UMD Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/3
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
6 results
Search Results
Item Unveiling secrets of brain function with generative modeling: Motion perception in primates & Cortical network organization in mice(2023) Vafaii, Hadi; Pessoa, Luiz; Butts, Daniel A; Physics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This Dissertation is comprised of two main projects, addressing questions in neuroscience through applications of generative modeling. Project #1 (Chapter 4) is concerned with how neurons in the brain encode, or represent, features of the external world. A key challenge here is building artificial systems that represent the world similarly to biological neurons. In Chapter 4, I address this by combining Helmholtz's “Perception as Unconscious Inference”---paralleled by modern generative models like variational autoencoders (VAE)---with the hierarchical structure of the visual cortex. This combination results in the development of a hierarchical VAE model, which I subsequently test for its ability to mimic neurons from the primate visual cortex in response to motion stimuli. Results show that the hierarchical VAE perceives motion similar to the primate brain. I also evaluate the model's capability to identify causal factors of retinal motion inputs, such as object motion. I find that hierarchical latent structure enhances the linear decodability of data generative factors and does so in a disentangled and sparse manner. A comparison with alternative models indicates the critical role of both hierarchy and probabilistic inference. Collectively, these results suggest that hierarchical inference underlines the brain's understanding of the world, and hierarchical VAEs can effectively model this understanding. Project #2 (Chapter 5) is about how spontaneous fluctuations in the brain are spatiotemporally structured and reflect brain states such as resting. The correlation structure of spontaneous brain activity has been used to identify large-scale functional brain networks, in both humans and rodents. The majority of studies in this domain use functional MRI (fMRI), and assume a disjoint network structure, meaning that each brain region belongs to one and only one community. In Chapter 5, I apply a generative algorithm to a simultaneous fMRI and wide-field calcium imaging dataset and demonstrate that the mouse cortex can be decomposed into overlapping communities. Examining the overlap extent shows that around half of the mouse cortical regions belong to multiple communities. Comparative analyses reveal that calcium-derived network structure reproduces many aspects of fMRI-derived network structure. Still, there are important differences as well, suggesting that the inferred network topologies are ultimately different across imaging modalities. In conclusion, wide-field calcium imaging unveils overlapping functional organization in the mouse cortex, reflecting several but not all properties observed in fMRI signals.Item Auditory cortical response to spectrotemporally dynamic stimuli during passive listening and behavior(2022) Liu, Ji; Butts, Daniel; Biology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Our sensory system is bombarded with information that can change whimsically and yet we make sense of the flow of the information effortlessly. How does the brain encode such richly dynamic stimuli? Specifically, how does the auditory system encode rich spectral and temporal aspects of the stimulus and how does it depend on the behavioral state of the animal? My study aims to answer these questions within the scope of mouse auditory cortex (ACX) using imaging techniques on various scales. Firstly, I studied how the ACX encodes one temporal aspect of the sound, specifically the onset and the offset. I found that offset responses dominated ACX at high sound levels and their strength depended on auditory cortical fields. Moreover, ACX neurons likely inherit their offset responses from thalamocortical input, which is further processed by local cortical microcircuit. Second, I studied the spectral tuning properties of layer 2/3 neurons in mouse ACX using two-tone stimuli. This study revealed the complex inhibitory sideband structures not only in excitatory and inhibitory neurons, but also in feedforward input from auditory thalamus. These complex structures showed a higher degree of feature selectivity of auditory neurons beyond what is predicted by conventional tuning, and thus auditory cortical responses are highly dependent on the spectral context. These two studies focused on passive listening, but cortical responses could depend on the behavioral state of the animal. The predictive coding theory proposes that sensory cortical responses are a form of error response signaling when sensory input failed to conform with predictions from higher order brain areas. Thus, to study the encoding of spectrotemporally dynamic stimulus under active engagement and to test the predictive coding theory, I designed a novel behavior paradigm that allowed the animal to interact with the sound stimulus and studied the cortical responses to not only the combination of sensory information and the animal’s action but also the introduced perturbation. Together, this dissertation combined advanced imaging techniques and innovations in experimental designs to provide new insight into how ACX encodes sound stimulus under various scenarios.Item Spatiotemporal Dynamics and Functional Organization of Auditory Cortex Networks(2021) Bowen, Zac; Kanold, Patrick O; Losert, Wolfgang; Biophysics (BIPH); Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The sensory cortices of the brain are highly complex systems that are uniquely adapted to reliably process any encountered sensory stimulus. Sensory stimuli such as sound are encoded in large populations of neurons that exhibit some functional organization in the cortex. For example, the auditory cortex has a characteristic organization of sound frequency by which neuronal responses are organized. However, this organization is a broad approximation of more complex and diverse functional properties of individual neurons. Furthermore, on a finer temporal scale, the moment-to-moment activity dynamics of populations of neurons are incredibly complex. Numerous studies have shown that spatiotemporal cascades of co-active neurons organize as neuronal avalanches possessing certain characteristics such as size, duration, and shape that fit the parameters of a critical system. Nevertheless, it remains that the exact manner in which neuronal populations encode information is still not fully understood. This dissertation makes use of neuroimaging data acquired with 2-photon calcium imaging of the auditory cortex in awake mice to investigate the spatiotemporal and functional organization of active neuronal populations in auditory cortex at a range of temporal and spatial scales. I aimed to gain a deeper understanding into how neuronal population dynamics and the underlying network organization contribute to sound encoding in auditory cortex. I studied input and associative layers of auditory cortex (L4 and L2/3) in a mouse model with normal hearing and another with age-related hearing loss due to loss of proper cochlear function to high-frequency sound. L4 and L2/3 contained populations of neurons with a large diversity in functional properties, though diversity was reduced in the hearing loss model due to paucity of high frequency tuned neurons. Despite the diverse tuning in both, similarly responding neurons tended to be co-localized in cortical space. I found that this result extended to volumetric samples of L2/3 where large populations of neurons contained a functional network architecture indicative of small-world topology. Furthermore, I demonstrated that L4 and L2/3 contain ensembles of co-active neurons indicative of critical dynamics in both the absence and presence of a stimulus. Finally, I developed software that facilitates real-time quantification of neuronal populations during an experiment which opens the door for novel closed-loop experiment design. This dissertation provides several avenues for further investigation into neuronal population coding and dynamics, functional network topology, and provides the groundwork for closed-loop experimental design.Item The activation of memory B cells to generate high affinity antibody responses in vitro and in vivo(2011) Richard, Katharina; Song, Wenxia; Cell Biology & Molecular Genetics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Immunological memory is the hallmark of the adaptive immune system. The humoral branch of the immunological memory is mediated by memory B-cells (mB). Memory B cells are marked by longevity, expression of antibodies with high affinity, and ability to generate robust antibody responses upon reencountering pathogens. However, requirements for the activation of mB cells and the induction of humoral memory responses are not well understood. This thesis examines the role of Toll-like receptors (TLRs) in mB activation using an immunized mouse model. TLRs are a family of receptors that recognize common molecular patterns of microbial pathogens and stimulate innate immune responses. Our study found that mouse mB expressed TLR9 and 4, and responded to their agonists in vitro by differentiating into high affinity IgG secreting plasma cells. However, TLR agonists alone were not sufficient to activate memory B cells in vivo. Antigen was required for the clonal expansion of antigen-specific memory B cells, the differentiation of mB cells to high affinity IgG secreting plasma cells, and the recall of high affinity antibody responses. The Ag- specific B cells that had not yet undergone isotype switching showed a relatively higher expression of TLR4 than memory B cells, which was reflected in a heightened response to its agonist, but in both cases of TLR4 and 9 yielded mostly low affinity IgM secreting plasma cells. When immunized together with the antigen, TLR agonists not only boosted the antigen-specific titers, but also increased affinity and isotype switching of the immunoglobulin. Thus, while TLR agonists alone are unable to activate mB in vivo, they can enhance humoral memory responses induced by the antigen.Item Development of a mouse model for the t(10:11)(p13;q14) chromosomal translocation associated with acute leukemia in humans(2008-08-08) Caudell, David L; Samal, Siba K; Aplan, Peter D; Veterinary Medical Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Acute leukemia is associated with a wide spectrum of gross chromosomal rearrangements. These acquired mutations include balanced and unbalanced chromosomal translocations. The analysis of chromosomal translocations has provided much insight into understanding the biology of hematologic malignancies, leading to improved diagnosis and classification, as well as identification of novel therapeutic targets. The rare but recurring chromosomal translocation [t(10;11)(p13;q21)] results in a CALM-AF10 fusion that occurs in patients with both acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). CALM-AF10 transgenic mice developed AML with lymphoid features and had Hoxa gene cluster upregulation. In this model, mice developed leukemia after a long latency period with incomplete penetrance. These findings suggest that additional genetic events are needed to complement CALM-AF10 mediated leukemic transformation. Retroviral insertional mutagenesis was used to identify complementary genetic events that might collaborate with CALM-AF10 during leukemic transformation. A cohort of CALM-AF10 mice was infected with the Mol4070LTR retrovirus; by 5.5 months of age, 50% of the transgenic mice developed AML, a clear acceleration of disease onset compared to either wild type littermates injected with the retrovirus or CALM-AF10 mice not injected with the retrovirus. The tumors assayed by Southern blotting for viral integration showed clonal to oligoclonal expansion. Ligation-mediated PCR and sequence analysis of DNA derived from leukemic cells was used to identify potential collaborating genes at the retroviral insertion sites including Evi1, Nf1, kRas, Zeb2, and Mnl. Identification of these genes as a potential collaborating gene with CALM-AF10 supports the emerging paradigm in leukemia biology that predicts that most, if not all leukemic cells must undergo at least two collaborative events to produce a fully transformed cell. One of these events typically leads to impaired differentiation and enhanced renewal of stem cells, whereas the second event leads to increased proliferation and/or decreased apoptosis. It has been shown here that retroviral infection accelerates the onset of acute leukemia, and identified genes that potentially collaborate with the CALM-AF10 fusion gene in the leukemic transformation process. This transgenic murine model serves as a model system for studying leukemogenesis similar to that observed in humans with leukemia.Item Nutritional and Physiological Control of Metabolic Pathways That Alter Milk Protein and Lactose Synthesis by the Mammary Gland(2006-12-08) Schoenberg, Katie Marie; Bequette, Brian J; Animal Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The objectives of this research were to develop a [U-<sup>13</sup>C]glucose tracer approach establishing the pathways and substrates for milk lactose and casein synthesis, and determine the influence of protein intake on murine mammary gland metabolism. Milk samples were collected after one, three and five days of feeding tracer (as 10% of dextrose). <sup>13</sup>C-Isotopic and isotopomer plateaus were attained by day three, establishing the time-course necessary for tracer feeding. 23% of lactose-derived glucose originated from sources other than blood glucose. Six paired (intake and pups equal) sets of lactating mice were fed either a normal (20%) or low (10%) protein diet. <sup>13</sup>C-mass isotopomer distribution (MID) in lactose-derived glucose and galactose did not differ, indicating common mammary metabolic pools. <sup>13</sup>C-MID in blood versus casein indicated significant mammary synthesis of glutamate (Normal:51%, Low:50%), alanine (Normal:32%, Low:29%), and serine (Normal:18%, Low:37%, P < 0.05), suggesting additional requirements for glucose and/or EAA for NEAA synthesis.