Biology subjects

Cho, E. S.

Publications and source records attributed to Cho, E. S..

3 recordsLinked to original sources

bioRxiv2026

Agentic AI enables cell segmentation-free cell-type inference and cell-type-guided cell segmentation in spatial transcriptomics

Cell-type identification in spatial transcriptomics often relies on cell segmentation, which can introduce inaccuracies in cell boundaries and transcript assignment. Segmentation-free or cell-free methods have been developed to completely avoid cell segmentation, but inaccurate cell typing hinder accurate downstream analysis. Here, we introduce SSAF-T, a segmentation-free agentic framework that identifies cell types directly from spatial gene expression without predefined reference datasets. For the reference-free cell typing, SSAF-T incorporates CeLLMan, which employs expert-level agents for each organ to annotate cell types using a literature-grounded knowledge graph. We benchmark SSAF-T on both array-based and imaging-based spatial transcriptomics datasets in terms of accuracy, reproducibility, and scalability. We further show that SSAF-T provides an accurate cell segmentation based on the segmentation-free mapping, which ensure compatibility to existing downstream analysis ecosystem, and eventually enable quantitative analysis at single-cell level.

bioinformatics↗

bioRxiv2022

Emergence of glycogen synthase kinase-3 interaction domain enhances phosphorylation of SARS-CoV-2 nucleocapsid protein

A structural protein of SARS-CoV-2, nucleocapsid (N) protein is abundantly expressed during viral replication. The N protein is phosphorylated by glycogen synthase kinase (GSK)-3 on the serine/arginine (SR) rich motif located in disordered regions. Although phosphorylation by GSK-3{beta} constitutes a critical event for viral replication, the molecular mechanism underlying N phosphorylation is not well understood. In this study, we found the putative alpha-helix L/FxxxL/AxxRL motif known as the GSK-3 interacting domain (GID), commonly found in many endogenous GSK-3{beta} binding proteins, such as Axins, FRATs, WWOX and GSKIP. Indeed, N interacts with GSK-3{beta} similarly to Axin, and Leu to Glu substitution of the GID abolished the interaction, with loss of N phosphorylation. Unlike with endogenous GID proteins, the N interaction neither disturbs endogenous GSK-3 activity nor regulates subsequent canonical Wnt activity and the Snail-EMT program. Notably, N abundance in SARS-CoV-2 is incomparably high compared to other coronaviruses, such as 229E, OC43 and HKU1. Compared to other coronaviruses, N harbors a CDK1 primed phosphorylation site and Gly-rich linker for enhanced phosphorylation by GSK-3{beta}. Furthermore, we found that the S202R mutant found in Delta and R203K/G204R mutant found in the Omicron variant allows increased abundance and hyper-phosphorylation of N. Our observations suggest that the emergence of GID and mutations for increased phosphorylation in N may have contributed to the emergence of SARS-CoV-2 and evolution of variants, respectively. Further study, especially in a BSL3-equipped facility, is required to elucidate the functional importance of GID and N phosphorylation in SARS-CoV-2 and variants.

cell biology↗

bioRxiv2020

Distinct VIP interneurons in the cingulate cortex encode anxiogenic and social stimuli

A hallmark of higher-order cortical regions is their functional heterogeneity, but it is not well understood how these areas encode such diverse information. The anterior cingulate cortex (ACC), for example, is important in both emotional regulation and social cognition. Previous work shows activation of the ACC to anxiety-related and social stimuli, but it is unknown how subpopulations or microcircuits within the ACC simultaneously encode these distinct stimuli. One type of inhibitory interneuron, which is positive for vasoactive intestinal peptide (VIP), is known to alter the activity of many cells in local cortical microcircuits, but it is unknown whether the activity of VIP cells in the ACC (VIPACC) encodes anxiety-related or social information. Using in vivo calcium imaging and miniscopes in freely behaving mice to monitor VIPACC activity, we identified distinct, non-overlapping subpopulations of VIPACC that preferentially activated to either anxiogenic, anxiolytic, social, or non-social stimuli. We determined that stimulus-selective cells encode the animals behavioral states and VIP interneuron clusters may co-activate, improving this encoding. Finally, we used trans-synaptic tracing to show that VIPACC receive widespread inputs from regions implicated in emotional regulation and social cognition. These findings demonstrate not only that the ACC is not homogeneous in its function, but also that there is marked functional heterogeneity even within disinhibitory interneuron populations. This work contributes to our understanding of how the cortex encodes information across diverse contexts and provides insight into the complexity of neural processes involved in anxiety and social behavior.

neuroscience↗