Biology subjects

Wang, C.

Publications and source records attributed to Wang, C..

At least 271 records · Page 15Linked to original sources

bioRxiv2025

A Bayesian Informative Shrinkage Approach for Large-scale Multiple Hypothesis Testing (BISHOT): with Applications in Differential Analysis of Omics Data

A major goal of many omics studies is to identify differential features, e.g. differentially expressed genes, between experimental groups. When performing differential analysis for a given dataset, relevant information from another platform or species is often available. Incorporating such prior information can help identify features that show consistent differential patterns across platforms or species, which are more likely to reflect shared biological processes, and thereby enhance the robustness and generalizability of the findings. However, existing differential analysis methods typically analyze only the data from the current study and do not leverage prior knowledge about the magnitude or direction of changes from other platforms or species. We address this challenge, and the associated multiple testing problem, using a Bayesian framework that enables the incorporation of prior knowledge obtained from different platforms or species. We propose a new test statistic, Bayesian Credible Ratio (BCR), based on a heteroscedastic global local shrinkage prior, and a new multiple testing criterion, sign-adjusted FDR (SFDR), that emphasize information regarding the direction of the differentially features. We prove that BCR achieves the largest count of sign-based true positives among all legitimate SFDR-controlling methods. Simulation results offer numerical evidence of its advantage compared to an empirical Bayesian method. The approach is demonstrated through the analysis of RNAseq and single-cell RNAseq datasets.

genetics↗

bioRxiv2025

Maximized field-of-view deep-brain calcium imaging through gradient-index lenses

Advances in genetically encoded fluorescent indicators have enabled increasingly sensitive optical recordings of neural activity. However, light scattering in the mammalian brain tissue restricts optical access to deeper regions. To address this limitation, researchers often employ implanted gradient-index (GRIN) lenses to reach deep brain areas. Nevertheless, the severe optical aberrations of GRIN lenses significantly reduce the effective field of view (FOV). In this work, we present a simple and robust imaging approach that combines low-NA telecentric scanning (LNTS) of laser excitation with high-NA fluorescence collection to increase the FOV. This configuration effectively eliminates common aberrations such as astigmatism and field curvature, resulting in a FOV [~]100% as large as the GRIN lens facet area -- corresponding to a [~]400% increase in imaging area compared with conventional approaches. We validate this method through both structural and functional in vivo imaging. The highly consistent imaging performance, fully maximized imaging FOV, and the very simple optical design make this method well-suited for broad dissemination in neuroscience research.

neuroscience↗

bioRxiv2025

ODC1 restricts meningeal B cell age-associated-like phenotype and function in multiple sclerosis: a human and experimental study

Meningeal inflammation, as a clinical feature of multiple sclerosis (MS), is associated with worse clinical disease outcomes. In both relapsing and secondary progressive MS and the experimental autoimmune encephalomyelitis (EAE) MS model, the meninges have been found to contain ectopic lymphoid follicles enriched with B cells. The metabolic requirement of meningeal B cell function in MS or EAE is not well elucidated. Using 7-Tesla MRI brain scans of MS patients and leptomeningeal enhancement (LME) as a marker, we found a correlation between meningeal inflammation and metabolites of the arginine/polyamine pathway, a finding recapitulated in the EAE model. Ornithine Decarboxylase (ODC1), the rate limiting enzyme for polyamine biosynthesis, as well as polyamine metabolism was diminished in the dura meningeal B cells from mice with MOG35-55 induced EAE mice as compared to naive controls. Pharmacological inhibition of ODC1 restricted meningeal T cells but promoted meningeal B cell proliferation. B cell-specific deletion of ODC1 resulted in expansion of B cells with age-associated B cell-like phenotype (CD11c+CD21/35-CD23-IgD-) and exacerbated disease in the MOG1-125 EAE model. Together, these findings demonstrate a divergent role of polyamines in regulating B and T cell responses in the meninges during autoimmunity. Significance StatementThis study identified the polyamine pathway to be associated with meningeal inflammation in multiple sclerosis, a clinical phenotype associated with worse disease outcomes and without targeted therapy. Using a mouse experimental model, we found that ODC1, the rate limiting enzyme of the polyamine biosynthesis pathway, was suppressed in meningeal B cells, restricted the development of age-associated B cells in the meninges and limited disease severity. This study elucidated a metabolic pathway regulating meningeal B cell function, informing its therapeutic applications in autoimmune diseases.

immunology↗

bioRxiv2025

Benchmarking long-read RNA-sequencing technologies with LongBench: a cross-platform reference dataset profiling cancer cell lines with bulk and single-cell approaches

Long-read RNA sequencing enables full-length transcript profiling and improved isoform resolution, but variable platforms and evolving chemistries demand careful benchmarking for reliable application. We present LongBench, a matched, multi-platform reference dataset spanning bulk, single-cell, and single-nucleus transcriptomics across eight human lung cancer cell lines with synthetic spike-in controls. LongBench in-corporates three state-of-the-art long-read protocols alongside Illumina short reads: Oxford Nanopore Technologies (ONT) PCR-cDNA, ONT direct RNA, and PacBio Kinnex. We systematically evaluate transcript capture, quantification accuracy, differential expression, isoform usage, variant detection, and allele-specific analyses. Our results show high concordance in gene-level differential analyses across protocols, but reduced consistency for transcript-level and isoform analyses due to length- and platform-dependent biases. Single-cell long-read data are highly concordant with bulk for high-confidence features, though single-nuclei data show reduced feature detection. LongBench provides one of the largest publicly available long-read benchmarking resources, enabling rigorous cross-platform evaluation and guiding technology selection for transcriptomic research.

genomics↗

bioRxiv2025

ATF4 alleviates doxorubicin-induced cardiomyopathy through H2S-mediated antioxidation

BackgroundDoxorubicin (DOX) remains a cornerstone in the treatment of various cancers. However, its clinical utilization is significantly hampered by dose-dependent cardiotoxicity. The generation of reactive oxygen species (ROS) constitutes the central component of the pathogenesis of DOX-induced cardiotoxicity. Activating transcription factor 4 (ATF4) has been demonstrated to exert a cardioprotective effect and augment cardiac antioxidative capacity in settings of heart failure. However, the role of ATF4 in DOX-induced cardiomyopathy (DIC) remains unknown. MethodsTo explore the role of ATF4 in DOX-induced cardiomyopathy, cardiac-specific ATF4 conditional heterozygous mice and AAV9 mediated ATF4 overexpression mouse models were utilized. Cardiac function was assessed by echocardiography. The upstream regulator and downstream mediator of ATF4 were evaluated using RNA-seq analysis and further verified using ChIP assay and luciferase reporter assay. ResultsWe found a substantial decrease in ATF4 expression levels in the heart of DIC mice. ATF4+/- mice exhibited a higher degree of susceptibility to DOX-induced cardiotoxicity in comparison with ATF4flox/flox mice, as evidenced by the manifestation of more severe cardiac dysfunction and a significantly earlier mortality rate. In contrast, cardiacc-specific overexpression ATF4 by AAV9 confers robust cardioprotection against DOX-induced cardiomyopathy. Mechanistically, we identified the upstream regulator of ATF4 as KLF16, which was significantly suppressed during DOX treatment. Further, the decrease of ATF4 led to a reduction in cystathionine {gamma}-lyase (CSE) transcription and hydrogen sulfide (H2S) production in the context of DOX-induced cardiotoxicity. ChIP and luciferase reporter assays revealed that ATF4 functioned as the transcription factor of the CSE gene, which is a key enzyme in the synthesis of H2S to counteract oxidative stress. Consistently, ROS scavengers or H2S donors was shown to mitigate the consequences of ATF4 deficiency. In contrast, the ectopic expression of ATF4 mitigated oxidative stress and apoptosis in DOX-induced cardiotoxicity, both in vivo and in vitro. ConclusionsOur study revealed a novel function of ATF4 in counteracting oxidative stress in DOX cardiotoxicity by promoting the transcription of CSE. ATF4 may represent a promising therapeutic target for the treatment of DOX-induced cardiomyopathy.

pathology↗

bioRxiv2025

Biodegradation of the endocrine-disrupting compound bisphenol F by Sphingobium yanoikuyae DN12

Bisphenol F (BPF) is an emerging environmental pollutant widely present in surface water and wastewater systems. Microbial activity is crucial in driving its degradation, offering a potential avenue for mitigating its environmental impact. Although the degradation pathway for BPF has been identified in various bacteria, the biodegradation mechanisms remain unclear. In this study, we isolated a highly efficient BPF-degrading strain of Sphingobium yanoikuyae DN12, which could utilize BPF as the sole carbon source and energy source for growth, from a river sediment in Anhui Province China. Through Ultra performance liquid chromatography high-resolution mass spectrometry (UPLC-HRMS) analysis, we found that oxidation and hydrolysis are key steps for BPF biodegradation. Utilizing whole-genome sequencing, comparative transcriptomics analysis and biochemical identification, a gene cluster bpf was identified to be involved in BPF degradation. BpfAB is a two-component oxidoreductase responsible for converting BPF to 4,4-dihydroxybenzophenone (DHBP). BpfC is a Baeyer-Villiger monooxygenase (BVMO) responsible for converting DHBP to 4-hydroxyphenyl-4-hydroxybenzoate (HPHB). Isotope tracing demonstrated that the oxygen atom incorporated by BpfAB originates from water, whereas that incorporated by BpfC derives from molecular oxygen (O2). BpfD is an /{beta} hydrolase responsible for converting HPHB to 4-hydroxybenzoate (4HB) and 1,4-hydroquinone (HQ). Analysis of the taxonomic and habitat of 325 prokaryotic genomes revealed that BpfA-like homologs are predominantly found in the phylum Pseudomonadota, primarily inhabiting soil and aquatic environments. This study enhances our understanding of the biodegradation mechanism of BPF, and provides guidance for the effective remediation of BPF-contaminated environments. IMPORTANCEBPF is a widely used alternative to bisphenol A and poses a growing threat to ecosystems and human health due to its environmental persistence and endocrine-disrupting effects. Although microbial degradation pathways for BPF have been reported, the key enzymes involved and their catalytic mechanisms remain unclear. This work reports the isolation of a Sphingobium strain capable of mineralizing BPF and the genetic basis for the catabolic pathway. Three enzymes--a two-component oxidoreductase, a Baeyer-Villiger monooxygenase, and an /{beta} hydrolase--were biochemically characterized and shown to catalyze the three critical steps in BPF degradation. These findings provide insights into the biochemical processes involved in the microbial degradation of BPF.

microbiology↗

bioRxiv2025

Mitochondrial dysfunction precedes neurodegeneration in DRPLA patient-derived neurons, and phenylbutyrate improves survival.

Dentatorubral-pallidoluysian atrophy (DRPLA) is a progressive autosomal-dominant neurodegenerative disorder caused by a CAG repeat expansion in the ATN1 gene, which encodes the transcriptional corepressor Atrophin-1. The temporal sequence of molecular mechanisms driving neuronal dysfunction and degeneration in DRPLA is poorly understood, limiting therapeutic development. We generated patient-derived induced pluripotent stem cells and differentiated them into cortical excitatory glutamatergic neurons to model early pathogenic processes. Early pathological features of DRPLA patient-derived neurons included mitochondrial dysfunction and oxidative stress. These alterations occurred before overt neuronal loss, highlighting bioenergetic stress as a key early driver of disease progression toward neurodegeneration. Pharmacological treatment with phenylbutyrate significantly improved neuronal survival and reduced mitochondrial reactive oxygen species production, demonstrating the therapeutic potential of targeting mitochondrial dysfunction and oxidative stress. These findings challenge the conventional aggregation-centric model of polyglutamine disease pathogenesis and position mitochondrial stress as a central and early promoter of neuronal degeneration in DRPLA. By providing mechanistic insight into early-stage disease processes, our study lays the foundation for therapeutic strategies targeting mitochondrial dysfunction in DRPLA and related polyglutamine disorders.

neuroscience↗

bioRxiv2025

Chemo-mechanotherapy of fibrosis: dynamic control of biological materials through tissue-architecture-dependent crosslink disruption

Extracellular matrix (ECM) stiffness drives cell fate, but tissues are nonlinear and difficult to modify chemically without harming patients. Here, we show that nonlinear mechanical responses of fibrous and porous tissues can be controlled by mechanically modulating crosslinks. Through integrated in vivo, in vitro, and computational approaches, we found that physiological dynamic stretching disrupts pathological advanced glycation end-product (AGE) crosslink-induced fibrogenesis in an architecture-dependent manner. This was effective in lung-like porous scaffolds but not in liver-like fibrous matrices, revealing a critical structure-property relationship. Computational and in vitro models established that this modulates cell-ECM feedback, and that controlling mechanical crosslinking modulates key nonlinearity of networked solids. This discovery inspired a non-invasive, "mechanotherapuetic" ventilation protocol, which reversed pulmonary fibrosis in mice by physically disrupting ECM pathological crosslinks. The therapeutic efficacy was further amplified when combined with pharmacological AGE inhibition. These findings establish design principles for dynamically reprogrammable biological materials and mechano-targeted therapies.

bioengineering↗

bioRxiv2025

Real-Time Viability Assessment of Ex Vivo Mouse Kidneys for Transplant Applications Using Dynamic Optical Coherence Tomography

Kidney transplantation remains the preferred treatment for patients with end-stage kidney disease. However, the ongoing shortage of donor organs continues to limit the availability of transplant treatments. Existing evaluation methods, such as the kidney donor profile index (KDPI) and pre-transplant donor biopsy (PTDB), have various limitations, including low discriminative power, invasiveness, and sampling errors, which reduce their effectiveness in organ quality assessment and contribute to the risk of unnecessary organ discard. In this study, we explored the dynamic optical coherence tomography (DOCT) as a real-time, non-invasive approach to monitor the viability of ex vivo mouse kidneys during static cold storage over 48 hours. The dynamic metrics logarithmic intensity variance (LIV), early OCT correlation decay speed (OCDSe), and late OCT correlation decay speed (OCDSl) were extracted from OCT signal fluctuations to quantify temporal and spatial tissue activity and deterioration. Our results demonstrate that DOCT provides complementary functional information to morphological assessment offered by conventional OCT imaging, showing potential to improve pre-transplant organ evaluation and clinic decision-making.

bioengineering↗

bioRxiv2025

Structural Role of Stomatin in Organizing Functional Membrane Microdomains

Stomatin, originally discovered in red blood cells, is a member of the SPFH (Stomatin, Prohibitin, Flotillin, and HflK/C) protein family, which has long been proposed to scaffold functional membrane microdomains (FMMs) enriched in saturated lipids such as cholesterol and sphingomyelin. Stomatin has been reported to associate with a variety of proteins involved in diverse physiological processes, including ion channel regulation, membrane fusion, mechanosensory regulation and vesicle trafficking; however, the mechanisms by which it modulates these interactions remain poorly understood. Here, we determined the cryo-electron microscopy (cryo-EM) structures of stomatin, revealing its hexadecameric assemblies. The SPFH1 domains insert into the cytosolic leaflet of membranes and, together with the N-terminal hairpin, form a potential cholesterol-binding pocket. Liposome reconstitution experiments demonstrated that stomatin self-organizes into clusters, and Spectrum and Polarization Optical Tomography (SPOT) further showed that these clusters enhance membrane lipid order, supporting the proposed role of stomatin in organizing functional membrane microdomains (FMMs). Proteomic mass spectrometry analysis identified numerous stomatin-associated cargo proteins, including solute carrier (SLC) transporters, RAB GTPases, and integrins, suggesting that stomatin regulates solute transport and contributes to vesicle trafficking and cell migration. Together, these findings elucidate the structural and functional roles of stomatin and underscore its potential as a therapeutic target for modulating cancer cell migration and metastasis.

biophysics↗

bioRxiv2025

Site- and Structure-Specific Characterization of Glycoproteins of H11: Potent Vaccine Candidates against Parasitic Worm Haemonchus

Parasitic worms (helminths) infections pose significant threats to global health and livestock economies. While mass spectrometry (MS)-based glycomics have revealed that helminths express complex, immunomodulatory N-glycans, site- and structure-specific characterization of intact glycopeptides remain challenging. Here we employed advanced MS-based intact glycoproteomics to explore the N-glycosylation profile of H11 antigen - an important vaccine antigen from pathogenic parasite Haemonchus contortus. We successfully identified seven glycosylated aminopeptidases carrying 19 N-glycosylation sites with 31 distinct N-glycan structures. Notably, 13 N-glycopeptides were significantly enriched by H11-induced protective IgG antibodies. Our findings revealed a high degree of structural heterogeneity and abundant core fucosylation among the identified N-glycopeptides. Additionally, molecular docking studies demonstrated those IgG-recognized N-glycopeptides are situated on the protein surface or adjacent to the substrate-access channels, strongly indicating their potential as antigenic epitopes. Overall, our work represents the first comprehensive glycoproteome of an economically important parasitic worm. These results hold important implications for the rational design of vaccines against H. contortus and other related metazoan parasites.

pathology↗

bioRxiv2025

Variation in Synaptic Inputs Drives Functional Versatility for Sound Encoding in Globular Bushy Cells

Synaptic convergence is fundamental to neuronal circuit function, underpinning computations such as coincidence detection and signal integration. Across sensory systems, convergence architecture and synaptic input strengths are key for extracting stimulus features and processing of sensory information. In the cochlear nucleus, globular bushy cells (GBCs) receive convergent inputs from multiple auditory nerve fibers via large endbulb of Held terminals. While these inputs vary considerably in size, even among those targeting the same cell, the functional consequences of this variation for sound encoding remain unclear. Here, we investigated how synaptic input variation shapes sound encoding in GBCs of Mongolian gerbils using in vitro conductance-clamp recordings and computational modeling. By simulating synaptic inputs with variable strength distributions, we found that increasing input variation enhances rate coding at the expense of temporal precision. These findings suggest that endbulb strength heterogeneity allows the GBC population to operate along a functional continuum, generating diverse information streams to downstream targets.

neuroscience↗

bioRxiv2025

Leveraging Large Language Models for Redundancy-Aware Pathway Analysis and Deep Biological Interpretation

Extracting coherent, biologically meaningful insights from vast, complex multi-omics data remains challenging. Currently, pathway enrichment analysis serves as a cornerstone for the functional interpretation of such data. However, conventional approaches often suffer from extensive functional redundancy caused by shared molecular components and overlapping pathway definitions across databases. This redundancy can obscure key biological signals and compromise the interpretability of pathway enrichment results. Here, we present MAPA (Functional Module Identification and Annotation for Pathway Analysis Results Using Large Language Models [LLM]), an open-source computational framework that resolves redundancy and enhances pathway analysis result interpretation. MAPA computes functional similarity between pathways using LLM-based text embeddings, enabling comparison across different databases. It constructs pathway similarity networks and identifies functional modules via community detection algorithms. Crucially, MAPA employs LLMs for automated functional annotation, integrating Retrieval-Augmented Generation (RAG) to generate comprehensive and real-time biological summaries and reduce hallucinations. Benchmarking demonstrated MAPAs superior performance: the biotext embedding similarity showed a large effect size (Cliffs {delta} = 0.96) compared with the Jaccard index ({delta} = 0.73), and module identification achieved high accuracy (Adjusted Rand Index [ARI] = 0.95) versus existing methods (ARI = 0.23-0.33). Human expert evaluation confirmed that MAPAs annotations match expert-quality interpretations. Finally, a multi-omics aging case study illustrates that MAPA uncovers coherent functional modules and generates insights extending beyond conventional pathway analyses. Collectively, MAPA represents a significant advance in redundancy-aware pathway analysis, transforming pathway enrichment results from fragmented lists into biologically coherent narratives. By leveraging the capabilities of LLMs, MAPA offers researchers a robust, scalable tool for deriving deep mechanistic insights from complex and vast multi-omics datasets, marking a new direction for AI-driven bioinformatics.

bioinformatics↗

bioRxiv2025

An activated wheat CCG10-NLR immune receptor forms an octameric resistosome

Nucleotide-binding and leucine-rich repeat (LRR) receptors (NLRs) are widespread intracellular immune sensors across kingdoms. In plants, the G10-type coiled-coil (CCG10)-NLRs form a distinct phylogenetic clade that remains poorly characterized. Here, we identified a gain-of-function (GOF) mutant of Wheat Autoimmunity 3 (WAI3), designated WAI3GOF, which encodes a constitutively activated CCG10-NLR protein due to an amino acid substitution. Cryo-EM structural analysis revealed that activated WAI3 assembles into a distinctive octameric resistosome. Arabidopsis RPS2, another CCG10-NLR, also forms an octamer, indicating a conserved structural property across monocot and dicot plants. The WAI3 resistosome mediates a prolonged and sustained increase in cytosolic calcium influx, facilitated by a unique channel architecture arising from its divergent CC domain configuration. Notably, this domain arrangement may be shared by many plant NLRs that lack the conserved EDVID motif in their CC domains. Our findings uncover a previously uncharacterized resistosome structure and provide insights into plant immune receptor plasticity.

plant biology↗

bioRxiv2025

Capturing trophectoderm-like stem cells enables step-wisely remodeling of placental development

The trophectoderm produced from totipotent blastomeres initiates trophoblast development, while placental deficiencies can cause pregnancy disorders. Yet, a culture system that fully recapitulates the entire placenta development is still lacking, greatly limiting related studies. Here, we captured mouse trophectoderm-like stem cells (TELSCs), which can give rise to all trophoblast lineages and be applied to generate trophoblast organoids. We achieved the induction and maintenance of TELSCs from totipotent blastomere-like stem cells or early embryos through a Hippo-YAP/Notch-to-TGF{beta}1 signaling switch. At the molecular level, TELSCs resemble E4.5 trophectoderm and are distinct from all previously known trophoblast-like stem cells. Functionally, TELSCs can generate all trophoblast lineages in both teratoma and chimera assays. We further applied TELSCs to generate trophoblast organoids containing various mature trophoblasts and a self-renewing extraembryonic ectoderm (ExE)-like progenitor population. Interestingly, we observed transiently formed rosette-like structures that rely on Itgb1, which are essential to induce ExE-like progenitors and to generate organoids eventually. Thus, the capture of TELSCs enables comprehensive insights into placental development. HighlightsO_LITELSCs were robustly induced and long-term maintained from TBLCs and early embryos using a newly developed Hippo-YAP/Notch-to-TGF{beta}1 signaling switch strategy C_LIO_LITELSCs resemble E4.5 trophectoderm at the molecular level and generate all trophoblast lineages both in vivo and in vitro functionally C_LIO_LITELSCs can be applied to generate trophoblast organoids containing multiple mature subtypes and a self-renewing ExE-like population C_LIO_LIRosette structure formation that relies on Itgb1 is critical to induce ExE-like progenitors and to eventually form trophoblast organoids C_LI

cell biology↗

bioRxiv2025

Lipid Droplets of Human Carotid Atherosclerotic Plaques: Cholesteryl Arachidonate Core and Cytoskeletal Fibrotic Cover

Lipid droplets (LDs) within atherosclerotic plaques are central to cardiovascular disease, yet their molecular composition remains largely unknown. More than 20 years experiences of developing LD purification method allowed us recently to successfully purify LDs from atherosclerotic plaques of patients undergoing carotid artery surgery, including three types, soft (lipid), mixed, and hard (calcified). In addition, the subpopulations of large LDs and small LDs were purified and studied, respectively. We report the first comprehensive proteo-lipidomic analysis of these LDs. Unexpectedly, about 19.3% of the plaque LD proteins belonged to cytoskeleton, while the content of lipid metabolism enzymes on plaque LDs was low. Our results show that the compositions of lipids and proteins were similar between two population LDs. Compared to soft (lipid) plaque LDs, LDs from mixed and hard (calcified) plaques had fewer metabolic enzymes and more cytoskeletal proteins. Lipidomics of the LD core uncovered an unexpected enrichment of polyunsaturated fatty acid cholesteryl esters (PUFA-CEs), which constituted 88.3% of all cholesterol esters. Notably, esters of the pro-inflammatory precursor arachidonic acid (AA) were among the most abundant species (16.5%). In addition, plaque LD were uniquely enriched in phospholipids including phosphatidic acid (PA), phosphatidylethanolamine plasmagens (PE-P), sphingomyelin (SM), and phosphatidylserine (PS). Our findings redefine the plaque LD as a dual-function "bad" organelle: a metabolically inert container for cholesterol ester, coated with cytoskeletons, and a massive reservoir for inflammatory precursors.

cell biology↗

bioRxiv2025

Shared latent representations of speech production for cross-patient speech decoding

Speech brain-computer interfaces (BCIs) can restore communication in individuals with neuromotor disorders who are unable to speak. However, current speech BCIs limit patient usability and successful deployment by requiring large volumes of patient-specific data collected over long periods of time. A promising solution to facilitate usability and accelerate their successful deployment is to combine data from multiple patients. This has proven difficult, however, due to differences in user neuroanatomy, varied placement of electrode arrays, and sparse sampling of targeted anatomy. Here, by aligning patient-specific neural data to a shared latent space, we show that speech BCIs can be trained on data combined across patients. Using canonical correlation analysis and high-density micro-electrocorticography (ECoG), we uncovered shared neural latent dynamics with preserved micro-scale speech information. This approach enabled cross-patient decoding models to achieve improved performance relative to patient-specific models facilitated by the high resolution and broad coverage of ECoG. Our findings support future speech BCIs that are more accurate and rapidly deployable, ultimately improving the quality of life for people with impaired communication from neuromotor disorders.

neuroscience↗

bioRxiv2025

Metabolomics-Guided Machine Learning Reveals Diagnostic and Mechanistic Biomarkers in CHB with MASLD

BackgroundMetabolic dysfunction-associated steatotic liver disease (MASLD) often coexists with chronic hepatitis B (CHB), yet early diagnosis remains challenging, particularly in non-obese patients or those with subclinical features. This study aimed to identify metabolic signatures of CHB-related MASLD and construct a predictive model using untargeted metabolomics integrated with machine learning. MethodsSerum metabolomics was performed on 160 subjects (80 CHB + MASLD and 80 healthy controls). Differential metabolites were identified and analyzed using KEGG enrichment and 4 machine learning algorithms (Random Forest, XGBoost, SVM, and LASSO). Metabolite-clinical correlations and diagnostic model performance were evaluated. ResultsA total of 924 differential metabolites were identified, with significant enrichment in pathways related to the TCA cycle, sphingolipid metabolism, and amino acid turnover. Machine learning prioritized six robust and biologically relevant metabolites: L-aspartic acid, succinic acid, caproic acid, sebacic acid, monomenthyl succinate, and glycolaldehyde, which consistently distinguished CHB + MASLD patients from controls (AUC > 0.75). These metabolites reflect key disruptions in mitochondrial function, lipid oxidation, and redox homeostasis. Integrated models combining metabolomics with clinical indices achieved perfect classification (AUC = 1.000). ConclusionCHB-associated MASLD is driven by systemic metabolic remodeling centered on mitochondrial overload, oxidative stress, and impaired amino acid metabolism. The identified metabolites provide mechanistic insights and hold promise for non-invasive MASLD screening in CHB patients. This study underscores the potential of multi-algorithmic metabolomics in advancing early diagnosis and personalized management of complex liver comorbidities.

bioinformatics↗