Biology subjects

Ma, Y.

Publications and source records attributed to Ma, Y..

At least 289 records · Page 16Linked to original sources

bioRxiv2023

DCBLD1 modulates angiogenesis by regulating of the VEGFR-2 endocytosis in endothelial cells

ObjectiveUnwanted angiogenesis is involved in the progression of various malignant tumors and cardiovascular diseases, and the factors that regulate angiogenesis are potential therapeutic targets. We tested the hypothesis that DCBLD1 (Discoidin, CUB, and LCCL domain-containing protein 1) is a co-receptor of VEGFR-2 and modulates angiogenesis in endothelial cells(ECs). Approach and ResultsA carotid artery ligation model and retinal angiogenesis assay were used to study angiogenesis using globe knockout or EC-specific conditional DCBLD1 knockout mice in vivo. Immunoblotting, immunofluorescence staining, plasma-membrane subfraction isolation, Co-immunoprecipitation and mass-spectrum assay were performed to clarify the molecular mechanisms.Loss of DCBLD1 impaired VEGF response and inhibited VEGF-induced EC proliferation and migration. DCBLD1 deletion interfered with adult and developmental angiogenesis. Mechanistically, DCBLD1 bound to VEGFR-2 and regulated the formation of VEGFR-2 complex with negative regulators: protein tyrosine phosphatases, E3 ubiquitin ligases(Nedd4 and c-Cbl), and also DCBLD1 knockdown promoted lysosome-mediated VEGFR-2 degradation in ECs. ConclusionsThese findings demonstrated the essential role of endothelial DCBLD1 in regulating VEGF signaling and provided evidence that DCBLD1 promotes VEGF-induced angiogenesis by limiting the dephosphorylation, ubiquitination, and lysosome degradation after VEGFR-2 endocytosis. We proposed that endothelial DCBLD1 is a potential therapeutic target for ischemic cardiovascular diseases by the modulation of angiogenesis through regulating of the VEGFR-2 endocytosis.

molecular biology↗

bioRxiv2023

Polymerization force-regulated Arp2/3 complex-actin filament interaction dominates self-adaptive cell migrations

Cells migrate by adapting their leading-edge behaviours to heterogeneous extracellular microenvironments (ECMs) during cancer invasions and immune responses. Yet it remains poorly understood how such complicated dynamic behaviours emerge from millisecond-scale assembling activities of protein molecules, which are hard to probe experimentally. To address this gap, we established a spatiotemporal "resistance-adaptive propulsion" theory based on the protein interactions between Arp2/3 complexes and polymerizing actin filaments, and a multiscale dynamic modelling system spanning from molecular proteins to the cell. Combining spatiotemporal simulations with experiments, we quantitatively find that cells can accurately self-adapt propulsive forces to overcome heterogeneous ECMs via a resistance-triggered positive feedback mechanism, dominated by polymerization-induced actin filament bending and the bending-regulated actin-Arp2/3 binding. However, for high resistance regions, resistance triggered a negative feedback, hindering branched filament assembly, which adapts cellular morphologies to circumnavigate the obstacles. Strikingly, the synergy of the two opposite feedbacks not only empowers cells with both powerful and flexible migratory capabilities to deal with complex ECMs, but also endows cells to use their intracellular proteins efficiently. In addition, we identify that the nature of cell migration velocity depending on ECM history stems from the inherent temporal hysteresis of cytoskeleton remodelling. We also quantitatively show that directional cell migration is dictated by the competition between the local stiffness of ECMs and the local polymerizing rate of actin network caused by chemotactic cues. Our results reveal that it is the polymerization force-regulated actin filament-Arp2/3 complex binding interaction that dominates self-adaptive cell migrations in complex ECMs, and we provide a predictive theory and a spatiotemporal multiscale modelling system at the protein level.

biophysics↗

bioRxiv2023

An Azapeptide Platform in Conjunction with Covalent Warheads to Uncover High-Potency Inhibitors for SARS-CoV-2 Main Protease

Main protease (MPro) of SARS-CoV-2, the viral pathogen of COVID-19, is a crucial nonstructural protein that plays a vital role in the replication and pathogenesis of the virus. Its protease function relies on three active site pockets to recognize P1, P2, and P4 amino acid residues in a substrate and a catalytic cysteine residue for catalysis. By converting the P1 C atom in an MPro substrate to nitrogen, we showed that a large variety of azapeptide inhibitors with covalent warheads targeting the MPro catalytic cysteine could be easily synthesized. Through the characterization of these inhibitors, we identified several highly potent MPro inhibitors. Specifically, one inhibitor, MPI89 that contained an aza-2,2-dichloroacetyl warhead, displayed a 10 nM EC50 value in inhibiting SARS-CoV-2 from infecting ACE2+ A549 cells and a selectivity index of 875. The crystallography analyses of MPro bound with 6 inhibitors, including MPI89, revealed that inhibitors used their covalent warheads to covalently engage the catalytic cysteine and the aza-amide carbonyl oxygen to bind to the oxyanion hole. MPI89 represents one of the most potent MPro inhibitors developed so far, suggesting that further exploration of the azapeptide platform and the aza-2,2-dichloroacetyl warhead is needed for the development of potent inhibitors for the SARS-CoV-2 MPro as therapeutics for COVID-19.

biochemistry↗

bioRxiv2023

Both CD8 and CD4 T cells contribute to immunosurveillance preventing the development of neoantigen-expressing autochthonous sarcomas

The adaptive immune system plays an essential anti-tumor role through immunosurveillance and response to immunotherapies. Characterizing phenotypic features and mechanisms of dysfunction of tumor-specific T cell populations may uncover novel immunotherapeutic targets and biomarkers of response. To study tumor-specific T cell responses in vivo, a tumor model must express a known neoantigen. While transplant models with known neoantigen expression are widely available, autochthonous tumor models in which the tumor coevolves with the immune system are limited. In this study, we combined CRISPR/Cas9 and sleeping beauty transposase technology to develop an autochthonous orthotopic murine sarcoma model with oncogenic KrasG12D, functionally impaired p53, and expression of known MHCI and MHCII sarcoma neoantigens. Using MHC tetramer flow cytometry, we identified a tumor-specific immune response in the peripheral blood as early as 10 days after tumor induction leading to tumor clearance. Tumors developed at high penetrance after co-depletion of CD8 and CD4 T cells, but depletion of either CD8 or CD4 T cells alone was insufficient to permit tumor growth. These results suggest that CD8 and CD4 T cells can independently contribute to immunosurveillance leading to clearance of sarcomas expressing MHCI and MHCII neoantigens.

cancer biology↗

bioRxiv2023

Loss of Grem1-articular cartilage progenitor cells causes osteoarthritis.

Osteoarthritis (OA), which carries an enormous disease burden across the world, is characterised by irreversible degeneration of articular cartilage (AC), and subsequently bone. The cellular cause of OA is unknown. Here, using lineage tracing in mice, we show that the BMP-antagonist Gremlin 1 (Grem1) marks a novel chondrogenic progenitor (CP) cell population in the articular surface that generates joint cartilage and subchondral bone during development and adulthood. Notably, this CP population is depleted in injury-induced OA, and with age. OA is also induced by toxin-mediated ablation of Grem1 CP cells in young mice. Transcriptomic analysis and functional modelling in mice revealed articular surface Grem1-lineage cells are dependent on Foxo1; ablation of Foxo1 in Grem1-lineage cells led to early OA. This analysis identified FGFR3 signalling as a therapeutic target, and injection of its activator, FGF18, caused proliferation of Grem1-lineage CP cells, increased cartilage thickness, and reduced OA pathology. We propose that OA arises from the loss of CP cells at the articular surface secondary to an imbalance in progenitor cell homeostasis and present a new progenitor population as a locus for OA therapy.

cell biology↗

bioRxiv2023

COMT inhibitor tolcapone represses the migration and invasion of trophoblast and results in preeclampsia-like phenotypes in mice

Pregnancy in young-onset Parkinsons disease (YOPD) is rare; however, medication for this condition is critical for maternal and fetal health. Tolcapone is an effective antiparkinsonian drug. However, its safety studies on mothers and fetuses are limited. In this study, we aimed to investigate the effects of tolcapone on the mother and the developing fetus during different stages of gestation. Tolcapone was administered to pregnant mice at a dose of 60 (H-Tol) or 30 mg/kg/day (L-Tol) during early or mid-gestation. We observed that tolcapone administration during early gestation causes abortion and delays fetal development in a dose-dependent manner. During mid-gestation tolcapone barely caused embryo lethality; however, the mice developed preeclampsia-like phenotypes, including maternal hypertension, proteinuria, and fetal growth restriction. Histomorphological analysis of placentas from tolcapone treated mice revealed abnormalities in the trophoblast layer and the impaired trophoblast invasion in the decidua. Mechanistically, we revealed that tolcapone inhibits the invasion and migration of trophoblasts in vitro, with changes in the protein expression of Snail, Twist, and E-cadherin. In conclusion, tolcapone caused embryo lethality and growth restriction during early gestation, whereas it caused preeclampsia-like phenotypes in mice with defective trophoblast invasion in mid-gestation. Collectively, our study provides novel insights into the effects of tolcapone on pregnancy.

pharmacology and toxicology↗

bioRxiv2023

Mitochondrial genome evolution in the Diprionidae: Major gene rearrangement in the basal Hymenoptera

In comparison to other non-parasitic basal lineages, Apocrita have consistently demonstrated a greatly accelerated rate of gene rearrangement. A number of mechanisms or correlates have been proposed for this observation, such as oxidative stress tolerated by exposure to the host immune system might lead to a high proportion of rearranged mt-genomes. Our studies reveal that gene rearrangements involving the protein-coding gene are present in the basal Hymenoptera lineage based on enriched sampling. We speculate the processes of diversification of rearrangements in the vicinity of nad2 involved tRNAs and NCRs by producing the chronogram of Diprionids. Furthermore, we investigated the relationship between rearranged genes and their nucleotide sequences. In conclusion, we demonstrate the great potential of gene order and associated sequence features as phylogenetic markers in the study of Hymenoptera evolution, offering a new perspective on studying organisms that undergo frequent gene rearrangements.

evolutionary biology↗

bioRxiv2023

Phage administration with repeated intravenous doses leads to faster phage clearance in mammalian hosts

ObjectivesPhage therapy has shown a great promise for the treatment of multidrug-resistant bacterial infections. However, the lack of a thorough and organized understanding of phage-body interactions has limited its clinical application. MethodsHere, we administered different purified phages (Salmonella phage SE_SZW1, Acinetobacter phage AB_SZ6, and Pseudomonas phage PA_LZ7) intravenously to healthy animals (rats and monkeys) to evaluate the phage-induced host responses and phage pharmacokinetics (PK) with different intravenous (IV) doses in healthy animals. The plasma and the organs were sampled after different IV doses to determine the phage biodistribution, the phage-induced cytokines, and antibodies. The potential side effects of phages on animals were assessed. ResultsA non-compartment model revealed that the plasma phage titer gradually decreased over time following a single dose. Repeated doses caused that the plasma phage titer at 5 minutes dropped 2-3 Log10 compared to the first dose regardless of phage types in rats. Host innate immune responses were activated including the upregulated expression (>10-fold) of TNF- and splenic enlargement following repeated doses. Phage-specific neutralization antibodies in animals receiving phages were detected. Similar results were obtained from monkeys. ConclusionsThe mammalian bodies were well-tolerant to the administered phages. The animal responses to the phages and the phage biodistribution profiles could have a significant impact on the efficacy of phage therapy.

pharmacology and toxicology↗

bioRxiv2023

Connectional Hierarchy in Human Brain Revealed by Individual Variability of Functional Network Edges

The human cerebral cortex exhibits intricate interareal functional synchronization at the macroscale, with substantial individual variability in these functional connections. However, the spatial organization of functional connectivity (FC) variability across the human connectome edges and its significance in cognitive development remain unclear. Here, we identified a connectional axis in the edge-level FC variability. The variability declined continuously along this axis from within-network to between-network connections, and from the edges linking association networks to those linking the sensorimotor and association networks. This connectional axis of functional variability is associated with spatial pattern of structural connectivity variability. Moreover, the connectional variability axis evolves in youth with an increasing flatter axis slope. We also observed that the slope of connectional variability axis was positively related to the performance in the higher-order cognition. Together, our results reveal a connectional axis in functional variability that is linked with structural connectome variability, refines during development, and is relevant to cognition. Significance StatementUnderstanding the spatial organization of inter-individual variability in functional connectivity across the human connectome can offer insights into which connections are most susceptible to insults and interventions. We identify a stable axis on functional connectome along which connectivity variability continuously varies, aligning with structural connectivity variability. We find that the pattern of this axis evolves during development, marked by a decrease in functional connectivity variability among association connections. Moreover, the variability axis is associated with individual differences in higher-order cognitions, driven by increased variability in higher-order association connections linked to enhanced cognitive performance. These findings highlight the importance of connectivity-specific plasticity in understanding both typical and atypical cognitive development across diverse populations.

neuroscience↗

bioRxiv2023

Light-Guided Rabies Virus Tracing for Neural Circuit Analysis

Neuronal tracing methods are essential tools to understand the fundamental architecture of neural circuits and their connection to the overall functional behavior of the brain. Viral vectors used to map these transsynaptic connections are capable of cell-type-specific and directional-specific labeling of the neuronal connections. Herein, we describe a novel approach to guide the transsynaptic spreading of the Rabies Virus (RV) retrograde tracer using light. We built a Baculovirus (BV) as a helper virus to deliver all the functional components necessary and sufficient for a nontoxic RV to spread from neuron to neuron, with a light-actuated gene switch to control the RV polymerase, the L gene. This design should allow for precisely controlled polysynaptic viral tracing with minimal viral toxicity. To use this system in a highly scalable and automated manner, we built optoelectronics for controlling this system in vitro with a large field of view using an off-the-shelf CMOS sensor, OLED display panel, and microcontrollers. We describe the assembly of these genetic circuits using the uLoop DNA assembly method and a library of genetic parts designed for the uLoop system. Combining these tools provides a framework for increasing the capabilities of nontoxic tracing through multiple synapses and increasing the throughput of neural tracing using viruses.

neuroscience↗

bioRxiv2023

Network Pharmacology and Bioinformatics Methods Reveal the Mechanism of Sishenwan in the Treatment of diabetic nephropathy

Objective. In order to decipher the bioactive components and potential mechanisms of the traditional Chinese medicine (TCM) formula Sishenwan (SSW) for diabetic nephropathy (DN), we integrated network pharmacology and bioinformatics. Methods. The candidate compounds of SSW and relative targets were obtained from the TCMSP, BATMAN-TCM, SiwssTartgetPrediction, STITCH, and ChEMBL web servers. The UniProt database was used to translate the target names into gene names, and then constructed the herbal-compound-target network. DN-related targets were ascertained based on OMIM, CTD, GeneCards, DisGeNET, and GEO. Furthermore, there was a protein-protein interaction (PPI) network to explore the overlapping targets between SSW and DN, which focused on screening the pivotal targets by topology. GO and KEGG enrichment analyses were carried out to further understand the potential functions associated with the effect of SSW against DN. Eventually, molecular docking simulations were performed to validate the binding affinity between major bioactive components and hub genes. Results. A total of 120 candidate active compounds and 542 corresponding drug targets were derived, in which 195 targets intersected with DN. Then, KEGG pathway analysis showed that several signaling pathways were closely related to the process of SSW against DN, including the AGE-RAGE signaling pathway in diabetic complications, the TNF signaling pathway, and IL-17 signaling pathway, ect. The PPI network analysis identified PTGS2, CREB1, ESR1, TNF, IL1B, INS, AKT1, PPARG, and JUN were the top 9 hub targets. The molecular docking confirmed that the bioactive compounds of SSW had a firm binding affinity with hub targets. Conclusions. As a whole, the present study revealed that SSW exerted therapeutic effects on DN via modulating multi-targets with multi-compounds through multi-pathways.

bioinformatics↗

bioRxiv2023

The Alfin-like transcription factor GmAlfin09 regulates endoplasmic reticulum stress in soybean via a peroxidase GmPRDX6

Soybean is a valuable oil crop cultivated throughout the world, but also highly susceptible to environmental stressors. The development of approaches to enhancing soybean stress resistance is thus vital to improving yields. In prior studies, Alfin has been shown to serve as an epigenetic regulator of plant growth and development. No studies of Alfin have yet been reported in soybean, however. In this study, the endoplasmic reticulum (ER) stress- and reactive oxygen species (ROS)-related transcription factor GmAlfin09 was identified. Screening of genes co-expressed with GmAlfin09 unexpectedly led to the identification of the peroxidase GmPRDX6. Further analyses revealed that both GmAlfin09 and GmPRDX6 were responsive to ER stress, with GmPRDX6 localizing to the ER under stress. Promoter binding experiments confirmed the ability of GmAlfin09 to directly bind the GmPRDX6 promoter. When GmAlfin09 and GmPRDX6 were overexpressed in soybean, enhanced ER stress resistance and decreased ROS levels were observed. Together, these findings suggest that GmAlfin09 can promote the upregulation of GmPRDX6, which subsequently localizes to the ER, reduced ROS levels, promotes ER homeostasis, and helps ensure the normal growth of soybean even under ER stress. This study highlights a novel genetic target for the future molecular breeding of stress-resistant soybean cultivars. HIGHLIGHTGmAlfin09 can increase the expression of GmPRDX6 to reduce ROS level under ER stress.

plant biology↗

bioRxiv2023

Single-cell transcriptomic analyses reveal the cellular and genetic basis of aquatic locomotion in scyphozoan jellyfish

The authors have withdrawn their manuscript because the authors need to re-organize the data and writing. More experimental evidence from whole-mount ISH assay and RNAi assays will be added to support the viewpoints. Therefore, the authors do not wish this work to be cited as reference for the project. If you have any questions, please contact the corresponding author.

ecology↗

bioRxiv2023

Timing of antibiotic administration determines the spread of plasmid-encoded antibiotic resistance during microbial range expansion

The spread of antibiotic resistance (AR) is a major threat to global health. Plasmid-mediated horizontal gene transfer is the main mechanism by which AR is transferred between cells lying within close spatial proximity to each other. In this study, we address a fundamental question regarding the spread of AR-encoding plasmids: Is there an optimal time to administer antibiotics to minimize plasmid spread within microbial communities? We addressed this question using microbial consortia consisting of two Pseudomonas stutzeri strains, where one is an AR-encoding plasmid donor and the other a potential recipient. We allowed the strains to co-expand across a nutrient-amended surface and administered antibiotics at different points in time. We found that the extents of plasmid transfer and transconjugant proliferation have unimodal relationships with the timing of antibiotic administration, where they reach maxima at intermediate administration times. Using an individual-based model, we found that these unimodal relationships result from an interplay between the probability to acquire plasmids from neighboring cells and the probability of plasmid loss upon cell division. Our study provides novel mechanistic insights into the transfer and proliferation of AR-encoding plasmids within microbial communities and identifies the timing of antibiotic administration as an important determinant of AR spread.

microbiology↗

bioRxiv2023

Physical contacts between sparse biofilms promote plasmid transfer and generate functional novelty

The horizontal transfer of plasmids is an important driver of microbial evolution, such as conferring antibiotic resistance (AR) to new genotypes. In biofilms, the abundance of cell-cell contacts promotes the frequent transfer of plasmids and their associated genes. In this study, we expand our knowledge about AR-encoding plasmids by investigating their transfer between discrete biofilms as the biofilms grow and physically collide with each other. Using an experimental system consisting of two fluorescently labelled Pseudomonas stutzeri strains and an Escherichia coli strain, we show that biofilm collisions promote plasmid transfer along the collision boundaries. The extent of plasmid transfer depends on the plasmid loss probability, the plasmid transfer probability, and the relative growth rates of plasmid-free and plasmid-carrying cells. We further show that the proliferation of plasmids after biofilm collision depends on the spatial positionings of plasmid-carrying cells along the collision boundary, thus establishing a link between the large-scale spatial distribution of discrete biofilms and the small-scale spatial arrangement of cells within individual biofilms. Our study reveals that plasmid transfer during biofilm collisions is determined by spatial factors operating at different organizational levels and length scales, expanding our understanding of the fate of plasmid-encoded traits in microbial communities.

microbiology↗

bioRxiv2023

Dopamine promotes Klebsiella quasivariicola proliferation and inflammatory response in the presence of macrophages

Klebsiella quasivariicola was a novel strain of Klebsiella species and had potential pathogenicity. Our previously studies showed dopamine, one of the most commonly used rescue drugs for critically ill patients, had clear effects on the growth of K. quasivariicola in culture medium, however, its effects on host immune system were ignored. Therefore, in consideration of the host immunity, the interactions of K. quasivariicola, dopamine and macrophages were explored. In this study, RAW264.7 cells and C57/BL6 mice were infected with K. quasivariicola, and the bacterial growth in macrophage, the production of inflammatory cytokines and the pathological changes of mice lung were detected, in the absence or presence of dopamine. Our results showed dopamine inhibited the K. quasivariicola growth in medium, but promoted the bacterial growth when K. quasivariicola was co-cultured with macrophages; The expression of proinflammatory cytokines in K. quasivariicola infected RAW 264.7 were increased, while a sharp rise was observed with the addition of dopamine; Infection of K. quasivariicola to mice induced an inflammatory response and lung injury, which were exacerbated by dopamine administration. It can be concluded that dopamine administration resulted in a significantly increase of K. quasivariicola burdens in the presence of macrophage, consequently, aggravated the inflammatory response and inflammatory injury. ImportanceDopamine is one of the most commonly used rescue drugs for critically ill patients. Here we indicated K. quasivariicola was a potential pathogen of pulmonary infection, and dopamine significantly increased the proliferation of K. quasivariicola when exposed to macrophage, subsequently result in severe inflammatory response and inflammatory injury. We also proposed an in vitro model of microbes-drugs-host immune cells that could better mimic in vivo environment and more suitable for the studies of inhibitor screening. This fundamental work had contributed to the present understanding of the crosstalk between pathogen, dopamine and host immune cells. Furthermore, our data showed dopamine was one of the risk factors for patients with K. quasivariicola infection, which provided a basis for clinical precision medicine.

microbiology↗

bioRxiv2023

Multi-region brain transcriptomes uncover two subtypes of aging individuals with differences in Alzheimer risk and the impact of APOEe4

The heterogeneity of the older population suggests the existence of subsets of individuals which share certain brain molecular features and respond differently to risk factors for Alzheimers disease, but this population structure remains poorly defined. Here, we performed an unsupervised clustering of individuals with multi-region brain transcriptomes to assess whether a broader approach, simultaneously considering data from multiple regions involved in cognition would uncover such subsets. We implemented a canonical correlation-based analysis in a Discovery cohort of 459 participants from two longitudinal studies of cognitive aging that have RNA sequence profiles in three brain regions. 690 additional participants that have data in only one or two of these regions were used in the Replication effort. These clustering analyses identified two meta-clusters, MC-1 and MC-2. The two sets of participants differ primarily in their trajectories of cognitive decline, with MC-2 having a delay of 3 years to the median age of incident dementia. This is due, in part, to a greater impact of tau pathology on neuronal chromatin architecture and to broader brain changes including greater loss of white matter integrity in MC-1. Further evidence of biological differences includes a significantly larger impact of APOE{varepsilon}4 risk on cognitive decline in MC-1. These findings suggest that our proposed population structure captures an aspect of the more distributed molecular state of the aging brain that either enhances the effect of risk factors in MC-1 or of protective effects in MC-2. These observations may inform the design of therapeutic development efforts and of trials as both become increasingly more targeted molecularly. One Sentence Summary: There are two types of aging brains, with one being more vulnerable to APOE{varepsilon}4 and subsequent neuronal dysfunction and cognitive loss.

neuroscience↗

bioRxiv2023

TSD: Transformers for Seizure Detection

Epilepsy is a common neurological disorder that sub-stantially deteriorates patients safety and quality of life. Electroencephalogram (EEG) has been the golden-standard technique for diagnosing this brain disorder and has played an essential role in epilepsy monitoring and disease management. It is extremely laborious and challenging, if not practical, for physicians and expert humans to annotate all recorded signals, particularly in long-term monitoring. The annotation process often involves identifying signal segments with suspected epileptic seizure features or other abnormalities and/or known healthy features. Therefore, automated epilepsy detection becomes a key clinical need because it can greatly improve clinical practices efficiency and free up human expert time to attend to other important tasks. Current automated seizure detection algorithms generally face two challenges: (1) models trained for specific patients, but such models are patient-specific, hence fail to generalize to other patients and real-world situations; (2) seizure detection models trained on large EEG datasets have low sensitivity and/or high false positive rates, often with an area under the receiver operating characteristic (AUROC) that is not high enough for potential clinical applicability. This paper proposes Transformers for Seizure Detection, which we refer to as TSD in this manuscript. A Transformer is a deep learning architecture based on an encoder-decoder structure and on attention mechanisms, which we apply to recorded brain signals. The AUROC of our proposed model has achieved 92.1%, tested with Temple Universitys publically available electroencephalogram (EEG) seizure corpus dataset (TUH). Additionally, we highlight the impact of input domains on the models performance. Specifically, TSD performs best in identifying epileptic seizures when the input domain is a time-frequency. Finally, our proposed model for seizure detection in inference-only mode with EEG recordings shows outstanding performance in classifying seizure types and superior model initialization.

bioengineering↗