Biology subjects

Yu, H.

Publications and source records attributed to Yu, H..

At least 109 records · Page 6Linked to original sources

bioRxiv2025

Discrete cytokine signaling networks instruct distinctsynovial pathotypes in inflammatory arthritis

Patients with rheumatoid arthritis (RA) display distinct patterns of synovitis. To define the inflammatory mechanisms driving this heterogeneity, we analyzed the inflamed synovium of wild-type (WT), Il6ra-/-, and Il27ra-/- mice with antigen-induced arthritis (AIA). Remarkably, each strain developed a joint pathology mirroring a major RA synovial pathotype: myeloid-rich (WT), fibroblast-rich/pauci-immune (Il6ra-/-), and lymphoid-rich (Il27ra-/-) synovitis. Histology confirmed minimal immune infiltration in Il6ra-/- joints, while WT and Il27ra-/- mice exhibited prominent immune involvement, including organized synovial lymphoid-like aggregates in Il27ra-/- mice. Transcriptomic and epigenomic profiling revealed both shared and distinct regulatory programs among genotypes. Il6ra-/- mice showed increased WNT, DKK, and AMPK signaling associated with fibroblast, chondrocyte, and osteoclast activation (e.g., Adamts19, Dkk1, Ecm1). Consistent with synovial ectopic lymphoid-like structures, Il27ra-/- mice showed enrichment of lymphocyte activation (e.g., Il17a, Il22, Bhlhe40). WT mice exhibited hallmarks of MAP kinase activation. These molecular signatures parallel those of fibroblast-, lymphoid-, and myeloid-rich synovitis in RA. Defining a STAT1-STAT3 regulatory interplay influencing transcriptional decisions in WT and Il27ra-/- mice, our findings offer insights into cytokine-driven disease heterogeneity. Together, these results establish a framework for mechanism-based classification of synovitis and introduce new mouse models to study the molecular drivers of synovial pathotypes and treatment response.

pathology↗

bioRxiv2025

Cross-species connectome comparisons reveal the network attributes of memory capacity and time series prediction

The brains connectome provides a powerful blueprint for designing efficient neural networks, yet the impact of incorporating its intricate, non-random architecture into machine learning models remains underexplored. Here, we integrate empirical structural connectomes from four model species--fruit fly, mouse, rat, and macaque--as the recurrent layer in echo state networks (ESNs). We demonstrate that biologically realized networks, particularly the macaque connectome, achieve superior performance in chaotic time series prediction and exhibit higher memory capacity compared to randomly shuffled controls. This computational advantage correlates with small-world topology, which scales with phylogenetic level. Crucially, we identify that weakly connected but highly central nodes are essential for optimal network dynamics; their targeted perturbation significantly degrades performance. Furthermore, functional connectomes from Alzheimers disease patients show computational deficits resembling those induced by weak-tie disruption in healthy networks. Our findings establish that evolved connectome topology is fundamental to efficient information processing, providing key principles for bio-inspired artificial intelligence.

neuroscience↗

bioRxiv2025

PlantScience.ai: An LLM-Powered Virtual Scientist for Plant Science

The accelerating growth of plant science literature presents a major challenge for researchers seeking to extract accurate, up-to-date knowledge from an increasingly fragmented and domain-specific corpus. General-purpose large language models (LLMs), while powerful, often misinterpret plant science terminology and lack mechanisms for source traceability. We created PlantScience.ai, a virtual plant biology scientist powered by our automated scientific knowledge graph construction pipeline (AutoSKG). PlantScience.ai exhibits expert-level reasoning in plant biology and maintains scholarly rigour in its citations. Through continuous learning, it integrates the latest research, ensuring that its knowledge base remains current and scientifically robust. Apart from providing the answers to the scientific questions, PlantScience.ai can interact with human scientists, follow instructions, and retrieve information with citation awareness, grounding each response in primary sources to ensure accuracy and verifiability. PlantScience.ai marks a pivotal advance toward a collaborative scientific paradigm in which virtual and human plant scientists work synergistically to accelerate discovery while preserving the unique value of human insight. PlantScience.ai is available at https://plantscience.ai.

plant biology↗

bioRxiv2025

Engineering Linear Cap-independent mRNA Vaccines with Intrinsic Adjuvanticity for Potent Cancer Immunotherapy

mRNA cancer vaccines have shown promising efficacy in early-phase clinical trials, but existing platforms struggle to boost antitumor efficacy without added cost or complexity. Here, we present a streamlined linear cap-independent RNA (LciRNA) cancer vaccine platform that achieves stable expression in vivo without 5 capping or ribonucleotide modification, and innately stimulates immune responses to enhance antitumor immunity. By fusing a UPA protective sequence, composed of a viral exoribonuclease-resistant RNA (xrRNA) and a poly(A) binding protein (PABP) motif, with an optimized Enterovirus A internal ribosome entry site, LciRNA resists 5 exonuclease decay and drives superior in vivo expression. Mechanistically, UPA sequence not only impedes exonuclease-mediated decay but also recruits RNA-binding proteins to stabilize LciRNA. Moreover, LciRNA robustly activates dendritic cell pattern-recognition receptor pathways, promotes dendritic cell maturation, and upregulates proinflammatory signals. In murine melanoma and HPV-associated tumor models, it elicits strong systemic and intra-tumoral T cell responses and superior tumor control, demonstrating how immune stimulation-translation synergy underpins its efficacy. This work establishes a next-generation cost-effective mRNA cancer vaccine platform with simplified production and enhanced efficacy, highlighting immuno-translation coupling as a paradigm for future mRNA cancer vaccines.

bioengineering↗

bioRxiv2025

METAHIT enables comprehensive and flexible genome-resolved microbiome analysis with metagenomic Hi-C

Metagenomic Hi-C (metaHi-C) adds within-cell proximity information to shotgun sequencing, but existing analyses are fragmented and short-insert intra-contig read pairs from Hi-C libraries are often underused. Here, we present METAHICT, a modular and configurable workflow spanning read processing, metagenome-assembled genome (MAG) reconstruction, and downstream analysis. METAHICT consolidates outputs from bin3C, MetaCC, and ImputeCC and uses an expectation-maximization model of mapped insert sizes to select short-insert, shotgun-like Hi-C pairs for per-bin reassembly. Across seven datasets, binning and consolidation generally recovered more near-complete and medium-quality MAGs than individual Hi-C binners. On the sheep-gut long-read assembly, METAHICT's binning and consolidation stage recovered 929 MAGs meeting at least the medium-quality threshold, including 487 near-complete MAGs. Within the reassembly module, adding EM-selected Hi-C reads to shotgun-only inputs yielded further significant reductions in contamination across multiple environments. The workflow also integrates scaffolding, taxonomic annotation, contact-map visualization, and analysis of candidate mobile genetic element (MGE)-host pairs. Together, these capabilities provide standardized outputs for comparative and hypothesis-driven genome-resolved microbiome analysis.

bioinformatics↗

bioRxiv2025

Identification of human RNA Polymerase II interactors at early stages of transcription.

RNA Polymerase II (Pol II) transcription is highly regulated at two early steps in the transcription cycle: Pre-Initiation Complex (PIC) assembly with its coupled initiation, and promoter-proximal pausing with its controlled release. Here, we developed an optimized biochemical purification method that captures endogenously tagged chromatin-bound Pol II complexes under native conditions at these rate-limiting steps. We then identified a large set of Pol II interactors by mass spectrometry and determined the footprints of these assemblies on promoters with high resolution. Many well-known and new or understudied factors were identified as associated with the PIC and promoter-proximal paused complexes, indicating that despite decades of efforts, these rate-limiting steps of the transcription cycle are far from being completely understood. The new and understudied factors implicate novel mechanisms of regulation that will need to be characterized to fully understand Pol II regulation.

molecular biology↗

bioRxiv2025

Neural signatures of harm aversion predict later willingness to exert effort for others rewards

Prosocial behaviours--actions that incur personal costs to benefit others--are central to human social life. Two key domains are moral harm aversion, where individuals forgo personal gains to prevent harming others, and prosocial effort, which involves exerting effort to benefit others. Although previous studies suggest a relationship between these behaviours, it remains unclear whether neural responses in one domain can predict prosocial motivation in another. Here, we tested whether neural sensitivity to morally salient information in harm aversion could predict prosocial effort later. Participants completed two tasks: a harm aversion task during fMRI, in which they traded off monetary profit against delivering electric shocks to another person; and, one week later, a prosocial effort task outside the scanner, in which they decided whether rewards for others were worth the required physical effort. We focused on three regions implicated in cost-benefit decision-making and social cognition: the anterior cingulate cortex (ACC), anterior insula (AI), and temporoparietal junction (TPJ). Behaviourally, greater harm aversion was associated with increased prosocial effort. Neurally, AI responses to others harm predicted sensitivity to others rewards in the effort task, consistent with a role in representing others outcomes across positive and negative valences. By contrast, TPJ responses to profit from harming others predicted decreased sensitivity to others rewards, suggesting a role in context-dependent valuation that may constrain prosocial behaviour. These findings demonstrate that neural responses to morally salient information in one context correlate with prosocial motivation in another, highlighting mechanisms that bridge moral sensitivity and effortful prosociality.

neuroscience↗

bioRxiv2025

Phosphatidylserine and RhoB connect phosphatidylinositol 4-phosphate and phosphatidic acid metabolism at the plasma membrane

Cells tightly control the homeostatic levels and subcellular localizations of membrane phospholipids through the regulation of the activities of numerous lipid-metabolizing enzymes and lipid transfer proteins. Yet, the mechanisms by which lipid imbalances are sensed and corrected to establish and maintain homeostasis are, in most cases, unknown. Here we present an expanded view of plasma membrane (PM) phosphoinositide metabolism by revealing an unexpected metabolic connection between two key anionic lipids in this membrane, phosphatidylinositol 4-phosphate (PI4P) and phosphatidic acid (PA). PM pools of PI4P are generated by PI 4-kinase Type III (PI4KIII/PI4KA), an essential enzyme whose partial dysfunction leads to numerous hereditary human diseases. We find that depletion of PI4P by pharmacological inhibition of PI4KA increases the activity of phospholipase Ds (PLDs) and the levels of their lipid product, PA, in the PM. Guided by RNA-seq analysis and proximity labeling proteomics, we elucidate how cells connect this PI4P decrease to a compensatory increase in PA levels. Loss of PM PI4P induces a concomitant decrease of phosphatidylserine (PS) levels, and this metabolic rewiring activates a reciprocal relationship between PS synthesis and PLD-mediated PA generation. These metabolic changes also lead to transcriptional and translational upregulation of the small GTPase RhoB, which enhances PLD-mediated PA synthesis and subsequent actin cytoskeletal remodeling. Our study reveals how disease-relevant perturbation of phosphoinositide synthesis induces an integrated response that ultimately boosts levels of PA, a key anionic lipid and metabolic intermediate in phosphoinositide resynthesis.

cell biology↗

bioRxiv2025

The regulatory effect of Ca2+/calmodulin dependent protein kinase II mediated by receptor interacting protein kinase 3 on necroptosis of hypertrophic cardiomyocytes

Receptor interacting protein kinase 3 (RIPK3) plays a crucial role in the signaling pathway of necroptosis and calcium/calmodulin dependent protein kinase II (CaMKII) is a novel substrate for RIPK3 induced necroptosis. The aim of this study is to investigate the regulation and mechanism of RIPK3 on AngII induced cardiomyocyte hypertrophy.AngII was used to stimulate myocardial cells for 72 hours, inducingcardiomyocytehypertrophy; Use RIPK3 inhibitor GSK872 to reduce RIPK3 expression.Detect indicators related to myocardial hypertrophy, cell damage, necroptosis,CaMKII activation and gene expression, oxidative stress, mitochondrial membrane potential, etc. After AngII stimulation of cardiomyocytes, the expression of hypertrophy markers ANP and BNP increased, LDH release increased, ATP levels decreased, splicing factors ASF and SC35 expression increased, CaMKII oxidation and phosphorylation levels increased, and CaMKII{delta} alternative splicing was disrupted. However, treatment with GSK872 can alleviate myocardial dysfunction, inhibit CaMKII activation, correct CaMKII{delta} variant splicing disorder and ultimately alleviating myocardial hypertrophy. In addition, pretreatment with RIPK3 can reduce the accumulation of reactive oxygen species (ROS) induced by AngII, decrease the activity of ASF and SC35, and restore mitochondrial membrane potential. RIPK3 inhibitor GSK872 can inhibit the activation of CaMKII, alleviate necroptosis and oxidative stress to alleviate myocardial hypertrophy. It has a protective effect on myocardial hypertrophy and is expected to become a new targeted drug for clinical treatment of dilated cardiomyopathy and heart failure.

cell biology↗

bioRxiv2025

The regulatory landscape of nascent transcription in human health and disease

Transcriptional regulatory elements (TREs) orchestrate gene expression programs fundamental to cellular identity and transitions across physiological and pathological states. Here, we present a high-resolution atlas of RNA Polymerase II-engaged TREs (enhancers and promoters) across all major human organ systems and a broad spectrum of developmental and disease states. This atlas is generated using PRO-cap, a highly sensitive method that detects nascent RNA at transcription initiation sites, a critical feature of active TREs. The base-pair resolution of PRO-cap enables systematic dissection of transcription initiation architecture, revealing associations among tissue specificity, evolutionary constraint, transcription factor usage, and regulatory connectivity. Integration with deep learning models such as ProCapNet further provides a framework for prioritizing noncoding variants from GWAS and eQTL studies. Moreover, this tissue-resolved atlas identifies lineage-specific regulatory programs and their alterations in diseases such as metastatic cancer, where TRE landscapes capture regulatory signatures reflecting both tissue of origin and adaptive responses to distant niches. Together, these findings establish transcription initiation at regulatory elements as a defining and mechanistically informative layer of gene regulation across development, physiology, and disease.

genomics↗

bioRxiv2025

SpaceSequest: A unified pipeline for spatial transcriptomics data analysis

BackgroundSpatial transcriptomics has emerged as one of the most powerful tools for gaining biological insights, enabling researchers to uncover intricate relationships between gene expression patterns and tissue architecture. Recent advances in the field have resulted in a variety of new platforms, including Visium, Visium HD, and Xenium from 10x Genomics, as well as GeoMx and CosMx from NanoString Technologies, which has now been acquired by the Bruker Corporation. However, the existence of diverse spatial transcriptomics platforms and various data formats poses challenges in standardizing data analysis. Thus, there remains a critical gap in the availability of a comprehensive pipeline capable of conducting end-to-end analysis that is necessary to extract biological insights from multiple spatial transcriptomics platforms. ResultsHere, we present SpaceSequest, a tailored pipeline that utilizes cutting-edge computational methodologies to conduct a thorough analysis, enabling the extraction of crucial biological insights from five major spatial transcriptomics technologies. SpaceSequest performs (1) standardized quality control and general data processing, (2) key analyses customized for each spatial platform, (3) automated cell type annotation and deconvolution, and (4) high-quality figure and analysis result generation. In addition, SpaceSequest allows for smooth integration with cellxgene VIP and Quickomics for user-friendly data access and interactive visualization. ConclusionsSpaceSequest is a unified and comprehensive pipeline designed for the analysis, visualization, and publication of spatial transcriptomics data from various platforms. The source code is available at https://github.com/interactivereport/SpaceSequest. To facilitate seamless installation and usage, we have also created a detailed Bookdown tutorial that can be accessed through https://interactivereport.github.io/SpaceSequest/tutorial/docs/index.html.

bioinformatics↗

bioRxiv2025

SpatialESD: spatial ensemble domain detection in spatial transcriptomics

Spatial transcriptomics (ST) measures gene expression while preserving spatial context within tissues. One of the key tasks in ST analysis is spatial domain detection, which remains challenging due to the complex structure of ST data and the varying performance of individual clustering methods. To address this, we propose SpatialESD, a Spatial EnSemble Domain detection method that integrates results from different spatial domain detection methods to improve spatial domain detection. SpatialESD captures both direct co-occurrence patterns and multiscale indirect relationships between clusters, improving the robustness and accuracy of spatial domain detection. We evaluated SpatialESD on simulated datasets and multiple 10x Visium spatial transcriptomics datasets, including human brain, breast cancer, and ovarian cancer samples. The results show that SpatialESD consistently outperforms individual methods and the existing EnSDD ensemble method in terms of clustering accuracy and stability. Based on the identified domains, we further detected region-specific differentially expressed genes and performed trajectory and cell-cell interaction analyses. These results reveal spatial patterns of gene expression and cellular communication, offering insights into tissue organization and disease mechanisms. Overall, SpatialESD provides a reliable and effective solution for spatial domain detection in ST data and facilitates downstream biological discovery. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=131 SRC="FIGDIR/small/675735v1_ufig1.gif" ALT="Figure 1"> View larger version (39K): org.highwire.dtl.DTLVardef@1344724org.highwire.dtl.DTLVardef@12ab936org.highwire.dtl.DTLVardef@1e59d57org.highwire.dtl.DTLVardef@1bcd55f_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2025

Disease-specific variant pathogenicity prediction using multimodal biomedical language models

Missense variants play a key role in the diagnosis of genetic disorders and in disease risk prediction. Existing methods focus primarily on the prediction of variant effects in terms of their deleteriousness, without taking into account the disease-specific context, and are therefore limited in terms of their utility in real-world diagnosis and decision making. Here, we introduce disease-specific variant pathogenicity prediction (DIVA), a novel deep learning framework that directly predicts specific disease types alongside the probability of deleteriousness for missense variants. Our approach integrates information from two different modalities - protein sequence and disease-related textual annotations - encoded using two pre-trained language models and optimized within a contrastive learning paradigm designed to align variants with relevant diseases in the learned representation space. Our results demonstrate that DIVA outperforms baselines and provides accurate disease predictions with high relevance to clinically curated disease annotations for missense variants. Variant deleteriousness prediction is enhanced by incorporating AlphaMissense scores through learnable weights derived from protein function annotations, which additionally boosts DIVAs ability to accurately classify deleterious variants. Our work provides new insights into variant pathogenicity prediction with awareness of disease specificity, addressing a hitherto unmet need in relation to clinical variant interpretation.

bioinformatics↗

bioRxiv2025

RNA splicing dynamics in CD8 T cells uncovers isoforms that impact T cell-mediated cancer immunotherapy

Immune checkpoint blockade has transformed cancer therapy, yet many patients fail to respond, and few new targets have emerged beyond PD-1 and CTLA-4. Alternative splicing dramatically diversifies the T cell proteome, but the functional roles of most isoforms remain unknown. Here we constructed the first single-cell splicing atlas of human CD8 T cells, capturing dynamic isoform programs across activation and subset differentiation. This revealed distinct splicing footprints that refine conventional transcriptomic states and highlight receptor families with isoform-level regulation. To functionally interrogate these candidates, we developed SpliceSeek, a CRISPR-based pooled screening platform that perturbs splice sites to redirect isoform usage. Using SpliceSeek, we uncovered isoform-specific immune checkpoints whose perturbation enhanced effector function and tumor control, including the LRRN3-203 isoform, which augmented cytokine secretion and antitumor immunity in mice models. Together, our results establish alternative splicing as a targetable layer of immune regulation and demonstrate the potential of isoform-focused screening to expand the landscape of cancer immunotherapy.

immunology↗

bioRxiv2025

Framework for analyzing MAE-derived immunopeptidomes from cell lines with shared HLA haplotypes

BACKGROUNDThe goal in vaccinology is to identify candidate antigens for clinical trials that will elicit an immune response for a significant portion of the target population. Unfortunately, promising data generated at the preclinical level often cannot be replicated in larger sample sizes. The goal of this project was to develop a methodology for processing MAE-generated data to identify MHC epitopes, minimize non-specific contaminants, find binding motifs, and utilize genetic connections among donors to determine which peptides were presented by specific MHC alleles. RESULTSOur approach demonstrated that mild acid elution of peptides from seven consanguineous B-lymphocyte lines accurately reflects the HLA genotypes within family members, highlighting the specificity of MAE. Additionally, the data successfully reproduced known MHC binding motifs and partially deconvoluted the originating HLA alleles of the epitopes. CONCLUSIONSThese findings suggest that our approach could be applied to numerous cell lines globally to evaluate a wide array of HLA haplotypes. This may help to reveal candidate vaccine antigens that induce immune protection for a wider population.

immunology↗

bioRxiv2025

Dysregulation of innate immunity and cellular metabolism through virus-induced deISGylation

Interferon-stimulated gene 15 (ISG15) regulates diverse cellular responses including antiviral immunity through its conjugation to proteins, a process known as ISGylation. Several pathogens, including SARS-CoV-2, subvert ISGylation by encoding deISGylating enzymes. However, the direct targets and physiological consequences of coronaviral deISGylation remain poorly defined. Here, we ablated the deISGylating activity of the SARS-CoV-2 papain-like protease (PLpro) and found that loss of deISGylation boosted innate immune activation, attenuated virus replication, and promoted viral clearance in human cells and in mice. Through untargeted metabolomics and ISGylome proteomics analyses, combined with functional studies, we discovered in molecular detail how the activities of key metabolic enzymes in glycolysis, the pentose phosphate pathway, and oxidative stress are controlled by PLpro deISGylation. These findings provide fundamental new insight into how reversible ISGylation regulates immunity and metabolic processes at the molecular level and highlight viral deISGylation as a major viral tactic for rewiring immunometabolism.

microbiology↗

bioRxiv2025

DNA barcoding for parallelised single-molecule characterisation of kinetic phenotypes

Single-molecule techniques provide exceptional resolution of biomolecular dynamics but are limited by low-throughput. We present a barcoding strategy that enables simultaneous kinetic profiling of multiple protein variants at the single-molecule level. Each protein is covalently linked to a unique DNA barcode, decoded via transient hybridisation of fluorescent probes distinguished by colour and binding duration. We applied this method to a library of 16 ClpS variants to systematically explore evolutionary trajectories towards a variant adapted to recognise N-terminal amino acids with binding kinetics optimised for single-molecule peptide sequencing. The approach uncovered [~]5-fold variation in median dissociation rates across variants. Several variants displayed bimodal kinetics, likely reflecting structural subpopulations, and differences in kinetic properties arose primarily from the relative abundance of these phenotypes. This approach offers a general framework for screening and optimising proteins for single-molecule applications, while revealing mechanistic insights that are inaccessible to traditional ensemble methods.

biophysics↗

bioRxiv2025

Evolutionary innovation within conserved gene regulatory networks underlying biomineralized skeletons in Bilateria

Biomineralized skeletons have evolved convergently across animals, yet exhibit remarkable diversity in structure and development. However, the evolutionary origins of gene regulatory networks underlying biomineralized skeletons remains elusive. Here, we report comprehensive developmental profiling of transcriptomic and chromatin dynamics in a bivalve mollusc, Crassostrea nippona. We provide evidence for a biphasic regulatory program orchestrating larval and adult shell formation, involving the coordinated activity of ancient transcription factors and dynamic chromatin remodeling. Comparative analyses suggest a conserved developmental toolkit was co-opted for larval exoskeleton formation in the common lophotrochozoan ancestor. In contrast, limited regulatory conservation was observed between lophotrochozoans and echinoderms with regards to the formation of biomineralized skeletons, despite both relying on a heterochronic activation of ancestral regulators. Together, our findings support a hierarchical model where dynamic chromatin decouples rapidly evolving effectors from deeply conserved regulators, allowing modular innovations within conserved gene regulatory networks. This study highlights how epigenetic dynamics bridge evolutionary conservation and novelty, offering a framework for understanding the independent evolution of biomineralization across Bilateria through combinatorial regulatory evolution.

evolutionary biology↗