Honors Projects
Showing 311 - 320 of 662 Items
Characterizing the Motor Activity Patterns of the Mammalian Thoracic Spinal Cord Neural Network This record is embargoed.
- Embargo End Date: 2027-05-16
Date: 2024-01-01
Creator: Sam McClelland
Access: Embargoed
A histological investigation of Arceuthobium pusillum infections in Picea rubens and Picea glauca
Date: 2024-01-01
Creator: Sade K. McClean
Access: Open access
- Arceuthobium pusillum is a hemiparasite that infects select Picea species. The hosts of A. pusillum do not experience the same symptoms of infection. A. pusillum infections are more fatal to P. marinara, and P. glauca. P. rubens, on the other hand, can survive longer with sustained infection. This presents itself as a contemporary issue because P. glauca, one of the parasite’s most vulnerable hosts, was untethered from ecological competition when old growth forests were subjected to large scale anthropogenic disturbances. These disturbances allowed P. glauca to proliferate, with A. pusillum following. A deeper understanding of the host-species specific responses to A. pusillum infection can broaden general knowledge of parasitic growth and development while also potentially inspiring conservation techniques. This study took advantage of the intrinsic differences between host and parasite to visualize infections in P. rubens and P. glauca, highlighting differences in infection outcome. By illuminating lignin and callose within cross sections of infected P. rubens and P. glauca branches, it was revealed that P. rubens forms dense bands of cells around the cortical strands of infection. These bands form more frequently in P. rubens than in P. glauca and are of a significantly larger area in P. rubens than in P. glauca (t(8), p=0.003, p=0.005). The discovery of the exterior bands is novel and exciting, as the bands are possibly made of callose and potentially facilitate P. rubens survival against A. pusillum infection. The foundational discoveries and results of this study should inspire, and warrant, further analysis.
Are People Blaming Artificial Intelligence More or Less for Incorrect Advice? Access to this record is restricted to members of the Bowdoin community. Log in here to view.
- Restriction End Date: 2029-06-01
Date: 2024-01-01
Creator: Anh Nguyen
Access: Access restricted to the Bowdoin Community
Creating Enantioselective Peptoid Catalysts with 2-Picolylamine and 2-Picolinic Acid Catalytic Sites Access to this record is restricted to members of the Bowdoin community. Log in here to view.
Date: 2024-01-01
Creator: Devin Kathleen O’Loughlin
Access: Access restricted to the Bowdoin Community
Statistically Principled Deep Learning for SAR Image Segmentation
Date: 2024-01-01
Creator: Cassandra Goldberg
Access: Open access
- This project explores novel approaches for Synthetic Aperture Radar (SAR) image segmentation that integrate established statistical properties of SAR into deep learning models. First, Perlin Noise and Generalized Gamma distribution sampling methods were utilized to generate a synthetic dataset that effectively captures the statistical attributes of SAR data. Subsequently, deep learning segmentation architectures were developed that utilize average pooling and 1x1 convolutions to perform statistical moment computations. Finally, supervised and unsupervised disparity-based losses were incorporated into model training. The experimental outcomes yielded promising results: the synthetic dataset effectively trained deep learning models for real SAR data segmentation, the statistically-informed architectures demonstrated comparable or superior performance to benchmark models, and the unsupervised disparity-based loss facilitated the delineation of regions within the SAR data. These findings indicate that employing statistically-informed deep learning methodologies could enhance SAR image analysis, with broader implications for various remote sensing applications and the general field of computer vision. The code developed for this project can be found here: https://github.com/cgoldber/Statistically-Principled-SAR-Segmentation.git.
Selective Procedural Content Generation Using Multi-Discriminator Generative Adversarial Networks This record is embargoed.
- Embargo End Date: 2025-05-16
Date: 2024-01-01
Creator: Darien Gillespie
Access: Embargoed
Poverty Ends with a 12 Year Old Girl: Empowerment and the Contradictions of International Development
Date: 2017-05-01
Creator: Meghan Elizabeth Bellerose
Access: Open access
- This thesis argues that international development programs focused on adolescent girls reproduce problematic and contradictory depictions of girls in the global South. Using Girl Effect marketing materials and interviews with INGO staff, I demonstrate that present-day international aid programs center on the neoliberal notion that an empowered adolescent girl holds the unique potential to end global poverty. Through empowerment programs, girls are encouraged to recognize their agency and take personal responsibility for improving the wellbeing of their communities. However, I argue that even as development leaders claim that an empowered adolescent girl is a source of indefatigable strength who can transform her community, they carry a deep conviction that such a feat is not possible without significant Western aid. Despite the empowerment rhetoric that The Girl Effect and related international initiatives espouse, their programs depict adolescent girls in the developing world as vulnerable and oppressed by poverty, local men, and their cultures. Thus, Western donors are called upon to save “Third World” adolescent girls. I argue that these contradictions in the language of international development contribute to the perception of girls in the global South as weak, inferior, and homogenous and lead to the establishment of programs that strengthen inequitable structures and sideline girls’ sexual rights.
Hybridization dynamics of a newly discovered parrotfish swarm in the Tropical Eastern Pacific
Date: 2017-05-01
Creator: Robert Barron
Access: Open access
- Hybrid zones and their dynamics are important in the understanding of the genetic basis of reproductive isolation and speciation. This study seeks to investigate the hybridization dynamics of a Scarus hybrid swarm within the Tropical Eastern Pacific (TEP) that includes four phenotypically distinct species: S. perrico, S. ghobban, S. rubroviolaceus, and S. compressus. Genetic and population structure analyses of four nuclear loci and a mitochondrial locus revealed that one of the four species, S. compressus, was the result of two different hybrid crosses: S. perrico ✕ S. rubroviolaceus and S. perrico ✕ S. ghobban. A NewHybrids model indicated that most of the S. compressus samples were F1 hybrids, but 21% of the S. compressus sample was classified as “parentals” which could also be explained by the presence of either F2 hybrids or backcrosses with S. compressus phenotypes, given the relatively low power of the nuclear data set (4 loci) to resolve complex hybrid genotypes. Significant mito-nuclear discordance in all three non-hybrid species is consistent with an evolutionary effect of backcrossing between F1 hybrids and “pure” species. This study reveals a relative ease of hybridization between parrotfish taxa separated by an estimated 4.5 million years of isolation and opens the door to further studies on the potential effects of gene flow across old species boundaries and perhaps the formation of new species by hybrid speciation in a diverse clade of tropical reef fish. Elucidating the nature of potentially “deep” F2 crosses and backcrosses within the TEP Scarus hybrid system will allow us to better understand the effects of hybridization on evolution and speciation on both a micro- and macro-ecological scale.
Being and Salvation: Environmental Implications of Martin Heidegger's Thought Access to this record is restricted to members of the Bowdoin community. Log in here to view.
- Restriction End Date: 2027-06-01
Date: 2024-01-01
Creator: Eleanor S. Huntington
Access: Access restricted to the Bowdoin Community
Investigating the Impacts of Drought on Turfgrass (Festuca arundinacea) Chlorophyll-a Fluorescence Emission
Date: 2024-01-01
Creator: Ayanna S Hatton
Access: Open access
- When photons from sunlight are absorbed by plants, they can take paths of photosynthesis, fluorescence, or energy dissipation. Instruments to quantify fluorescence have expanded in scale to allow measurements from satellites and flux towers using Solar-Induced Chlorophyll Fluorescence (SIF). Studies have found a positive correlation between SIF and gross primary productivity (GPP; representative of photosynthesis), suggesting SIF is a proxy for GPP. This conclusion encourages the use of SIF to inform decisions about carbon budgets and responding to climate change. Studies of fluorescence on the single-leaf scale have revealed that SIF measurements do not account for all variables nor is there an understanding of the impact of environmental factors, such as drought, on these measurements. In this project, tall fescue turfgrass was placed in one of four differing drought severities for 19 days. Leaf-level measurements of photosynthesis and pulse-amplitude modulated fluorescence were made, demonstrating stomatal closure and inhibition of photosynthesis. This physiological change caused greater photon allocation to energy dissipation. Changes in greenness and the utilization of photoprotective mechanisms such as senescence and anthocyanin accumulation were observed. This study has provided an understanding of the temporal, physiological, and visible impacts of drought on turfgrass to inform interpretations of SIF in future experiments. Caution is crucial in utilizing SIF as a proxy for GPP before further research into the impact of drought on SIF is completed.