https://scholarworks.unist.ac.kr/handle/201301/11 Wed, 08 Apr 2026 00:41:18 GMT 2026-04-08T00:41:18Z https://scholarworks.unist.ac.kr/handle/201301/90911 Title: Analysis of 10,239 high-resolution mitogenomes reveals Korean maternal lineages and population-specific variants defining novel ethnically distinct clusters Author(s): Choi, Yookyung Abstract: This thesis presents the first population-scale characterization of the Korean mitochondrial landscape using 10,239 whole-genome sequences. Through an optimized and scalable assembly pipeline, we generated 10,166 complete and circularized mitogenomes within three days, enabling highly accurate mitochondrial reconstruction at national cohort scale. These high-fidelity assemblies made it possible to reconstruct a whole-mitogenome phylogenetic tree directly from de novo sequences rather than relying on predefined haplogroup markers. Unsupervised principal component analysis of complete mitogenomes revealed four distinct Korean Mitochondrial (KM) clusters, representing fine-scale maternal lineage structure not captured by traditional haplogroup classifications. The KM clusters were defined by 24 population-informative variants and corresponded to separate branches of the M, N, and R macrohaplogroups, indicating unexpectedly deep maternal lineage divergence within Korea. Among the cluster-defining variants, one corresponded to the 9-bp deletion in the COX2–tRNA Lys intergenic region (MIC9D), a well-known anthropological marker in East Asian and Pacific populations. Although MIC9D showed strong contribution to Korean clustering patterns, previous reports of its co- transmission with pathogenic mtDNA mutations suggest that its relevance should be investigated further. Overall, this study demonstrates that whole-mitogenome, assembly-based phylogeny surpasses marker- based haplogrouping by resolving population-specific mitochondrial lineages. The resulting high- resolution reference lays the foundation for future studies of Korean mitochondrial evolution and disease association. Major: Department of Biomedical Engineering Sat, 31 Jan 2026 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/90911 2026-01-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/90910 Title: Injectable Adhesive Hydrogel for Sustained Local Delivery of CRISPR-RNP Loaded Lipid Nanoparticles Author(s): Im, Jisu Abstract: CRISPR-Cas9 therapeutics hold great promise for the treatment of solid tumors but are limited by inefficient and short-lived delivery of gene-editing components, particularly at the viable tumor rim and adjacent hypoxic regions. Lipid nanoparticles (LNPs) are widely used as non-viral carriers, yet both systemic administration and conventional intratumoral injections result in rapid dispersion, off-target biodistribution, and poor intratumoral retention. This study presents a bio-adhesive, surface-adherent alginate hydrogel as an in situ-forming local depot for the sustained delivery of CRISPR-Cas9 ribonucleoprotein (RNP)-loaded LNPs The C12-200 based LNP platform was formulated and characterized to ensure consistent encapsulation and physicochemical properties suitable for CRISPR-RNP delivery. Alginate was functionalized with N-hydroxysuccinimide (Alg-NHS) and ionically crosslinked with calcium ions to generate an adhesive hydrogel exhibiting an interconnected porous microstructure, strong tissue adhesion, and stable gelation under physiological conditions. Release studies using LNP-loaded hydrogels demonstrated diffusion- controlled, sustained release that could be tuned by polymer and crosslinking density. Fully crosslinked Alg-NHS hydrogels showed negligible intrinsic cytotoxicity, and prolonged, lower-intensity exposure to LNPs via hydrogel release mitigated non-specific formulation-related toxicity while preserving potent CRISPR-RNP mediated tumor cell killing in vitro. In vivo, subcutaneously injected Alg-NHS hydrogels formed stable depots that adhered to surrounding tissues and were retained for extended periods. In a subcutaneous HCT116 xenograft model, local administration of CRISPR-RNP (MT50) loaded alginate adhesive hydrogels around the tumor suppressed tumor growth without apparent systemic toxicity. Further optimization of hydrogel composition and payload dose enhanced the antitumor efficacy. Overall, these findings indicate that bio-adhesive, surface-adherent alginate hydrogels can function as a long-acting local depot for CRISPR-LNP formulations and offer a conceptually distinct framework for designing sustained, localized gene-editing therapies in solid tumors. Major: Department of Biomedical Engineering Sat, 31 Jan 2026 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/90910 2026-01-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/90909 Title: Serial Dependence Under Transparent Distraction: Encoding vs. Decoding Influences Author(s): Yoon, Hoyeon Abstract: The perceptual system must reliably interpret incomplete sensory input that inherently contains both internal and external noise. To achieve stable perception, the system exploits the temporal statistics of visual stimuli, which are correlated over time. As a result, the perceptual system incorporates information from previous stimuli and previous responses when processing current inputs, producing biases in current perception, a phenomenon known as serial dependence. Serial dependence has been reported across multiple visual features, including orientation, motion, and face identity. However, prior studies have primarily focused on serial dependence for attended stimuli and have not examined how the presence of distractors alters this effect. In natural environments, targets often appear alongside transparent distractors that must be suppressed, and these distractors may modulate serial dependence. The present study investigates how distractor stimuli presented in the current and previous trials influence the processing of sequential sensory information. Participants performed a direction judgment task in which two different colored moving dots were presented transparently. The results show that serial dependence driven by previous responses does not directly influence the perceived direction of the current target; instead, it acts on the distractor-biased direction that has been shifted by the current distractor. This indicates that the influence of previous responses arises during the decoding stage, after the target signals have been encoded. Additionally, the effects of previous target and distractor stimuli were determined by their motion directions rather than the color feature that distinguished the two stimuli. Importantly, the influence of the previous distractor exhibited a different pattern from that of the previous target, suggesting that distractors alter the encoding stage by contributing direction specific information that differs from the target driven bias. Overall, this study demonstrates that the presence of distractors modulates the processing of sequential sensory input. Furthermore, it provides empirical evidence that previous response effects emerge during the decoding stage, whereas previous stimulus effects originate during the encoding stage. The significance of this work lies in showing how the perceptual system integrates not only target information but also transparent distractor signals when forming perceptual estimates over time. Major: Department of Biomedical Engineering Sat, 31 Jan 2026 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/90909 2026-01-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/90908 Title: Temporal Dynamics of Emotion in Magnetoencephalography: From Dataset Validation to Quantitative Trajectory Analysis Author(s): Yoo, MoonA Abstract: While emotion is fundamentally time-varying process, the dominant paradigm in brain-based emotion recognition has remained focused on static classification. This focus on single, aggregate labels, rather than on the unfolding temporal dynamics, has persisted due to two primary challenges: (1) a scarcity of high-resolution, richly labeled neural datasets, and (2) a lack of analytical frameworks to quantify the properties of emotional trajectories. This thesis addresses both challenges through a sequential, two-part study. Study 1 addressed the dataset challenge by constructing and validating a novel, large-scale Magnetoencephalography (MEG) dataset. This dataset was built using culturally optimized video stimuli for Korean young adults and employed a comprehensive assessment method combining dimensional (SAM) and fine-grained (PrEmo) labels. Validation results confirmed the dataset’s high quality: robust four-class neural classification (89.87% accuracy) was achieved, and critically, we found evidence of a brain-behavior correspondence: videos that elicited ambiguous SAM behavioral responses correspondingly yielded lower classification accuracy. Study 2 built upon this validated dataset to address the analysis challenge. The study proposed and validated a novel set of trajectory-based indices designed to quantify the dynamic properties of emotion estimation trajectories derived from the MEG data. The findings demonstrate that these temporal indices, which quantify dynamic properties such as stability and variability, successfully differentiated not only the broad emotional categories but also the more fine-grained emotional differences within those categories-distinctions missed by traditional classification. By first establishing a foundational dataset and then validating a new dynamic analysis method upon it, this thesis provides a validated pathway for enriching emotion recognition, introducing temporal trajectory indices as a new analytical dimension. This work establishes a robust methodological groundwork for the detailed and temporally precise assessment and quantification of human affective dynamics. Major: Department of Biomedical Engineering Sat, 31 Jan 2026 15:00:00 GMT https://scholarworks.unist.ac.kr/handle/201301/90908 2026-01-31T15:00:00Z