Category: Graduate

Faith Sheedy: Faith and Spirituality On-Campus: Exploring the Spiritual Needs of Students

Posted on

Co-author: Laura Bell

https://youtu.be/uYhyAVC7yC4

This project will focus on analyzing student engagement with their spirituality through the Office of the Chaplain and Religious Life as well as other faith-based campus resources. The project will use both qualitative and quantitative methods to understand and gauge if the spiritual needs of students are being met and how SMU can support of students' spiritual lives. Utilizing Astin, Astin, and Lindholm (2010) national study of undergraduates' spiritual growth we seek to understand the spiritual and faith needs of on-campus residential students. Results may inform the work of religious and spiritual life centers.

Faith Sheedy
Program: Master’s in Theological Studies
Faculty mentor: Dustin Grabsch

Megan Simons: Transition-potential coupled cluster

Posted on

https://youtu.be/oTt_Ut-7E4o

The problem of orbital relaxation in computational core-hole spectroscopies, including x-ray absorption and x-ray photoionization, has long plagued linear response approaches, including equation-of-motion coupled cluster with singles and doubles (EOM-CCSD). Instead of addressing this problem by including additional electron correlation, we propose an explicit treatment of orbital relaxation via the use of “transition potential” reference orbitals, leading to a transition-potential coupled cluster (TP-CC) family of methods. One member of this family, in particular, TP-CCSD(12/ 1 2 ), is found to essentially eliminate the orbital relaxation error and achieve the same level of accuracy for the core-hole spectra as is typically expected of EOM-CCSD in the valence region. These results show that very accurate x-ray absorption spectra for molecules with first-row atoms can be computed at a cost essentially the same as that for EOM-CCSD.

Megan Simons
Program: PhD in Theoretical and Computational Chemistry
Faculty mentor: Devin Matthews

Jase Sitton: Indirect Bridge Monitoring Using Crowdsourced Smartphone Data from Passing Vehicles

Posted on

https://youtu.be/mcIlXNE2kl4

United States bridge infrastructure is aging, with more than 231,000 bridges currently in need of repair, rehabilitation, or replacement. In light of the urgent needs surrounding aging bridge infrastructure, and in the absence of sufficient funding to address all required bridge repairs, it is crucial to identify which bridges have the most immediate need for the allocation of available funds for bridge repairs. Typically, structural health monitoring comprises the installation of permanent sensors on a bridge superstructure and the collection of data for analysis to determine bridge condition. These systems, however, are often expensive and logistically difficult to install on a large number of bridges. Smartphones contain a variety of sensors, including accelerometers and gyroscopes, that may be used to record bridge vibration response data from within vehicles as they traverse bridges. This data may then be used isolate the bridge's vibration response and make bridge condition assessments at regular intervals, and bridges requiring immediate attention can be flagged.

Jase Sitton
Program: PhD in Civil and Environmental Engineering
Faculty mentor: Brett Story

Elyssa Sliheet: Genetic Network Analysis

Posted on

Co-author: Molly Robinson

https://youtu.be/OiPjD0fuQvI

There exists the need to develop and study the effects of novel immunotherapies for the treatment of various cancer types. Unlike chemotherapy and radiotherapy which directly target the cite of a tumor, immunotherapies serve as catalysts to activate the body's immune response. Studies have shown the benefits of immunotherapies. However, there is still need to computationally analyze genetic data to identify relevant biomarkers as related to clinical significance for the overall advancement of targeted medicine. In this research, we construct and analyze important properties of a protein-protein interaction network. These properties shed light on potential biomarkers relevant to patient immune response and sensitivity.

Elyssa Sliheet
Program: PhD in Mathematics
Faculty mentor: Andrea Barreiro

Zilin Song: Quantifying Energy Contribution in Enzyme Catalysis using Machine-Learning based Regression Analysis

Posted on

Winner: Theoretical and Computational Chemistry (Graduate)

https://youtu.be/3IXmP6v7jRM

The bacterial enzyme class of β-lactamases are involved in benzylpenicillin acylation reactions, which are currently being revisited using hybrid quantum mechanical molecular mechanical (QM/MM) chain-of-states pathway optimizations. Minimum energy pathways are sampled by reoptimizing pathway geometry under different representative protein environments obtained through constrained molecular dynamics simulations. Predictive potential energy surface models in the reaction space are trained with machine-learning regression techniques. Herein, using TEM-1/benzylpenicillin acylation reaction as the model system, we introduce two model-independent criteria for delineating the energetic contributions and correlations in the predicted reaction space. Both methods are demonstrated to effectively quantify the energetic contribution of each chemical process and identify the rate limiting step of enzymatic reaction with high degrees of freedom. The consistency of the current workflow is tested under seven levels of quantum chemistry theory and three non- linear machine-learning regression models. The proposed approaches are validated to provide qualitative compliance with experimental mutagenesis studies.

Zilin Song
Program: PhD in Theoretical and Computational Chemistry
Faculty mentor: Peng Tao

Siavash Tabrizian: A sampling-based branch and cut algorithm for two-stage stochastic mixed integer programs

Posted on

Winner: Operations Research (Graduate)

https://youtu.be/6E7GQBXBa3c

Stochastic mixed-integer programs are among the most challenging class of optimization problems that finds many applications in practice. In this presentation, we describe a novel algorithmic framework for solving two-stage stochastic mixed-integer programs using internal sampling.

Siavash Tabrizian
Program: PhD in Operations Research
Faculty mentor: Harsha Gangammanavar

Micah Thornton: Examining Uses of DFT distance metrics in SARS-CoV-2 Genomes

Posted on

Co-authors: Monnie McGee

https://youtu.be/bFW9xMpdSp0

The Fourier Coefficients (FC) of a genomic sequence can be calculated according to a method proposed earlier this decade by Yin et al. Here we are concerned with the efficacy of these coefficients in capturing useful information about viral sequences. The FCs are rapidly computable and comparable which allows for speedy real-time numerical analyses of sequences. In this work we investigate using the FCs as summaries of SARS-CoV-2 sequences by applying regional classification procedures, and graphical examination. Specifically we extract geographic submission location from sequences submitted to the GISAID Initiative, and attempt to use the FCs to classify these sequences in addition to displaying them visually utilizing dimensionality reduction. We show that the FCs may serve as useful numerical summaries for sequences which allow manipulation, identification, and differentiation via classical mathematical and statistical methods not readily applicable to character strings. Further we argue that subsets of the FCs may be usable for the same purposes, indicating a reduction in storage requirements. We conclude by offering extensions of the research, and potential future directions for subsequent analyses and further theoretical development of techniques specific to the FCs and suggesting different kinds of series transforms for discretely indexed signals like genomes.

Micah Thornton
Program: PhD in Biostatistics
Faculty Mentor: Monnie McGee

Hao Tian: Prediction of allosteric sites through ensemble learning

Posted on

Winner: Theoretical and Computational Chemistry (Graduate)

Co-author: Xi Jiang

https://youtu.be/qKg_rBE1UQM

Allostery is the process by which proteins transmit perturbations caused by the binding effect at one site to another distal site. Allostery is considered important in regulating protein's activity. Drug development depends on the understanding of allosteric mechanisms, especially the identification of allosteric sites, which is prerequisite in drug discovery and design. Many computational methods have been developed for allosteric site prediction using pocket features and dynamics information. Here, we provide a novel ensembled model, consisting of eXtreme gradient boosting (XGBoost) and graph convolutional neural network (GCNN) to predict allosteric sites. Our model can learn both physical properties and topology structure without any prior information and exhibited good performance under several indicators. Prediction results have shown that 84.9% of allosteric pockets in the testing proteins appeared in the top 3 positions. An interactive and quick-response server, PASSer: Protein Allosteric Sites Server (https://passer.smu.edu), is provided to facilitate further analysis in drug discovery.

Hao Tian
Program: PhD in Theoretical and Computational Chemistry
Faculty mentor: Peng Tao

Claire Trotter: Reduced resting beat-to-beat blood pressure variability in multiple sclerosis

Posted on

Co-authors: Alex D. Smith, Ben E. Young, Mu Huang, Dustin R. Allen, Paul J. Fadel and Scott L. Davis

https://youtu.be/db5yGxbEgRk

Multiple Sclerosis (MS) is an autoimmune disease associated with increased cardiovascular risk. Greater resting beat-to-beat blood pressure variability (BPV) is a known predictor of cardiovascular risk. Therefore, we hypothesized that resting beat-to-beat BPV is increased in individuals with MS compared to matched healthy controls. In 7 patients with relapsing-remitting MS (2 males) and 7 sex-, age-, and weight-matched healthy controls, beat-to-beat blood pressure (Finometer) was recorded during 10 minutes of quiet supine rest. Individuals with MS had similar resting mean blood pressure (BP) compared to healthy controls (P= 0.736), however the BP standard deviation (SD) and coefficient of variation (CV) was less in MS (BP SD; MS: 3.2 +/- 0.2 vs. CON: 4.0 +/- 0.2, P=0.022 and BP CV; MS: 3.8 +/- 0.3 vs. CON: 4.7 +/- 0.2, P=0.025 ). Similarly, mean resting systolic blood pressure (SBP) was not different between MS and healthy controls (P=0.207) but the SBP SD and CV was less in MS (SBP SD; MS: 4.7 +/- 0.4 vs. CON: 6.6 +/- 0.5, P=0.013 and SBP CV; MS: 4.3 +/- 0.4 vs. CON: 5.8 +/- 0.4, P=0.033). In contrast, there was no difference in the DBP SD or CV between the two groups (P= 0.321 and P=0.295; respectively). Contrary to our hypothesis, individuals with MS exhibited reduced resting beat-to-beat BPV compared to healthy controls which may be related to altered autonomic function.

Claire Trotter
Program: PhD in Education-Applied Physiology
Faculty mentor: Scott Davis

Francesco Trozzi: UMAP as Dimensionality Reduction Tool for Molecular Dynamics Simulations of Biomacromolecules

Posted on

https://youtu.be/J3CfmaX6vXg

Proteins are the molecular machines of life. The multitude of possible conformations that proteins can adopt determines their free energy landscapes. However, the inherently high dimensionality of a protein free energy landscape poses a challenge on the rationalization of how proteins perform their functions. For this reason. dimensionality reduction (DR) is an active field of research for molecular biologists. The Uniform Manifold Approximation and Projection (UMAP) is a dimensionality reduction method based on a fuzzy topological analysis of data. In the present study, the performance of UMAP is compared to other popular dimensionality reduction methods such as t-Distributed Stochastic Neighbor Embedding (t-SNE), Principal Component Analysis (Analysis (PCA), and time-structure Independent Components Analysis (t-ICA) in context of analyzing Molecular Dynamics simulations of the circadian clock protein Vivid. A good dimensionality reduction method should accurately represent the data structure on the projected components. The comparison of the raw high-dimensional data with the projections obtained using different DR methods, showed the superiority of UMAP compared to linear reduction methods (PCA, t-ICA) and comparable performance with t-SNE, thus far the state-of-the-art method.

Francesco Trozzi
Program: PhD in Theoretical and Computational Chemistry
Faculty mentor: Peng Tao

Menglin Wang: Advanced Capacitors for Future Power Conversion Systems

Posted on

https://youtu.be/11AOmLE9dpo

Capacitors are critical for voltage source converter functionality. DC-link capacitors are known to have reliability issues. Therefore, the overall goal of this program is to improve the understanding of high voltage breakdown in high dielectric constant inorganic ceramic capacitors. This year the focus has been on improving the understanding of resistance degradation in BaTiO3 at elevated temperature, using single crystal BaTiO3 as a model system to understand the impact and control of oxygen vacancy migration to maximize the long-term reliability of high voltage inorganic multi-layer ceramic capacitors.

Menglin Wang
Program: PhD in Electrical Engineering
Faculty mentor: Bruce Gnade

Min Wang: Instructional Technology to Support Students’ Mathematical Problem-Posing

Posted on

Co-author: Candace Walkington

https://youtu.be/aRPhgmNoeeE

Problem-posing is an instructional activity that has been suggested to be beneficial for students' mathematical learning. However, the gap between problem-posing research and classroom implementation remains. The purpose of this presentation is to demonstrate how instructional technology can be integrated in mathematics classrooms to support students' problem-posing in different contexts. In this project, students were instructed to create geometry proof problems using the Hidden Village motion capture game, Algebra word problems using the ASSISTments web-based platform, and general mathematical problems based on students' surroundings using the online walkSTEM Gameboard. In addition, this presentation discusses students' learning behaviors, problem-posing performances, and dispositions toward mathematics when participating in these problem-posing activities.

Min Wang
Program: PhD in Education
Faculty mentor: Candace Walkington

Sharyl Wee: Emerging Adults’ and Parents’ Perceptions of Supportive Parenting: Associations with Family and Parent-Child Relationship Quality

Posted on

Co-authors: Naomi Ekas, Chrystyna Kouros

https://youtu.be/Z7FWiSsmcC0

Parents' and children's reports of parenting are often incongruent, with parents reporting themselves more favorably. The extent to which these discrepancies in perceptions of parenting predict family relationship quality later in development has not been extensively studied. 79 emerging adults and their caregiver completed the Supportive and Unsupportive subscales of Coping with Children's Negative Emotions. EAs completed the Family Environment Scale and the Inventory of Parent and Peer Attachment Scale. Multiple regression models showed greater discrepancies between EAs' reports and parents' reports of parent' unsupportive responses predicted lower current family relationship quality b=-1.94,SE=0.76,p=.01. A main effect of EAs' perceptions of supportive parent responses to their negative emotions in childhood predicted higher levels of EA-reported parent-child relationship quality b=5.49,SE=1.52,p=.001. The results suggest that when EAs and parents agree that the parent was unsupportive of their child's negative emotions in childhood, EAs rated the family relationship quality as worst. Moreover, when EAs remembered their parents as responding supportively to their negative emotions in childhood, they rated their current relationship with their parents better. Results stress the implications of parents' and youths' perceptions of parenting on family relationship quality.

Sharyl Wee
Program: PhD in Psychology
Faculty mentor: Chrystyna Kouros

Ann Marie Wernick: Coaching in the time of coronavirus 2019: how simulations spark reflection

Posted on

Co-authors: Jillian Conry & Paige Ware

https://youtu.be/yCHaUUOkuSE

This study investigates how debrief conversations unfold during virtual coaching sessions that provide embedded opportunities to practice teaching within a mixed reality simulation (MRS). We examine how teacher and coach topical episodes function (agreeing, explaining, clarifying, probing, recapping, reflecting and suggesting) to activate reflection as part of virtual coaching. Grounded in Vygotsky's sociocultural theory and the belief that learning is collaborative and impacts how pre- and in-service teachers construct knowledge, this exploratory case study draws on insights from 15 graduate students (5 pre-service teachers (PSTs) and 10 in-service teachers (ISTs)) who participated in virtual coaching with embedded practice opportunities. Data sources were video recordings and transcripts of 15 virtual coaching sessions, and one-on-one post coaching interviews. Coding categories were determined through the constant comparative analysis method. Findings indicate that MRS provides an immediate context for reflection, which guided the debrief conversations. Additionally, functions occurred with varying frequency among PSTs and ISTs, and across both groups, probing questions often led directly to reflecting and recapping the shared simulation context. In times of remote teaching, like during coronavirus 2019 (COVID-19), opportunities to simulate clinical experiences become vital.

Ann Marie Wernick
Program: PhD in Education
Faculty mentor: Annie Wilhelm

Doran Wood: The Value of a Multistage Dynamic Approach for Radiation Therapy Planning

Posted on

Co-authors: Sila Çetinkaya, Harsha Gangammanavar

https://youtu.be/9RUaZPAkGs0

The goal of intensity modulated radiation therapy is to distribute a prescribed dose of radiation to cancerous tumors while sparing the surrounding healthy tissue. Current approaches implement a radiation plan such that the prescribed tumor dosage is divided equally and delivered through several treatment sessions. This equally distributed (uniform) approach involves solving an optimization problem at each treatment session without consideration of the future sessions or tumor evolution. Herein, we develop a generalization of the uniform formulation that does not automatically assume equal session tumor prescriptions (nonuniform) and also takes future decisions into consideration. This nonuniform multistage framework allows for a natural connection between treatment sessions as well as consideration for sources of uncertainty due to tumor evolution. For the proposed formulation, a sequence of prostate cancer scans provide numerical results revealing drastic improvement in tumor delivery precision while using a total dosage no more than the current practice methods.

Doran Wood
Program: Ph.D in Operations Research
Faculty mentor: Sila Çetinkaya

Bin Xia: Rheology of Particulate Suspensions in 3-D Printing

Posted on

Co-authors: Paul Krueger

https://youtu.be/FTlzRrGvPBg

Multi-functional 3-D printing has been developing quickly and is playing an increasingly vital role in manufacturing technologies. To satisfy the requirements of different functionalities, particulate composites have been widely utilized in this area. These types of materials are usually formulated with different functional particles and matrix materials such as polymer melts and silicone. The materials are particulate suspensions during formulation and printing, and their rheology is a key factor for the processing. This work will concentrate on the suspension rheology in capillaries scaled appropriately for 3-D printing applications (around 1 mm ID). The impact of particle volume fraction and the ratio of the capillary ID to the particle size were analyzed both theoretically and experimentally. Using the results of this investigation, the 3-D printing process based on particulate composites can be optimized, defects can be avoided, its efficiency and quality can also be improved.

Bin Xia
Program: PhD in Mechanical Engineering
Faculty Mentor: Paul Krueger

Xin Yang: Kernel Independent Treecode Accelerated Kernel Smoothing

Posted on

https://youtu.be/HIP3Wghr8vI

A kernel-independent treecode (KITC) is presented for fast summation of particle interactions. A regular treecode algorithm computes particle-cluster or cluster-cluster interactions instead of particle-particle interactions. The KITC uses barycentric Lagrange interpolation instead for the far-field approximation when particle and cluster are well-separated. It reduces the operation count and therefore it is more efficient.

Xin Yang
Program: PhD in Mathematics
Faculty mentor: Weihua Geng

Yuqui Ian Yang: A decentralized sparse fare splitting algorithm

Posted on

https://youtu.be/hAgmXErY-xI

A fare splitting algorithm that is guaranteed to remove all redundant transactions is proposed. We proved the equivalence of the non-existence of redundant transactions and the minimization of the total transaction amount in the system. We also showed that although the resulting transaction amount is unique, the final transaction graph is not. By selecting a basic feasible solution, however, we can achieve a sparse solution.

Yuqui Ian Yang
Program: PhD in Biostatistics
Faculty mentor: Daniel Heitjan

Mai Zaru: Storybook Reading Practices for Children from Low Income Families

Posted on

https://youtu.be/Yp5Pt_8EG5s

This literature review describes effective features of storybook reading practices captured across three decades. The selected studies consist of (a) randomized control trials and quasi-experimental studies, (b) participants from low socioeconomic backgrounds, and (c) an age range of 2 to 7 years old. The selection of studies began with an exploration of two education databases and later cross-referenced with a manual search of storybook reading interventions approved by the Institute of Educational Science (IES). In the seven well-cited studies was a collection of classics published as early as the 1990s, with a total of 735 students, researchers reported the use of similar video-training techniques across. While read aloud interventions were found to have profound impacts on students' achievement across different types of implementers (teacher, parents, and researchers), many students remained unresponsive to storybook reading interventions. Finally, the differences detected across this small scope of studies made it challenging to compare their methodological rigor; however, implications and directions for future research are described in greater depth.

Mai Zaru
Program: PhD in Education
Faculty mentor: Stephanie Al Otaiba

Wenzhong Zhang: FBSDE based deep neural network methods for solving high-dimensional quasilinear parabolic PDEs

Posted on

https://youtu.be/6VSy90uB3Xg

We propose forward and backward stochastic differential equations (FBSDEs) based deep neural network (DNN) learning algorithms for the solution of high dimensional quasilinear parabolic partial differential equations (PDEs). The algorithms relies on a learning process by minimizing the path-wise difference of two discrete stochastic processes, which are defined by the time discretization of the FBSDEs and the DNN representation of the PDE solutions, respectively.

Wenzhong Zhang
Program: PhD in Mathematics
Faculty mentor: Wei Cai

Student Presentations – Research Days 2021

Posted on

Student presentations appear below.

These are listed in alphabetical order by last name of the lead student author (click “Older posts” to keep scrolling). You can use the search function to find a particular person, or look under “Project categories” to browse by discipline.

Comments are allowed and encouraged!