Biology subjects

Joo, J.

Publications and source records attributed to Joo, J..

7 recordsLinked to original sources

bioRxiv2026

Limits of marker information transfer]{Limits of marker information transfer in cross-panel cytometry imputation

Marker imputation can bridge cytometry datasets acquired with different panels and platforms. Its validity depends on whether common markers retain information about unmeasured target markers and whether those relationships transfer across assay contexts. We evaluated these limits in two paired flow cytometry-CyTOF cohorts comprising 134 patients and 257 specimens. A patient-held-out framework distinguished within-platform same-cell reconstruction from cross-platform distribution translation without cell correspondence. We compared conventional machine-learning and deep generative estimators with median imputation and used permutation controls to estimate the contributions of patient matching and source-exclusive markers. Evaluation covered marker values, specimen distributions, phenotype utility, and directional predictive relationships among markers. Within platforms, exact k-nearest neighbors reduced mean absolute error by 12.8-35.0% relative to median imputation. However, the median-based reconstructions were contracted, with a median ratio of predicted to observed interquartile range below 1 in every context and evaluable population. Patient matching improved cross-platform distribution recovery, whereas adding source-exclusive markers or higher-capacity estimators did not consistently produce net gains. In the independently annotated acute myeloid leukemia (AML) cohort, reconstructed markers largely preserved broad-lineage classification but did not recover fine-phenotype information available through direct measurement. In AML-CyTOF, the measured full panel outperformed the reconstructed representation by 10.25 percentage points for fine phenotypes. Predictive relationships among markers were context-specific. Most targets were better explained by several markers than by one. Marker selection based on these relationships showed a trade-off between panel size and pairwise coverage. Paired samples and imputation can therefore connect selected marker distributions across panels but should not be treated as substitutes for direct measurement. Cross-panel cytometry should be evaluated with patient-held-out tests matched to the resolution of available ground truth and the intended downstream use.

bioinformatics↗

bioRxiv2025

Strain-specific differences in toxT expression and virulence gene activation in Vibrio cholerae

The transcriptional activator ToxT is a key regulator of Vibrio cholerae virulence, controlling the expression of cholera toxin (CT) and the toxin co-regulated pilus (TCP). Insights into the regulation and function of toxT have largely come from studies on the classical biotype strain O395, which exhibits robust toxT expression. However, toxT expression is not fully consistent across V. cholerae strains. Here, we show that toxT transcription and protein production differ significantly among strains, particularly in El Tor biotype, which exhibit allele-dependent regulatory constraints. Using qRT-PCR and Western blotting, we demonstrate that alternative toxT alleles confer distinct transcriptional activator functions across El Tor biotype strains. Notably, IB5230, a pandemic El Tor strain, exhibits enhanced toxT and virulence gene expression even at 37{degrees}C, suggesting a potential adaptation for increased pathogenicity. Our results suggest that toxT autoregulatory feedback plays a more critical role in virulence gene activation than previously recognized. This study underscores previously unrecognized complexity in toxT regulation, emphasizing the need to reassess virulence control mechanisms in epidemic and emerging V. cholerae strains. Understanding these strain-specific differences is essential for refining models of cholera pathogenesis. Author SummaryCholera is a severe diarrheal disease caused by the bacterium Vibrio cholerae, which continues to pose a major threat to global health. The ability of V. cholerae to cause disease depends on the expression of two key virulence factors--cholera toxin (CT) and the toxin-coregulated pilus(TCP)-- controlled by the transcriptional activator ToxT. While much of our understanding of ToxT regulation is based on classical biotype strains, most recent cholera outbreaks have been driven by El Tor biotype strains. In this study, we demonstrate that toxT expression and its ability to activate virulence genes vary markedly across V. cholerae strains, depending on both the specific toxT allele and the genetic background. Using isogenic strains engineered to express different toxT alleles, we show that some alleles are transcriptionally silent in certain El Tor strains while others are active, despite having similar functional capabilities when expressed. Notably, a hypervirulent strain linked to the Haitian cholera outbreak maintained high toxT expression even at human body temperature, hinting at possible adaptive mechanisms for enhanced pathogenicity. Our findings suggest that virulence regulation in V. cholerae may be more variable than previously recognized, highlighting strain- and allele-specific differences in toxT expression. Understanding these strain- and allele-specific differences in virulence gene regulation is crucial for improving models of cholera pathogenesis and could inform the development of more effective interventions and vaccines against this evolving pathogen.

microbiology↗

bioRxiv2023

SUBTLE: An unsupervised platform with temporal link embedding that maps animal behavior

While huge strides have recently been made in language-based machine learning, the ability of artificial systems to comprehend the sequences that comprise animal behavior has been lagging behind. In contrast, humans instinctively recognize behaviors by finding similarities in behavioral sequences. Here, we develop an unsupervised behavior-mapping framework, SUBTLE (spectrogram-UMAP-based temporal-link embedding), to capture comparable behavioral repertoires from 3D action skeletons. To find the best embedding method, we devise a temporal proximity index as a metric to gauge temporal representation in the behavioral embedding space. The method achieves the best performance compared to current embedding strategies. Its spectrogram-based UMAP clustering not only identifies subtle inter-group differences but also matches human-annotated labels. SUBTLE framework automates the tasks of both identifying behavioral repertoires like walking, grooming, standing, and rearing, and profiling individual behavior signatures like subtle inter-group differences by age. SUBTLE highlights the importance of temporal representation in the behavioral embedding space for human-like behavioral categorization. One Sentence SummaryUnsupervised behavior-mapping from 3D action skeletons achieves superior performance, captures behavioral repertoires, and identifies inter-group differences, emphasizing how temporal representation is critical in the behavioral embedding space.

neuroscience↗

bioRxiv2022

Rebound Excitation of Epileptiform Activities by Transcranial Focused Ultrasound Stimulation

BackgroundConventional neurostimulations to treat epilepsy have adverse effects caused by post-inhibitory rebound excitations. Although ultrasound brain stimulation is feasible in inducing anticonvulsant effects, its association with paradoxical rebound excitations is unknown. ObjectiveThis study aimed at demonstrating rebound excitations with transcranial focused ultrasound. The modulations of epileptiform activities toward both suppressive and excitatory responses were investigated by changing ultrasonic transmit sequences. MethodsIn a pentylenetetrazol-injected acute epilepsy rat model, transcranial focused ultrasound stimulation was applied on the thalamus to modulate epileptiform activities. The parameters differentiated for pulse sequences were the pulse length, pulse pressure, and interval between the pulses. Sonication effects were assessed by electroencephalography (n=38), immuno-histochemical analysis (n=24), and optical measurement of cerebral blood volume changes (n=18). ResultsWhile ultrasonic patterns of stimuli at long intervals showed antiepileptic effects on electroencephalography, those at short intervals showed rebound excitatory responses followed by inhibitory activities. Further, suppressive states induced by inhibitory stimulations were transformed into excitatory states by applying a consecutive series of short bursts at higher acoustic pressure. Cerebral blood volume changes demonstrated consistent results with electroencephalography. Immunohistochemistry revealed that both inhibitory and excitatory neuronal cells were activated to generate rebound excitatory conditions, while inhibitory cells were activated for suppressive conditions. ConclusionsIn our study, variations in ultrasound stimulation patterns could modulate epileptiform activities in both upregulated and downregulated directions.

bioengineering↗

bioRxiv2020

Nanoparticle-mediated microRNA-145 Delivery for Vascular Smooth Muscle Cell Phenotype Modulation and Atherosclerosis Treatment

Vascular smooth muscle cells (VSMCs) change from contractile to the synthetic phenotype during atherogenesis and 30-70% of cells that make up plaques have been elucidated to be of VSMC origin. MicroRNA-145 (miR-145) is responsible for regulating VSMC phenotypic switching, and low miR-145 levels in circulation have been linked with atherosclerosis. Hence, we developed nanoparticles for targeted delivery of miR-145 by synthesizing micelles co-assembled with miR-145 and the CCR2-binding peptides for plaque targeting. The miR cargo was protected in micelles from premature endosomal degradation and rescued contractile markers in synthetic VSMCs and SMCs isolated from patient arteries in vitro. In ApoE-/- mid-stage atherosclerotic mice, miR-145 micelles halted plaque growth and maintained contractile phenotypes similar to baseline levels. In early-stage atherosclerosis, a single dose of miR-145 micelles prevented lesion growth by 49%. We present the potential of miR-145 micelles as a therapeutic that can be applied longitudinally and intervene throughout atherosclerosis pathogenesis.

bioengineering↗

bioRxiv2020

Increased capillary stalling is associated with endothelial glycocalyx loss in subcortical vascular dementia

Proper regulation and patency of cerebral microcirculation is crucial for maintaining a healthy brain. Capillary stalling, i.e., the brief interruption of microcirculation mainly by leukocytes, has been observed in several diseases and contributes to disease pathogenesis or progression. However, the underpinning mechanism for leukocyte capillary plugging remains elusive. Therefore, we investigated the mechanism of capillary stalling in mice during the development of subcortical vascular dementia (SVaD), the most common type of vascular dementia characterized by impaired microcirculation and associated pathological features. Longitudinal optical coherence tomography angiography showed increased number of stalled segments as the disease progressed, while two-photon microscopy indicated a less extensive endothelial glycocalyx (EG) in the stalled segments. We also found that increased gliosis and blood-brain barrier leakage were correlated with the increased number of stalled segments. Based on the above, we conclude that EG potentially mediates capillary stalling and can be a therapeutic target of SVaD.

neuroscience↗

bioRxiv2017

Inference about causation between body mass index and DNA methylation in blood from a twin family study

BackgroundSeveral studies have reported DNA methylation in blood to be associated with body mass index (BMI), but only a few have investigated causal aspects of the association. We used a twin family design to assess this association at two life points and applied a novel analytical approach to investigate the evidence for causality.\n\nMethodsThe methylation profile of DNA from peripheral blood was measured for 479 Australian women (mean age 56 years) from 130 twin families. Linear regression was used to estimate the associations of methylation at ~410 000 cytosine-guanine dinucleotides (CpG), and of the average methylation at ~20 000 genes, with current BMI, BMI at age 18-21 years, and the change between the two (BMI change). A novel regression-based methodology for twins, Inference about Causation through Examination of Familial Confounding (ICE FALCON), was used to assess causation.\n\nResultsAt 5% false discovery rate, nine, six and 12 CpGs at 24 loci were associated with current BMI, BMI at age 18-21 years and BMI change, respectively. The average methylation of BHLHE40 and SOCS3 loci was associated with current BMI, and of PHGDH locus was associated with BMI change. From the ICE FALCON analyses with BMI as the predictor and methylation as the outcome, a womans methylation level was associated with her co-twins BMI, and the association disappeared conditioning on her own BMI, consistent with BMI causing methylation. To the contrary, using methylation as the predictor and BMI as the outcome, a womans BMI was not associated with her co-twins methylation level, consistent with methylation not causing BMI.\n\nConclusionFor middle-aged women, peripheral blood DNA methylation at several genomic locations is associated with current BMI, BMI at age 18-21 years and BMI change. Our study suggests that BMI has a causal effect on peripheral blood DNA methylation.

epidemiology↗