VTechWorks

VTechWorks provides global access to Virginia Tech scholarship, including journal articles, books, theses, dissertations, conference papers, slide presentations, technical reports, working papers, administrative documents, videos, images, and more by faculty, students, and staff. Faculty can deposit items to VTechWorks from Elements, including journal articles covered by the University open access policy. Email vtechworks@vt.edu for help.

Recent Submissions

Strategic Communication as a Mechanism of Organizational Change: A Mixed-Methods Examination of Self-Efficacy for Transformation and Change Support

Thrash, Julian Earl (Virginia Tech, 2026-11-06)

This three-study sequential explanatory mixed-methods dissertation examines how strategic communication is associated with employees' self-efficacy for transformation and change support during the U.S. Department of Veterans Affairs' implementation of the Non-Expendable Equipment initiative, a large-scale change involving more nationally structured acquisition and ordering processes for non-expendable medical equipment. Instrumental strategic communication, which provides procedures and implementation guidance, and inspirational strategic communication, which emphasizes mission and purpose, are specified as primary communication predictors. Organizational identification is modeled as a proposed boundary condition, and self-efficacy for transformation is positioned as the central employee capability pathway linking communication to change support. Study A provided the primary quantitative test of the hypothesized communication-to-self-efficacy-for-transformation-to-change-support pathway using observed-variable path analysis. Results showed that self-efficacy for transformation was the strongest immediate correlate of change support and that both communication forms were most consistently associated with change support indirectly through self-efficacy for transformation. Moderation by organizational identification was not supported. Study B used directed qualitative content analysis of early implementation narratives to explain how the quantitative pathway was experienced in practice. Study C used guided interviews conducted later in implementation as a temporally separated qualitative cross-section to examine whether those patterns remained visible in a later implementation phase. Findings indicated that later employee responses differentiated into four recurring profiles: capable champions, instrumental performers, stretched idealists, and strained compliers. Taken together, the studies indicate that strategic communication is most strongly associated with change support when it is linked to employees' practical confidence to function effectively within the new change environment. Across the three studies, self-efficacy for transformation emerged as the central employee capability pathway through which strategic communication was associated with change support. Collectively, the findings suggest a recurring sequence in which communication is associated with implementation clarity, clarity is associated with employee capability, and capability is associated with more constructive engagement with organizational change. Keywords: strategic communication; self-efficacy for transformation; change support; organizational identification; mixed methods; organizational change

SAR-Driven Semisynthetic Optimization of Chloramphenicol Analogues for Improved Antibacterial Activity and Resistance Evasion

Shuvo, Suzzudul Islam (Virginia Tech, 2026-11-05)

Antibiotic resistance continues to pose a major global health threat, exacerbated by the limited discovery of new antibiotic classes. This challenge has renewed interest in the redevelopment of legacy antibiotics through semisynthetic strategies that enable targeted structural modification of complex scaffolds. Chloramphenicol, a historically important broadspectrum antibiotic, represents a compelling candidate for such efforts due to its well-defined mechanism of action, synthetic accessibility, and known liabilities associated with toxicity and resistance. This dissertation describes a systematic semisynthetic structure–activity relationship (SAR) investigation of chloramphenicol focused on diversification at the para position of the aryl ring and modification of the C3 functionality. A key enabling advance was the development of an improved reduction platform to access chloramphenicol aniline intermediates, overcoming longstanding limitations associated with low-yielding nitro reductions. This intermediate served as a versatile branching point for the synthesis of multiple derivative classes. Using this platform, three major libraries were developed and evaluated: (i) amide derivatives, which established baseline SAR trends and demonstrated activity comparable to thiamphenicol; (ii) azo-linked derivatives, which in several cases exhibited improved broadspectrum antibacterial activity relative to the amide series; and (iii) p-triazole analogues, which provided further enhancements, with multiple compounds surpassing the potency of chloramphenicol against selected strains. Across these series, systematic SAR analysis revealed the importance of electronic effects, linker identity, and substituent positioning in modulating antibacterial activity, with electron-withdrawing and conjugated systems generally favoring potency. III In parallel, a multistep semisynthetic strategy was developed to introduce a 3-fluoro substituent into the chloramphenicol scaffold, a modification known to confer resistance to chloramphenicol acetyltransferase (CAT)-mediated deactivation. This work addressed key synthetic challenges associated with site-selective fluorination of a densely functionalized natural product and provided access to a series of 3-fluoro analogues across the amide, azo, and triazole scaffolds. Comparative biological evaluation demonstrated that incorporation of the 3-fluoro group frequently enhanced antibacterial activity, particularly among the most potent derivatives, and offered insight into the interplay between para substitution and C3 modification. Collectively, this work establishes an efficient semisynthetic platform for the diversification of chloramphenicol, defines key SAR trends governing antibacterial activity, and demonstrates that strategic combination of para-substitution and C3 fluorination can improve potency and potentially overcome resistance mechanisms. These findings provide a foundation for the rational design of next-generation phenicol antibiotics with improved efficacy against both Gram-positive and Gram-negative pathogens.

Scam Shield: A Hierarchical Framework for Robust and Efficient Adversarial Scam Detection

Chang, Chen-Wei (Virginia Tech, 2026-11-05)

Can we trust large language models (LLMs) to accurately detect scams? This thesis investigates the vulnerabilities of LLMs when facing adversarial scam messages. We created a dataset with fine-grained labels, including original scam messages, adversarial scam messages, and non-scam messages. These labels allow us to examine detection performance across different scam types in addition to the binary classification task. We evaluated GPT, Claude, and LLaMA on this dataset. Under the selected rewriting conditions, the evaluated models miss scam messages even when the deceptive intent is preserved. To address this problem, we propose a Hierarchical Scam Detection System (HSDS). The system combines four lightweight classifiers, Decision Tree, XGBoost, Random Forest, and K-Nearest Neighbors, with a fine-tuned LLaMA 3.1 8B Instruct model. The classifiers first predict whether each message is a scam. When all four agree, the system returns their prediction. When they disagree, the message is sent to the language model. We fine-tuned this model with Low-Rank Adaptation (LoRA) using a domain-specific adversarial dataset. The fine-tuned model achieved higher detection scores with zero-shot prompting than the evaluated few-shot baselines. The full HSDS achieved an accuracy of 0.90 and an F1-score of 0.90, with higher scores than the evaluated proprietary LLM baselines. It also reduced average inference time by approximately 56.7% compared with the fine-tuned model alone. Romance scams remained difficult to detect. Further work could examine whether semantic and contextual features improve detection of these emotionally manipulative messages.

Dynamic integrin expression changes and adhesion-driven mitobiogenesis in ovarian cancer metastasis

Bano, Nazia (Virginia Tech, 2026-11-05)

Ovarian cancer is the 6th leading cause of cancer-related deaths in women, with survival rates of only 31% in metastatic disease. The unique transcoelomic dissemination of ovarian cancer cells includes detachment from the primary site, aggregation into multicellular spheroids, and attachment and invasion of distant tissues. Successful peritoneal metastasis requires cancer cells to adapt to a changing microenvironment, which necessitates coordinated changes in cell adhesion, ligand regulators, and mitochondrial function. Prior studies have demonstrated that aggregation promotes survival by fragmenting mitochondria and reducing metabolic activity, which is reversed by mitobiogenesis, high metabolic activity, and proliferation after adhesion to secondary sites. Here, we investigated the role of integrin-mediated adhesion in regulating mitochondrial and metabolic adaptation at the secondary adhesion site during ovarian cancer dissemination. By modeling the primary adherent, aggregation (spheroid), and adherent spheroid stage under physiologically relevant oxygen and nutrient conditions, we demonstrate that integrin expression is highly dynamic throughout dissemination and is accompanied by coordinated changes of other adhesion molecules, extracellular matrix (ECM) components, and their regulators. Spatial expression analyses revealed that integrins and CD44 exhibit distinct localization patterns within spheroids and were frequently observed in the nucleus, suggesting a novel role in transcriptional regulation that may contribute to survival, cellular plasticity, and metastatic progression. Expression and functional studies identified ITGαVβ1 as key mediators of spheroid adhesion and outgrowth at secondary sites. While mitochondrial activity is suppressed during aggregation, adhesion induces metabolic reprogramming characterized by improved mitochondrial respiration through enhanced mitobiogenesis to support outgrowth and invasion for successful metastasis. This process is driven by signaling through the ITGαVβ1/FAK pathway, confirmed by the pharmacological inhibition of ITGαVβ1 or FAK, which reduced mitochondrial mass, decreased mitochondrial membrane potential, oxygen consumption rate, and inhibited or reduced spheroid outgrowth. These findings establish the ITGαVβ1/FAK axis as a central link between adhesion-mediated signaling and mitobiogenesis and are a promising therapeutic target to limit ovarian cancer metastasis and improve patient outcomes.

Physics-Informed Machine Learning for Predicting Rubber–Surface Friction from Dynamic Friction Tester Measurements: Load–Speed Interaction, Load, Roughness, and Material Response

Hamedi, Sajad (Virginia Tech, 2026-09-11)

The interaction between tires and the road is one of the main factors determining how safely and effectively a vehicle can operate. Predicting this friction, however, is far from straightforward because it depends on several variables, such as structural and material properties of tire, road roughness, tire slip, applied load, temperature, and velocity, that interact in complex and nonlinear ways. Existing models can handle simplified conditions but often fall short when applied to real driving situations. This thesis addresses the problem through physics-informed machine learning. Approximately 600 steady-state friction observations produced by a Dynamic Friction Tester (DFT), collected by Su [8] at the CenTiRe Laboratory at Virginia Tech, were assembled and used to evaluate a range of machine learning techniques. Eight rubber compounds were tested across three standardized abrasive surfaces spanning coarse to fine texture. Random Forest, Support Vector Regression, Artificial Neural Network, and a Hybrid Stacking Ensemble were trained and compared against the measured friction coefficients. Because the measurements were made on abrasive papers rather than on pavement, the friction coefficients span a wider and higher range than a tire encounters in service; the models are therefore calibrated to a DFT-on-abrasive regime rather than to a road regime. The central contribution of this work lies in how the learning problem is posed rather than in the volume of data analyzed. Raw experimental inputs such as sliding velocity and viscoelastic loss modulus are coupled through the excitation frequency relationship 𝜔 ≈ 𝑞𝑉s described in Persson’s viscoelastic friction theory [2,59–61]. The feature basis was therefore re-expressed through two mechanism-guided interaction terms: Fn×Vs, a power-dimensional load–speed interaction, and Ts×E″, a strength–damping coupling formed from the compound’s maximum tensile strength and viscoelastic loss modulus. The physical interpretation of the first term is deliberately limited: Fn×Vs is not the actual frictional power delivered at the interface, which is 𝐹f𝑉s= 𝜇𝐹n𝑉s. Su [8] likewise expresses sliding work using friction force, 𝑊 = 𝐹f𝑠. The second term is also an engineered interaction rather than a derived physical quantity; its units, Pa², are not those of energy or dissipation. Models trained on this reparameterized feature set achieved high accuracy while producing feature-importance rankings that are interpretable in tribological terms. Overall, the study shows that a feature basis built from tribological reasoning yields accurate friction predictions whose importance structure is interpretable in mechanistic terms. A matched raw-feature control was subsequently trained on {Dp, Ts, Fn, Vs, E″, grit} under the same split, preprocessing and hyperparameters, and all four architectures were re-evaluated under leave-one-compound-out validation. The control shows that the composite terms are not universally more accurate: across twenty repeated splits they give the Random Forest a small but reproducible gain (ΔR² = +0.005, p = 0.024), while the raw basis is better for SVR and ANN and indistinguishable for the Hybrid Ensemble. Compound-holdout validation shows that the Random Forest predicts entirely unseen formulations with pooled R² = 0.921, including R² = 0.858 when trained on compounds A–G and tested on the withheld compound H. The claim advanced here is therefore one of interpretability, physical consistency and demonstrated material generalization rather than of universal predictive superiority. The framework is a step toward friction models that can be audited against theory as well as against data.

VTechWorks

Open Access Policy

Theses and Dissertations

Open Textbooks

Communities in VTechWorks

Recent Submissions