Environmental and Occupational Health Sciences Institute
Rutgers University Community RepositoryNot a member yet
58345 research outputs found
Sort by
Development and validation of the positive rumination in disordered eating questionnaire
No full textAnorexia nervosa (AN) has low rates of early detection and prevention, in part because mechanisms driving the onset of AN are poorly understood. It is unclear how dieting, a commonplace behavior, turns into a full-blown eating disorder (ED) for a subset of individuals. Positive rumination about ED behaviors (e.g., “I feel powerful when I suppress my appetite”) may heighten reward responsivity and motivate the persistent engagement in ED behaviors during the onset of AN. The ‘Positive Rumination in Disordered Eating’ (PRIDE) questionnaire was developed to assess these cognitive mechanisms amplifying ED-related reward. 265 individuals with clinically significant ED pathology and 592 college students completed self-report surveys containing the PRIDE and other measures of psychopathology at baseline and two-month follow-up. Exploratory and confirmatory factor analysis revealed a two-factor structure characterized by Reward Anticipation and Reward Experiencing. The final 14-item scale had good model fit (SRMR=.05, RMSEA=.08, CFI=.95, TLI=.94) in the ED sample. The PRIDE showed excellent internal consistency (α=.93), good convergent/discriminant validity, and strong predictive validity for ED cognitions and behaviors at two-month follow-up (r=.80, p<.001) in the ED sample. These results were replicated in the college sample. Positive rumination may play an important role in the development of AN, but research is limited by the lack of ED-specific measures capturing this construct. The PRIDE is a novel measure that can be used in research and clinical settings to increase understanding of understudied reward processes, and to improve detection of prodromal AN symptoms.M.S.Includes bibliographical reference
Rational design and synthesis of bismuth-doped metal borate and defect-rich metal oxide catalysts for water splitting and chemical conversions
No full textAs the world grapples with the harmful effects of climate change due to the excessive use of fossil fuels that emit greenhouse gases, sustainable and carbon-neutral energy sources are urgently needed. Hydrogen produced from water is considered the potential fuel that can replace our current dependence on fossil fuels. The hydrogen production can be achieved by water electrolysis or photocatalytic water splitting. The water electrolysis involves two half-reactions - the hydrogen evolution reaction (HER), which produces hydrogen, and the oxygen evolution reaction (OER), which produces oxygen. Among these two reactions, the OER is kinetically sluggish and is detrimental to the overall efficiency of water electrolysis to produce hydrogen. In the case of photocatalytic water splitting, excellent photocatalysts that can work in the visible region of sunlight with high efficiency are still lacking. Thus, catalysts that are low cost and abundant and have high catalytic activity are required for producing hydrogen from water. This thesis explores doping and introducing defect sites as strategies to improve the catalytic activity of existing catalysts for water splitting. Furthermore, sustainable synthesis of catalysts for chemical conversions using ionic liquids as reaction media is also studied.In Chapter 1, a broad overview of climate change and how it is fueled by our fossil fuel consumption are discussed. Since the hydrogen economy is one of the potential ways forward to mitigate climate change, the production of hydrogen via water electrolysis and photocatalytic water splitting are reviewed in detail. As excellent catalysts are required to produce hydrogen from water, various strategies to design and develop such catalysts are also presented. Finally, brief overviews of metal borates, transition metal oxides, titanium dioxide, and zeolitic materials, which are parts of the studies in this thesis, are also included.
In Chapter 2, a series of Bi-doped cobalt borates are synthesized, and their electrocatalytic activities towards OER in alkaline media are tested. Here, we demonstrate that the incorporation of oxophilic p-block metals such as Bi into cobalt borate can enhance its electrocatalytic activity towards OER. Furthermore, we show that pyrolyzing the catalyst under various temperatures enhances its catalytic activity because it increases the synergistic interactions between Bi with Co and B atoms. The best electrocatalytic performance is obtained when the Co-to-Bi ratio in the catalyst is 9:1, and the pyrolysis temperature is set at 450 oC.
In Chapter 3, bismuth’s ability to enhance the electrocatalytic activities of first-row transition metal oxides towards acidic OER is demonstrated. First, a series of Bi-doped cobalt oxides are synthesized in situ on fluorine-doped tin oxide (FTO) glass slides and are then tested as electrocatalysts for acidic OER. In this study, the incorporation of Bi into cobalt oxides is found to not only improve their electrocatalytic activity towards acidic OER but also not compromise the durability of the catalysts. The best catalytic performance is obtained when the catalyst is synthesized with a Co:Bi ratio of 9:1. The experiments and density functional theory-based calculations indicate that the Bi incorporated into cobalt oxide can act as catalytic sites during acidic OER. The study is extended further to understand the versatility of Bi in improving the catalytic activity of other transition metal oxides by making a series of Bi-doped iron oxides and nickel oxides on FTO. The results demonstrate that Bi can also improve the catalytic activities of these metal oxides towards acidic OER.
Chapter 4 presents the synthesis of defect-rich TiO2 or reduced TiO2 (which is also called TiO2-x) materials using polyaniline-derived mesoporous carbon (PDMC) as a hard template and the investigation of their photocatalytic activities towards hydrogen production from water. Here, specifically, titanyl sulfate, which is an industrially relevant titania precursor, is used to make the defect-rich TiO2 materials, which makes the work scalable and commercially viable in the end. The level of defect sites in the materials can be varied by varying the TiO2 loading in the PDMC template. These defect-rich TiO2 materials exhibited unusual charge-discharge phenomena when exposed to slow electron beams. Preliminary studies also show that they are active towards photocatalytic hydrogen production from water.
In Chapter 5, the synthesis of aluminophosphates (AlPOs) using double salt ionic liquids (DSILs) prepared from imidazole-based ionic liquids (ILs) with fast resistive conventional and microwave heating is demonstrated. Notably, by using DSILs as both solvent and structure-directing agents, highly crystalline AlPOs are synthesized without the use of corrosive mineralizers such as hydrofluoric acid (HF). Additionally, the composition of DSILs is found to dictate the morphology of the obtained AlPO crystals. The fast resistive conventional heating, which mimics microwave-based heating, also enables rapid crystallization of AlPOs.
In summary, this dissertation explores (i) the enhancement in the catalytic activities of metal borates and first-row transition metal oxides by doping them with the p-block metal bismuth for electrocatalytic OER, (ii) the improvement in the photocatalytic activities of TiO2 by introducing defect sites in them, and (iii) the synthesis of catalytically active AlPOs using DSILs through a facile synthetic route.Ph.D.Includes bibliographical reference
Structural studies of hiv-1 reverse transcriptase inhibition, drug resistance, and function: novel inhibitor discovery and nnrti design, second-strand initiation mechanisms, and host-factor complex formation
No full textHIV-1 reverse transcriptase (RT) is the enzyme that converts the single-stranded viral RNA genome into double-stranded DNA, which is irreversibly integrated into the host chromosome. HIV is the etiological agent of acquired immunodeficiency syndrome (AIDS), which kills ~630,000 people each year, making effective treatment and prevention of critical. RT is the target of nearly half of the FDA-approved antiretroviral therapies. Despite decades of research, many questions remain regarding drug resistance, fundamental mechanistic understanding, and the role of host factors in reverse transcription. We leverage structural biology (X-ray crystallography, cryo-EM) to answer these questions. First, we describe the rational design and structure determination of novel non-nucleoside reverse transcriptase inhibitors (NNRTIs) that potently inhibit a clinically significant drug-resistant triple-mutant RT. These inhibitors show no phenotypic cross-resistance with FDA-approved NRTIs and some NNRTIs, indicating the potential for synergistic use in antiretroviral therapy. Next, we determined the first known structures of an HIV-1 second-strand reverse transcription initiation complex. These complexes are unique in their selective use of a polypurine tract (PPT) RNA primer, which, unlike typical RNA, is not cleaved by RT during first-strand synthesis, making it a unique substrate for second-strand initiation. Moreover, they are uniquely susceptible to drug inhibition relative to other functional states. Our structures of the PPT RNA, pre-incorporation, and NNRTI-bound second-strand initiation complexes reveal a unique architecture. In these structures, the primer adopts unusual placement and induces the opening of the cryptic NNRTI binding pocket, even in the absence of an NNRTI. Together, these structures provide a basis for understanding the unusual sensitivity of second-strand initiation to NNRTI inhibition and fill a critical mechanistic gap in our knowledge of the reverse transcription cycle. Finally, we characterize the binding of HIV-1 RT to host factor eukaryotic translation elongation factor 1 alpha 1 (eEF1A1), which is highly abundant in the cell and required for reverse transcription in vivo as part of the reverse transcription complex (RTC). Preliminary cryo-EM experiments provide insight into the spatial arrangement of the RT/eEF1A1 complex. Disruption of the binding interface by small molecules is an attractive means of developing a new class of HIV-1 inhibitors. These three projects contribute significantly to our fundamental understanding of HIV-1 reverse transcriptase structure, function, and inhibition and advance our collective knowledge of this incredible molecular machine of HIV.Ph.D.Includes bibliographical reference
Analysis and optimization of storage tank design for variable frequency drive compressor
No full textCompressed Air Systems (CAS) are prevalent in industrial applications due to their flexibility, reliability, and ability to power various tools and equipment. CAS usually consists of storage tanks to buffer fluctuations in air demand, helping to maintain stable pressure and reduce the frequency of compressor cycling. Conventional on/off controllers in CAS with tanks to maintain air pressure is an efficient solution but comes with extra expenses, space needs, and maintenance requirements. This research aims to investigate systems with Variable Frequency Drive (VFD)-controlled air compressors, if operated without or with minimal storage tank, can maintain the necessary pressure stability in real-time industrial applications. As the VFD can change the compressor's speed in real-time according to feedback from the demand side, and if it can react quickly to changes in the demand, there is the possibility of having a small or no tank in such systems. The mathematical modeling of air compressors and tanks is created to understand the storage requirements more accurately. The study also investigates the potential optimization of the setpoint of the VFD-controlled compressed air system to save energy. The outcome of this study will provide valuable insights for optimizing the size of tanks for a more flexible, space-saving, and potentially cost-effective VFD-controlled CAS.M.S.Includes bibliographical reference
Machine learning and automation for data-driven design of polymer biomaterials and bioformulations
No full textSynthetic polymers offer exemplary materials for a wide range of applications in drug delivery, tissue engineering, protein stabilization, and mimicry of biologic molecules. This enormous potential of synthetic polymers ultimately stems from polymers capacity for chemical and architectural diversity as they have many tunable parameters including chemical composition, chain length, valency, sequence, tacticity, and architecture. While these represent the properties of a single polymer, additional modulation and diversity is achieved by post polymerization modifications such as functionalization or conjugation. Currently, methods for designing polymer biomaterials involve either a rational design approach based on human expertise and experience, high-throughput (HTP) screening experiments, or a combination of both. Rational design relies on scientists' knowledge and experience to predict optimal chemical properties for a specific goal. However, this approach can be difficult, as it demands extensive knowledge of polymerization chemistry to anticipate complex interactions and synergies that influence the behavior of these complex macromolecules. Further, much of the available chemistries for polymers remain un-explored as the chemical diversity of synthetic polymers is practically infinite. Thus, techniques and approaches that enable HTP and efficient synthesis of polymers, as well as strategies to efficiently explore polymer designs are hypothesized to be critical in identifying synthetic polymers for advanced material applications. In this dissertation, we hypothesize that the pairing of high throughput automated polymer synthesis with machine learning strategies can be utilized to efficiently perform data-driven design of polymer biomaterials. To support this hypothesis, we propose the idea of a Biomaterials Acceleration Platform (BioMAP) in which laboratory robotics and automation is coupled with machine learning (ML) and artificial intelligence (AI) to optimize biomaterials and bioformulations in an efficient, high-throughput, and potentially fully autonomous manner. While a full BioMAP remains unrealized in the field, here we have both developed a platform for autonomous high-throughput polymer synthesis and combined this synthetic capacity with machine learning strategies to optimize polymer designs for enzyme stabilizing polymer-protein hybrids (PPHs) and the design of single chain nanoparticles (SCNPs), two highly complex and diverse biomaterial challenges.Our first chapter involves developing automated tool to develop polymer materials. In biological and medical contexts, polymers have exhibited remarkable utility in drug delivery systems, tissue engineering, and protein stabilization, underscoring the importance of elucidating the intricate relationships between polymer structure, composition, and function. To this end, HTP techniques have emerged as a powerful tool for deconvoluting the complex relationships governing material behavior, enabling the discovery of novel materials and mechanisms. However, high-throughput synthesis of polymers has been historically limited by the oxygen intolerance of most polymerization chemistries. Reactions generally required large sample volumes and inert atmospheres provided by glove boxes, sparging with inert gases such as argon and nitrogen, or freeze-pump-thaw cycling. With the recent advent of oxygen tolerant polymerization techniques that provide chemical mechanisms to sequester and consume oxygen, it has become possible to perform polymer chemistries at small volumes (<300 µL) and in some cases in open-vessels. Leveraging these discoveries here, we: (1) adapted a Hamilton MLSTARlet liquid handling robot to carry out oxygen tolerant photoinduced electron/Energy transfer reversible addition–fragmentation chain transfer (PET-RAFT) and enzyme-assisted reversible addition-fragmentation chain transfer (Enz-RAFT) by custom python scripts, and (2) demonstrate the capacity to perform multi-step synthesis such as block-copolymer synthesis and automated postpolymerization functionalization.
The second chapter of this dissertation is to leverage this capacity for HTP automated polymer synthesis to design highly stable polymer-protein hybrids (PPHs). While a myriad of approaches have been used to stabilize proteins; abiotic environments where proteins are exposed to extreme temperatures, organic solvents, or pHs remain largely challenging domains to maintain protein structure and activity within. Polymer-protein hybrids in which a copolymer is tailored to protein surface chemistry, have demonstrated remarkable success in this regard with examples such as retaining hours-long enzyme activity in pure solvents such as toluene. However, identifying these polymers whether by rational design or high-throughput screening is challenging due to the near infinite combinatorial design space. In this chapter we investigated coupling automated high throughput polymer synthesis with active machine learning (ML) to drive the design of enzyme specific-PPHs. Utilizing a “Design-Build-Test-Learn” strategy, we iteratively synthesized polymer candidates, characterized their ability to maintain enzyme activity after thermal stress, and trained machine learning models on the resulting data to drive the design of new candidate polymers. Remarkably, for every enzyme evaluated we: (1) identified high performing PPHs capable preventing significant denaturation above enzyme melting temperatures, (2) identified unique polymer chemistry preferences required for stability in each enzyme case, (3) obtained quantitative structure activity relationships by SHAP, and (4) observed chaperone refolding behavior of key PPHs through biophysical characterization.
The third chapter of this dissertation focuses on developing a data-driven approach to improve the efficiency of identifying single chain polymer nanoparticles (SCNPs). Single-chain polymer nanoparticles are synthetic molecules that can self-assemble into complex structures, similar to proteins. SCNPs have gained significant interest in the polymer science community as drug delivery vehicles, diagnostic and biosensing materials, anti-bacterial applications, or as biological mimetics for proteins and enzymes. However, given the complexity of SCNP design in which polymers must undergo hydrophobic collapse and remain stable through charge-charge interactions, hydrogen bonding, steric, and Vander walls forces, rational design is extremely challenging. Therefore, in this chapter, we hypothesized that HTP automated polymerization and machine learning could be utilized to learn SCNP structural design features. Leveraging HTP automated PET-RAFT, we synthesized a set of over 1000 polymers and polymer conjugates, and performed HTP biophysical characterization through dynamic light scattering (DLS) and small angle x-ray scattering (SAXS). Combining this data with polymer chemical featurizations, we then trained evidential neural networks (ENNs), a probabilistic machine learning model to learn key parameters for SCNP structural design, including Porod exponent and Rg. After model training, we attempted to predict synthesizable polymer compositions with significant compactness (Rg/Rh < 0.80) and flexibility (Porod exponent between 3.0-4.0). This approach successfully identified 10 novel SCNPs that exhibited high compactness and flexibility, out of 30 synthesized polymers used to validate the model.
In conclusion, this work encompasses initial case studies on how the combination of HTP automated polymer synthesis and machine learning strategies can be used for tailored biomaterial development. Advances here have demonstrated the potential to use these techniques to efficiently discover polymers as effective protein stabilizers, and compact SCNPs. Further, beyond simply identification of materials, these approaches have enabled us to uncover design rules specific to each biomaterials challenge. In the future, these methods could be expanded to accelerate research across a wide berth of biomaterial and bioformulations challenges.Ph.D.Includes bibliographical reference
Parkinson’s Disease gene, synaptojanin1, regulates dopamine neuron function
No full textParkinson’s disease (PD) is a prominent neurodegenerative disease characterized by dopamine signaling dysfunction, resulting in the deterioration of dopaminergic neurons in the substantia nigra. A fundamental question in the field is why dopaminergic neurons are susceptible in genetic PD conditions. Mutations of SYNJ1 (synaptojanin1/Synj1) are associated with PD or Parkinsonism. Studies from our lab and others have shown that Synj1 deficiency in mice leads to dopamine (DA) neuron-specific synaptic defects such as impaired synaptic vesicle endocytosis and abnormal dopamine transporter (DAT) clusters. Dysregulation of DAT trafficking alters brain DA levels, which may be associated with early synaptic changes in PD. Despite mounting evidence for the role of Synj1 in synaptic function, specific roles of Synj1 on DAT trafficking have not yet been reported. In this dissertation I conducted several studies combining in vitro and in vivo analyses to understand 1) Synj1 mediated pathways in DAT trafficking and 2) the role of Synj1 in DA signaling and PD pathogenesis. I found that overall, DAT expression was altered in Synj1+/- mice. Interestingly the steady state surface DAT (sDAT) is reduced in the soma and axons of cultured Synj1+/- dopamine neurons. Consistently, Synj1+/- mice exhibit blunted pharmacosensitivity to a DAT stimulant. In N2a cells expressing human SYNJ1 cDNAs harboring PD mutations, we found significantly decreased sDAT expression associated with impaired 5’-phosphatase activity of SYNJ1. Furthermore, we showed that axonal sDAT is maintained by the PI(4,5)P2-PKCβ signaling pathways. These results revealed molecular underpinnings of Synj1-regulated sDAT trafficking in the Synj1 haploinsufficient mouse.
To further understand the in vivo role of Synj1 in DA neuron function and dopaminergic degeneration, we generated a conditional deletion (cKO) mouse specifically deleting Synj1 in the DA neurons. A battery of locomotor behavioral studies was carried out for a cohort of sex matched Synj1 cKO and control littermate mice across different ages from 3 to 12 months. I found that cKO mice exhibit age-dependent impairments in locomotor function, reduced DA and enhanced DA turnover in the dorsal striatum at 3 and 12 months of age.
Taken together, my dissertation work enlightened pathogenic cellular signaling downstream of Synj1 deficiency and may be valuable to a multitude of DAT-related disorders including Parkinson’s disease.Ph.D.Includes bibliographical reference
DEP Bulletin: Volume 44 Issue 10
No full textThe DEP bulletin is a semi-monthly publication of the New Jersey Department of Environmental Protection. It contains a list of construction permit applications recently filed or acted upon by the Department along with a calendar of events of interest, a schedule of public hearings, and Environmental Impact Statements acted upon during the period
The effect of urbanization and anthropogenic disturbance on a widespread snake species, Dekay's brown snake (Storeria dekayi)
No full textThe rapid urbanization of the earth’s terrestrial land surface over the last century has led to habitat loss and fragmentation worldwide, posing a major threat to biodiversity. To date, the impact of urbanization has already been evaluated on a variety of organisms regarding their abundance, behavior, physiology, phenology, and genetics, which has yielded results of varying response directionalities and intensities. Studies of such have been conducted primarily on a small set of taxa, namely birds, arthropods, and plants. Little is known about how other taxa, such as elusive and reclusive reptiles respond to urban environments. Dekay’s brown snake (Storeria dekayi) is a small, semi-fossorial natricid snake. Storeria dekayi is a habitat generalist commonly found throughout central and eastern North America. Because it is frequently encountered in developed areas, S. dekayi makes a good model organism to examine the effect of urbanization on reptiles. In this dissertation, I used a combination of ecological, GIS, and genomic data, to explore the effect of urbanization on S. dekayi. In Chapter 1, I constructed ecological niche models to investigate the importance of anthropogenic disturbance in shaping the current S. dekayi distribution, and to predict the future S. dekayi range shift. I found no significant effect of anthropogenic disturbance on the current distribution, but potential range expansion in the future due to anthropogenic climate change. In Chapter 2, I examined whether S. dekayi populations along a rural-urban gradient displayed morphological differences using a combination of linear and geometric morphometrics. I uncovered an elevated degree of morphological divergence and a lowered magnitude of sexual dimorphism among populations dwelling in more urbanized habitats. In Chapter 3, I explored the natural history of a S. dekayi population in an urbanized setting concerning its population abundance and fall migration. S. dekayi in this urban population were found to occur in high densities. They were also vulnerable to car strikes in fall, and their fall movements were associated with certain roadside environmental features. In Chapter 4, I evaluated whether urban landscapes could shape the genetic diversity and structure of S. dekayi populations by using a genome-wide genetic dataset (ddRADseq data) of S. dekayi sampled along a rural-urban gradient. Though without significant reduction of intrapopulation genetic diversity, I uncovered a certain amount of genetic structuring and reduced migration and diversity rates among the populations, especially for those inhabiting more urbanized areas. Altogether, this dissertation provides valuable information to expand our knowledge of urban biodiversity and to guide future urban conservation management.Ph.D.Includes bibliographical reference
An exploration and analysis of the microstructure, dimensional changes and scattering effects of stereolithography additive manufactured alumina
No full textThis dissertation examines the microstructure and quantitative analytical assessment of porosity, scattering, and lateral dimensional changes in stereolithography (SL), additive manufactured (AM), alumina (Al2O3). The accompanying research is divided into two primary areas of focus: an in-depth analysis of the microstructure of alumina, printed under multiple light intensities, layer sizes and ceramic powder particle sizes including an investigation of quantifying lateral dimensional spread and scattering and its impact on the microstructure. In the analysis of the microstructure, advanced techniques including backscattered and secondary electron imaging, provide for sample imaging of the layers of the AM pieces. Each sample is investigated within each layer as well as the adhesion between layers. This microstructure visible by analysis on a scanning electron microscope demonstrates the correlation between printing parameters and the challenges introduced by scattering and lateral dimensional changes. The result of this investigation and the computational analysis of the properties assists in determining the best parameters for optimal printing of the ceramic materials. The phenomena studied here and the factors that have been quantified provide the opportunity for consideration of potential strategies for optimizing the SLA process including resin composition, adjusting the parameters of the light source, and developing post-processing techniques.
The multi-faceted approach discussed in this study is one pathway to conduct experiments and create characterization techniques to address the challenges of printing and post-processing to ensure the optimization of the printing process based on the unique needs of each resin as well as the design of each finished part.
This study conducts a series of experiments comparing the microstructure, scattering and lateral spread under various experimental conditions. The microstructure of each of these sample conditions are then introduced into a computational model to ensure that predictable and reproducible results are achieved. Additionally, the potential of machine learning algorithms for predicting and compensating for dimensional changes pre-sintering is explored, highlighting the role of real-time adjustments in printing parameters for enhanced accuracy and consistency.
The findings presented here contribute to the field of additive manufacturing, offering insights into optimizing SLA for ceramic resins and paving the way for its application in industries requiring high precision and specific material properties, such as strength, toughness, and corrosion resistance. Additionally, the research presented provides for the use of Artificial Intelligence (AI) in the form of Machine Learning (ML) to assist in quantifying and understanding the changes and differences in each part based on printing and curing parameters. The research also provides opportunities for future investigations into the SLA manufacturing processes and the expansion of this technology and quantitative technique to other ceramic powders.Ph.D.Includes bibliographical reference
Research universities in search of our kind of people: Black women doctoral student socialization
No full textAs research universities become more visibly “diverse,” scholarship must contend with whether inclusion is also a part of higher education’s future. This project examines doctoral education's attitudes, beliefs, and ideologies as a site of persistent exclusionary practices and impetus for enduring change. It has been documented that knowing how to navigate academia's social dynamics is one of the most critical lessons doctoral students learn. Hence, this project focused on the social aspects of doctoral training, seeking to move beyond quantifiable benchmarks. This research project explored the lived experience of Black women doctoral students in Ph.D. programs in the Humanities at research universities. Data collection included semi-structured interviews examining the experience of Black women doctoral students enrolled or formerly enrolled at three R1 public universities in New England, the Mid-Atlantic, and the Southern United States from across 10 disciplines. Three research questions guided this project, covering the perception of academic socialization, relationship building, and the history of the universities as racialized organizations. Three significant findings from my project include (1) the significance of the advisor/advisee relationship; (2) Black women’s experiences were informed by racialized/gendered ideologies; and (3) the (mis)alignment of beliefs between the advisor and advisee complicated relationship dynamics and access to resources.Ph.D.Includes bibliographical reference