TTU DSpace Repository
DSpace is a digital service that collects, preserves, and distributes digital material. Repositories are important tools for preserving an organization's legacy; they facilitate digital preservation and scholarly communication.
Communities in DSpace
Select a community to browse its collections.
- Digitized books and images
- Digital repository featuring a searchable,online collection of scholarly legal research and information
- Digital repository featuring a searchable, online collection of scholarly research and information
Recent Submissions
Cross-sectional survey exploring current intake practices for dogs admitted to animal shelters in Texas: a descriptive study
(2023) Cranford, Mackenzie (TTU); Bing, Abbey (TTU); Cisneros, Alissa (TTU); Carroll, Amber D. (TTU); Porter, Hannah (TTU); Stellato, Anastasia Chiara (TTU)
Introduction: Entering an animal shelter is a stressful experience for dogs that can impair their welfare, adoptability, and shelter staff safety; thus, it is crucial to reduce the stress experienced during intake. This study investigated the current intake practices for dogs admitted in animal shelters in Texas, United States. Methods: To gather data, an online survey was designed and distributed to shelter employees responsible for intake at animal shelters. The survey collected information about examination procedures, the type of information collected from owner-surrenders, as well as the housing environment for the dogs. Results: Survey participants (n = 64) were shelter staff from municipal (59%, 38/64) and private shelters (23%, 15/64) in 47 counties. Handling techniques reported to be used during intake exams varied depending on the dog’s behavior, with participants reporting higher restraint for aggressive dogs and lower restraint for calm dogs. If the dog was displaying fear, participants reported offering food and attention (89%, 47/53), using towel restraint (64%, 34/53) and conducting the exam on someone’s lap (49%, 26/53). In cases of aggression, it was commonly reported to use muzzles (81%, 42/52) and catch poles (77%, 40/52), and shorten the exam (71%, 37/52). After the exam, most reported placing dogs on the adoption floor (45%, 27/60) or placing them wherever space was available (20%, 12/60). Discussion: Results provide descriptive information on current intake procedures and routine handling techniques used in Texas shelters. Future research should explore shelter dog responses to routine handling techniques to support the development of evidence-based protocols during routine intake examinations and procedures.
Associations between dust exposure and hospitalizations in a dust-prone city, Lubbock, TX, USA
(2023) Herrera-Molina, Estrella; Gill, Thomas E.; Ibarra-Mejia, Gabriel; Jeon, Soyoung; Ardon-Dryer, Karin (TTU)
Although it is a growing area of investigation in the Global Dust Belt, only a few population-level studies have evaluated the human health associations of windblown dust in North America. We investigated whether acute, short-term dust exposures (DE), in Lubbock, Texas (a medium-sized, dust-prone city in the southern Great Plains, USA) were associated with significant increases in hospitalizations on the day of the exposure and up to 7 days afterward. We used the distributed lag non-linear models in time series analysis to describe non-linear relationship between response outcomes and the delayed effects of exposure over time. We found that increased relative risks of hospitalizations for multiple conditions were associated with the two DE approaches that occurred between 2010 and 2014. Consistent with prior studies of dust health effects in other cities in North America, we identified increased hospitalization risks in Lubbock due to neurodegenerative, atherosclerosis, renal, respiratory, asthma, mental, stroke, neoplasms, ischemia, hematologic, musculoskeletal, and associated diseases (aggregation of all causes each associated with at least 5% of hospitalizations) at various dust exposure days. Associations were modified by age, gender, day of the week, and holiday effects. As climate change increases water stresses on dryland agriculture and long periods of drought, dust exposures are likely to increase for residents of dryland cities and with it the likelihood of adverse health effects on people with preexisting conditions. Additional investigations are needed for other dust-prone population centers worldwide to document the health effects of dust exposures and investigate their causes.
Lipid modulation contributes to heat stress adaptation in peanut
(2023) Spivey, William W.; Rustgi, Sachin; Welti, Ruth; Roth, Mary R.; Burow, Mark D. (TTU); Bridges, William C.; Narayanan, Sruthi
At the cellular level, membrane damage is a fundamental cause of yield loss at high temperatures (HT). We report our investigations on a subset of a peanut (Arachis hypogaea) recombinant inbred line population, demonstrating that the membrane lipid remodeling occurring at HT is consistent with homeoviscous adaptation to maintain membrane fluidity. A major alteration in the leaf lipidome at HT was the reduction in the unsaturation levels, primarily through reductions of 18:3 fatty acid chains, of the plastidic and extra-plastidic diacyl membrane lipids. In contrast, levels of 18:3-containing triacylglycerols (TGs) increased at HT, consistent with a role for TGs in sequestering fatty acids when membrane lipids undergo remodeling during plant stress. Polyunsaturated acyl chains from membrane diacyl lipids were also sequestered as sterol esters (SEs). The removal of 18:3 chains from the membrane lipids decreased the availability of susceptible molecules for oxidation, thereby minimizing oxidative damage in membranes. Our results suggest that transferring 18:3 chains from membrane diacyl lipids to TGs and SEs is a key feature of lipid remodeling for HT adaptation in peanut. Finally, QTL-seq allowed the identification of a genomic region associated with heat-adaptive lipid remodeling, which would be useful for identifying molecular markers for heat tolerance.
CYBERSLACKING IN THE WORKPLACE: ANTECEDENTS AND EFFECTS ON JOB PERFORMANCE
(2023) Venkatesh, Viswanath; Cheung, Christy M.K.; Davis, Fred D. (TTU); Lee, Zach W.Y.
Employees’ nonwork use of information technology (IT), or cyberslacking, is of growing concern due to its erosion of job performance and other negative organizational consequences. Research on cyberslacking antecedents has drawn on diverse theoretical perspectives, resulting in the lack of a cohesive explanation of cyberslacking. Further, prior studies have generally overlooked IT-specific variables. To address cyberslacking problems in organizations, as well as research gaps in the literature, we used a combination of a literature-based approach and a qualitative inquiry to develop a model of cyberslacking that includes a 2×2 typology of antecedents. The proposed model was tested and supported in a three-wave field study of 395 employees in a U.S. Fortune-100 organization. This study organizes antecedents from diverse research streams and validates their relative impact on cyberslacking, thus providing a cohesive theoretical explanation of cyberslacking. This study also incorporates contextualization (i.e., IT-specific factors) into theory development and enriches the IS literature by examining the nonwork aspects of IT use and their negative consequences to organizations. In addition, the results provide practitioners with insights into the nonwork use of IT in organizations, particularly regarding how they can take organizational action to mitigate cyberslacking and maintain employee productivity.
Performance-Aware Energy-Efficient GPU Frequency Selection using DNN-based Models
(2023) Ali, Ghazanfar (TTU); Side, Mert (TTU); Bhalachandra, Sridutt; Wright, Nicholas J.; Chen, Yong (TTU)
Energy efficiency will be important in future accelerator-based HPC systems for sustainability and to improve overall performance. This study proposes a deep neural network (DNN)-based learning model for execution time and power consumption of workloads across GPUs DVFS design space. Micro-architectural data obtained by running SPEC-ACCEL, DGEMM, and STREAM benchmarks are used for model training. These features are consistent for a workload unaffected by frequency and input size reducing the data required significantly. For real-world applications - LAMMPS, NAMD, GROMACS, LSTM, BERT, and ResNet50 power and time models show 89% - 98% accuracy on NVIDIA Ampere. Multi-objective functions help select optimal frequencies that lower power and minimize performance impact showing maximum energy savings of 27% at a performance loss of 1.8%. The same models trained on Ampere showed an accuracy of greater than 93% on an NVIDIA Volta, thereby demonstrating model portability across architectures.