is Failing to Protect the Children It is Mandated to Serve”

Abstract: Within the Grand Challenge of ensuring healthy development of all youth, the child welfare system is failing to prevent harm to the children it is mandated to protect. The norm that holds this problem into place is the default setting that child welfare knows what is best for the children and families it serves. Adultcentrism (Petr, 1992) is held in place by the need to rescue the child, unconscious bias, blinding following “old” practices, and the notion that confidentiality supersedes transparency. This norm effect millions of youth, costs millions of dollars, and is further complicated by social inequalities that target youth from poor families, homelessness, minorities, immigrants, domestic violence, substance abuse, and unemployment. In her examination of Pittsburgh, Pennsylvania’s child welfare system, Eubanks (2017) found that poor families often feel “forced to trade their rights to privacy, protection from unreasonable searches, and due process for a chance at the resources and services to keep their children safe” (p.158). Subsequently, upper and middle-class families do not experience the same feelings when they reach out to privately funded resources.

The goal of any child welfare service should be to seek to understand before seeking to be understood while creating a cooperative alliance with the client and their systems. In addition, they should focus on collaborative interventions (prevention), accountability, and increased transparency with the public including giving client’s a voice.

To disrupt the norm of adultcentrism, research supports the best practice of shared decision making. Shared decision making can unlock and empower the voice of those being served so that they can be seen as their own experts by child welfare. In collaborative pilot with CASA (Court Appointed Special Advocates) in Kansas City, Missouri, this researcher has developed an innovation entitled Operation Hope, and designed an interactive comic book within a shared decision framework that teaches children how to have and use their own voice while assisting them to better understand child welfare policies and practices.

Operation Hope is supported by three reasons including the best practice of shared decision making, comic book fans can connect to their hero’s experiences of trauma and loss, and it is grounded in empowerment theory. It will teach and give those being served by child welfare a voice that will lessen the negative impacts of child welfare.

Julie A. Jedlicka (Biology), “Beneficial Birds Foraging in Temperate and Tropical Agricultural Lands”

Abstract: Determining the ecosystem function of high-order predators is critical for evaluation of food web interactions. Insectivorous birds are abundant predators in many ecosystems yet because they forage upon such small taxa, it remains largely unknown whether birds are providing ecosystem services in the form of pest control or disservices by preying upon predaceous arthropod species, functioning as intraguild predators. I provide a detailed look into three approaches used to quantify the ecosystem services provided by insectivorous birds comparing stomach-content analysis to more recent molecular scatology methods. In Mexican shade-coffee farms strong annual changes in diets from year to year and across shade management systems highlighted the importance of long-term studies and emphasized the plasticity of insectivorous and omnivorous birds to respond to fluctuations of arthropod populations. In California vineyards, adult and nestling Western Bluebirds (Sialia mexicana) were found to consume mostly herbivorous insects, primarily from the orders Hemiptera and Lepidoptera, representing over half (56%) of the prey items. Intraguild predation (of predator or parasitoid arthropods) represented only 3% of adult and nestling dietary items. Finally, I present a case study of recent work analyzing the arthropod diets from insectivorous birds foraging in Kenyan coffee farms and on Missouri Western campus. The focus is to determine whether bird foraging is providing avian-induced ecosystem services of pest control.

Joseph Kendall-Morwick (Computer Science), “The Scholarly Impact and Potential of Open-Source Software Development”

Abstract: Beyond being a model for software development, free and open source software (FOSS) development is having a cultural impact on the commercial software industry. Traditionally, software developers seeking employment would burnish their credentials through their employment history and recommendations from supervisors and co-workers. However, employers are increasingly judging applicants’ potential through analysis of their contributions to open source projects. This raises the importance of student-faculty FOSS collaborations both as applied learning activities and as opportunities for portfolio and reputation building for the student.

FOSS is also having a cultural impact on academia, particularly in computer science. The open and transparent development of software that is freely available, reproducible, and modifiable is inherently scholarly in its application of computer science to solving important problems faced by the communities served by the university. For many researchers in CS, a paper is written and published largely to legitimize the researcher’s more substantial software development efforts
as scholarly work, but emerging metrics for measuring the impact of FOSS contributions are driving FOSS development towards being a first-class scholarly activity.

Thus, those in academia should endeavor to make a greater impact with their work through development of free and open source software. Although FOSS has seen many successes (such as the Linux operating system), it has not made considerable headway into many domains where private cloud-based software dominates (including social media, citizenship, health care, transportation, communication, and many others). University collaborations in FOSS development, whether through research projects or applied-learning projects, have the potential to bring universities’ problem-solving capacity to bear on the problems faced by their communities and better realize the ultimate goals of scholarship within academia. This presentation will expose the scholarly nature of FOSS contributions currently developed by both students and faculty at universities worldwide. It will identify the alignment of goals of FOSS with those of universities involved in research and applied learning. Finally, it will make the case for the scholarly value FOSS development at universities and identify future avenues for this work to enhance the university’s civic and societal impact.

Yipkei Kwok (Computer Science), “Assessing the Improvement Space and Performance Tuning of Learning From OPT, a State-of-the-Art Artificial Intelligence-based Caching Scheme for Content Delivery Networks”

Abstract: A content delivery network (CDN) is a network of cache servers that cache web objects (e.g., YouTube videos) to provide low-latency web accesses. When an end-user requests an object that is available on a nearby cache server, the server services the request and sends the object directly to the end-user to facilitate a low-latency access. Essentially, a CDN’s performance is measured by object hit ratio (OHR) – the percentage of requests that its servers service. The lower the OHR, the more requests the CDN forwards to distant content providers (e.g., YouTube). The key to achieve high OHR is by caching objects that will most likely be accessed soon. Such decisions are made by the caching scheme on each server. A caching scheme decides when to  admit and evict objects and identify objects for admission and eviction. Traditionally, object access recency and frequency have been popular heuristics for guiding caching decisions. Lately, researchers have demonstrated the promise of using Artificial Intelligence (AI) for making caching decisions. We studied a recent AI-based caching scheme known as Learning From OPT (LFO), which continuously trains a gradient-based decision tree to mimic the decisions of OPT, a theoretical caching scheme. Though OPT is impractical to implement given the unrealistic assumption of knowing when future accesses happen, its caching decisions are proven to be optimal, which provides a direction for training the decision tree. Despite showing great promise, no published work has (1) investigated how far LFO’s effectiveness, in terms of OHR, is from the optimal or (2) explored the opportunities of performance optimization.

The contributions of this work are twofold. First, we evaluated LFO using production web request traces and measured how far LFO’s performance is from OPT’s theoretical optimal (i.e., space for improvement). To do so, we used a technique known as practical flow-based offline optimal (PFOO), which, given a request access trace and the cache size, calculates the theoretical optimal hit ratio. Second, we fine-tuned LFO and evaluated the effectiveness of our optimization. Even though LFO is a sophisticated caching scheme, there are opportunities where performance optimization is possible. This work laid the groundwork for performance optimization of other AI-based caching schemes. In the future, we plan to investigate the potentials of using other AI-based techniques, such as machine learning, for caching schemes.

Jones Mutua (Math), “Modeling the Effects of Antibody Responses under Drugs of Abuse on HIV Dynamics”

Abstract: Mathematical models are useful in studying the dynamics of infectious diseases, both in within-host and between-hosts. Here, we present within-host disease modeling projects which consider ‘human immunodeficiency virus (HIV) under drugs of abuse’ to demonstrate how modeling can provide great insights into complex phenomena of infectious diseases. The within-host models presented here can help address some of the issues related to the HIV dynamics in the presence of drugs of abuse. The models help evaluate morphine-induced alterations in target cell susceptibility and HIV-specific immune responses that may cause higher viral replications and accelerated disease progressions in HIV infected drug abusers.

Amit Verma (Business), Mark Lewis (Business), and Todd Eckdahl (Biology), “Qfold: A New Modeling Paradigm for the RNA Folding Problem”

Abstract: Ribonucleic Acid (RNA) molecules play informational, structural, and metabolic roles in all living cells. RNAs are chains of nucleotides containing bases {A,C,G,U} that interact via base pairings to determine higher-order structure and functionality. The RNA folding problem is to predict one or more secondary RNA structures from a given primary sequence of bases. From a mathematical modeling perspective, solutions to the RNA folding problem come from minimizing the thermodynamic free energy of a structure by selecting which bases will be paired, subject to a set of constraints. Here we report on a Quadratic Unconstrained Binary Optimization (QUBO) modeling paradigm that fits naturally with the parameters and constraints required for RNA folding prediction. Three QUBO models are presented along with a hybrid metaheuristic algorithm. Extensive testing results show a strong positive correlation with benchmark results.

Pi-Ming Yeh (Nursing) and Gavin Waters (Math), “Path Analysis Testing the Development of Personality and Psychological Well-Being Model”

Abstract:

Background: A person’s psychological well-being and suicidal ideation are impacted by family interaction, especially parenting. Little is known about the long term impacts of parenting styles on psychological well-being in later life.

Purpose: The purpose of this study was to test the Development of Personality and Psychological Well-Being Model. This was a cross sectional, descriptive design.

Methods: The 448 participants were recruited from older adults living in a Midwest community in the United States. After IRB approval, the trained researchers explained this study to older adults who lived in the community. After agreeing to participate, they signed an informed consent form. Upon completion of a questionnaire, the participants received a pack of chocolate. The structured questionnaires were used to perform data collection. Path Analysis was used to examine this model. SPSS 23 version was employed to examine the instruments’ reliabilities and descriptive data.

Results: The mean age was 71.33 (SD = 7.61). There were 184 male (41.1%) and 264 female (58.9%) participants. In this model, family interaction and spiritual well-being had significant influences on personality, which had a significant influence on selecting coping strategies. Using different coping strategies influenced on the outcome variables (e.g., psychological well-being, anxiety, depression, and suicidal ideation). The family interaction and spiritual well-being played a critical role on the outcome variables. The following four hypotheses had been examined and had statistically significant relationships.

Study Hypotheses:

The first hypothesis: Family interaction and spiritual well-being had significant influences on the development of a person’s personality.

The second hypothesis: A person’s personality had a significant influence on selecting coping strategies.

The third hypothesis: Using different coping strategies had significant influences on the outcome variables (psychological well-being, anxiety, depression, and suicidal ideation).

The fourth hypothesis: The family interaction and spiritual well-being had significant influences on the outcome variables (psychological well-being, anxiety, depression, and suicidal ideation).

Jeff Woodford (Chemistry) and Nelson Maxey (MWSU Undergraduate Co-Author), “Fully Numerical Computation of Vibrational Properties of Small Molecules”

Abstract: With the advancement of more powerful computing resources, it is becoming more feasible to approach solving the Schrödinger Equation without having to rely on approximate models. In this presentation, the Fourier Grid Hamiltonian (FGH) approach will be presented as one possible method for numeric computation of the vibrational Schrödinger Equation. Current attempts to implement this method at MWSU will be presented.