Virginia Tech\'s Computational Bioinformatics and Bio-imaging Laboratory (CBIL) is exploring integrative network analysis techniques to identify subnetworks or genetic pathways that contribute to various cancers. Chen et. al. developed a bagging Markov random field (BMRF)-based approach which examines gene expression data with prior biological information to reliably identify significant genes and proteins. Using random resampling with replacement (bootstrapping or bagging) is essential to confident results but is computationally demanding as multiple iterations of the network identification (by simulated annealing) is required. The MATLAB implementation is computationally demanding, employs limited concurrency, and thus time prohibitive. Using strong software development discipline we optimize BMRF using algorithmic, compiler, and concurrency techniques (including Nvidia GPUs) to alleviate the wall clock time needed for analysis of large-scale genomic data. Particularly, we decompose the BMRF algorithm into functional blocks, implement the algorithm in C/C++ and further explore the C/C++ implementation with concurrency optimization. Experiments are conducted with simulation and real data to demonstrate that a significant speedup of BMRF can be achieved by exploiting concurrency opportunities. We believe that the experience gained by this research shall help pave the way for us to develop computationally efficient algorithms leveraging concurrency, enabling researchers to efficiently analyze larger-scale data sets essential for furthering cancer research. / Master of Science

Natural hybridization is reproduction (without artificial influence) between two or more species/populations which are distinguishable from each other by heritable characters. Natural hybridizations among marine fishes were highly underappreciated due to limited research effort; it seems that this phenomenon occurs more often than is commonly recognized. As hybridization plays an important role in biodiversity processes in the marine environment, detecting hybridization events and investigating hybridization is important to understand and protect biodiversity. The first chapter sets the framework for this disseration study. The Cohesion Species Concept was selected as the working definition of a species for this study as it can handle marine fish hybridization events. The concept does not require restrictive species boundaries. A general history and background of natural hybridization in marine fishes is reviewed during in chapter as well. Four marine fish hybridization cases were examed and documented in Chapters 2 to 5. In each case study, at least one diagnostic nuclear marker, screened from among ~14 candidate markers, was found to discriminate the putative hybridizing parent species. To further investigate genetic evidence to support the hybrid status for each hybrid offspring in each case, haploweb analysis on diagnostic markers (nuclear and/or mitochondrial) and the DAPC/PCA analysis on microsatellite data were used. By combining the genetic evidences, morphological traits, and ecological observations together, the potential reasons that triggered each hybridization events and the potential genetic/ecology effects could be discussed. In the last chapter, sequences from 82 pairs of hybridizing parents species (for which COI barcoding sequences were available either on GenBank or in our lab) were collected. By comparing the COI fragment p-distance between each hybridizing parent species, some general questions about marine fish hybridization were discussed: Is there any correlation between genetic similarity and the potential for hybridization in marine fishes? In some particular geographic locations that have the existence of several different hybridization reports, are the species involved in hybridization among those reports all closely related or distantly related? Can any associations between parent species’ similarities and hybrid spots be found?

Lubrication is used during the drilling of aluminum to counter the formation of a built-up-edge (BUE), among other reasons. The elimination of the use of lubricants in drilling of aluminum is important because of the associated high costs of cleaning and disassembly involved in lubrication. The optimal conditions sought in this work include the elimination of the use of lubricants along with the possible attainment of a high material removal rate, which could help in reduction of cost and increase productivity at the same time. BUE has been found to be almost always present in the process of metal cutting at low to moderate speeds. It has been found that a necessary condition for the formation of a BUE is the presence of a negative stress gradient away and normal to the tool rake face. The quantitative equivalence of the effects of temperature and strain rate on flow stress described by the Zener-Hollomon parameter (Z) [5]. The relationship between the Zener-Hollomon parameter (Z) and chip flow stress implies that a negative Z gradient could be considered equivalent to a negative stress gradient. A series of computer simulations with varied cutting conditions were analyzed to determine the combination of machining variables which yielded a low predicted BUE preferably with a high material removal rate. The results presented here include cases which have a low predicted BUE as well as a high material removal rate.

The concept of unconditional self-love is fundamental to many theoretical perspectives in social science and is referred to periodically in the literature. This study addressed the problem that scientific literature refers to unconditional self-love, but does not clearly define it, differentiate it from other types of feelings that comprise self-esteem, measure it, or even attempt to substantiate its existence. The purpose of this study was to define, describe, and measure unconditional self-love so that it may be used in the treatment of problems caused by low self-esteem. To achieve this purpose, the concept of self-love was defined in relation to self-liking, separated from the other components of self-esteem, and operationalized through the creation of the Unconditional Self-Liking (USL) model and the Unconditional Self-Liking (USL) scale. The USL scale is a self-report questionnaire that simultaneously measures two variables in relation to one another. This was done using the intersect of the variables of personal success and self-liking as a measure of unconditionality across nine areas of self-identity. The USL model allows for the measurement and description of four primary types of self-liking: (a) conditional self-likers who like themselves only when they succeed, (b) unconditional self-dislikers who dislike themselves even when they succeed, (c) conditional self-dislikers who dislike themselves when they fail, and (d) unconditional self-likers who like themselves even when they fail. The USL scale was administered to a convenience sample of 164 undergraduate university students who also completed the Modified Barrett-Lennard Relationship Inventory (BLRI) and the 10-question Rosenberg Self-Esteem (RSE) scale. The subjects' responses on the three scales were compared. Major findings indicated that individual levels of self-liking varied between areas of self-identity. Moreover, the level of self-liking was not necessarily dependent on their level of success. Each of the corresponding measures of the three instruments showed positive correlations, except the measures of unconditionality. Finally, each of the four primary patterns of self-liking described by the USL model characterized some subjects. Findings support the concept of unconditional self- liking, which suggests that high levels of self-liking arc not limited to only the most capable, intelligent, talented, or attractive people.

Blackfeet youth suicide Attempters and a sample of non-suicide Attempters were compared on the Family Adaptability and Cohesion Evaluation Scales (FACES- 111), the Family Environment Scale (FES), the Scale for Suicide Ideation (SSI), the revised Beck Depression Scale (BDI) and a biographical inventory. The purpose of this study was to define a set of variables that identify Indian youth with a high potential for suicide. The study sample consisted of 60 participants between the ages of 15-24, 30 suicide Attempters and 30 non-suicide Attempters, from the Blackfeet reservation. Fifteen identified variables were proposed to differentiate between the two groups. Analyses involving nine variables revealed a significant correlation between the revised Beck Depression Inventory variable of depression and the Biographical Inventory self-report variable of suicide attempt. Results indicated that suicide Attempters scored higher on the revised Beck Depression Inventory than did those subjects who did not attempt suicide.

No description available.

No description available.

No description available.

No description available.

No description available.