Biology subjects

Zhang, Z.

Publications and source records attributed to Zhang, Z..

At least 325 records · Page 18Linked to original sources

bioRxiv2025

Deep learning reveals a microprotein atlas in maize and uncovers novel regulators of seed amino acid metabolism

Microproteins represent a vast and functionally important class of genes that remain largely unexplored in plant genomes. Here, we developed DeepMp, a deep learning framework that integrated multi-omics evidence and built the most comprehensive plant microprotein atlas to date, identifying 18,338 high-confidence candidates in maize. The majority of these appear to have originated de novo from regions previously annotated as noncoding, and they show hallmarks of rapid, lineage-specific evolution and pronounced tissue specificity. These novel microproteins were found integrated into core regulatory networks, particularly in organ development and nutrient storage. Focusing on the maize kernel, population-scale analyses linked microprotein expression to natural variation in amino acid content. We functionally validated three grain-filling-specific candidates originating from noncoding regions by CRISPR-Cas9 knockouts, which confirmed their roles as precise modulators of arginine, aspartate, and methionine levels, without pleiotropic effects on kernel morphology. Our study provides a foundational resource and analytical framework, establishes microproteins as a new and functionally important coding layer in maize, and uncovers a previously untapped source of targets for crop improvement.

genomics↗

bioRxiv2025

Modular mRNA lipid nanoparticle platform rescues diverse genetic male infertility

Genetic male infertility arises from diverse mutations that disrupt spermatogenesis, yet therapeutic strategies capable of restoring germ-cell function remain limited. Messenger RNA (mRNA) therapeutics offer a programmable and non-integrating approach, but their functional compatibility in structured and immune-privileged tissues remains unclear. Here we identify a temporal mismatch between mRNA delivery kinetics and the progression of spermatogenesis as a key barrier to functional rescue. We show that delivery efficiency does not predict outcome: formulations with high peak expression but short duration fail to support spermiogenesis, whereas sustained expression aligned with developmental timing enables recovery. Guided by this principle, we establish a testis-targeted mRNA delivery platform that restores spermatogenesis across multiple genetic models and generates functional sperm capable of supporting embryo development and multigenerational inheritance. The approach is modular, extends to human seminiferous tubules ex vivo, and is supported by comprehensive safety analyses. Together, these findings define a design constraint for RNA therapeutics in structured tissues: functional rescue requires temporal alignment between delivery kinetics and the therapeutic window defined by tissue-specific biological programs.

bioengineering↗

bioRxiv2025

Integrating Millions of Years of Evolutionary Information into Protein Structure Models for Function Prediction

BackgroundUnderstanding life processes relies on accurate protein function prediction, fundamentally requiring the integration of evolutionary information encoded in sequences with spatial characteristics from 3D structures. Existing approaches often face limitations, however, by either over-relying on sequence, using simplified structural representations instead of fine-grained spatial details, or failing to capture the synergistic relationship between sequence and structure, compounded by challenges in acquiring annotated data. ResultsTo address these issues, we propose a novel contrast-aware pre-training framework, ESMSCOP. ESMSCOP leverages a state-of-the-art protein language model to harness evolutionary insights embedded in sequences, and introduces a new encoder to fuse topological and fine-grained spatial structural features. By employing a contrastive pre-training strategy with auxiliary supervision, ESMSCOP effectively bridges the sequence-structure gap, yielding rich and informative representations. ConclusionsExtensive experiments conducted on multiple benchmark datasets demonstrate that ESMSCOP achieves superior performance in protein function prediction tasks compared to existing methods. Furthermore, it shows strong performance even when utilizing relatively less pre-training data compared to some large-scale models.

bioinformatics↗

bioRxiv2025

Circulating chromatin reveals the effects of disease-associated variants on gene regulation

A fundamental challenge in human genetics is determining how variation in regulatory DNA shapes complex traits and disease risk. Chromatin quantitative trait loci (cQTLs) can address this challenge by revealing the effects of disease-linked genetic variants on regulatory element activity. Discovering cQTLs in disease-relevant tissues at scale remains challenging, however. To address this limitation, we leveraged advances in epigenomic liquid biopsy. We profiled histone modifications in circulating chromatin from patients with cancer to identify cell-free chromatin QTLs (cfcQTLs). By sampling cancer-derived chromatin in plasma, we captured cfcQTLs affecting regulatory elements from diverse non-hematologic tissues, as well as developmentally restricted elements that are reactivated in cancer (enriched 16-fold). Applying a cistrome-wide association study (CWAS), we linked 4,891 cfcQTLs to 1,011 traits and diseases. Developmentally restricted cfcQTLs that were not found in white blood cells were associated with 22.7 traits per 100 QTLs, compared to 0.58 for WBC-restricted cQTLs, underscoring the power of cfcQTLs for capturing genetic variation that shapes phenotypes. We extended our approach beyond germline variants to non-coding somatic mutations in cancer by measuring the activating effects of TERT promoter mutations from circulating chromatin. This work provides a path to scalable discovery of cQTLs across tissues, diseases, and populations to dissect the genetics of common diseases, enabled by the ease of sampling blood.

genomics↗

bioRxiv2025

Dissecting mammalian cortical circuit development at single-cell resolution using inducible barcoded rabies virus

Highly organized circuits of connected neurons enable diverse brain functions. Improper development of these circuits is associated with neurodevelopmental disorders, and understanding how circuits are formed is crucial for unraveling the mechanisms of these diseases. We currently have an incomplete picture of how specific brain circuits develop and how they are affected in disease, because we lack methods to study them at scale and with single-cell resolution. Monosynaptic rabies tracing is the gold standard method to study circuit architecture. However, it suffers from cellular toxicity, low throughput, lack of control over the timing of labeling, and the inability to access the molecular profiles of individual neurons. To address these issues, we developed an inducible barcoded rabies virus (ibRV) to enable temporal-controlled labeling of synaptic circuits followed by high-throughput single-cell genomics readout. ibRV allows for dissecting neuronal circuit changes over time at single-cell and spatial resolution. We applied ibRV to study the development of specific mouse cortical circuits during late prenatal and postnatal life using single-cell genomics and unbiased spatial transcriptomics as readouts. We characterized and quantified developmental connectivity patterns and molecular cascades that underlie their formation. Additionally, we constructed functional in silico circuit models that enable interrogation of circuit function and dysfunction at specific developmental stages. Our study provides novel tools for circuit analysis and can provide new insights into the mechanisms of mammalian brain development.

neuroscience↗

bioRxiv2025

Flexible parental care in a songbird correlates with sex-specific responses to seasonal phenology, mating opportunity and reproductive success

O_LISpecies with diverse parental care patterns often exhibit flexibility in response to environmental changes. Such changes can influence parental care decisions by altering the trade-offs between current and future reproduction. While previous studies have revealed correlations between specific social and ecological factors and parental care patterns, the joint effects of multiple factors could be rather complex and may result in different outcomes within and across populations, and between sexes. C_LIO_LIIn this study, we revealed parental care systems and their seasonal variations across three geographic populations in a polygamous species, the Chinese penduline tits (Remiz consobrinus), by monitoring breeding events over years. To investigate how local conditions (including social and ecological conditions) correlate with sex-specific parental care patterns, we compared breeding phenology and local conditions (including season length, mating opportunity and reproductive success) across the populations. C_LIO_LIStriking population differences in parental care strategies are found in this species: biparental care predominated in one population, whereas female-only care was more common in the other two. Our analyses reveal that, despite a large difference in season length, the time of egg-laying peaked similarly across the three populations. Female penduline tits determined the breeding phenology, which varies male mating opportunities dramatically over the season. For all populations, male Chinese penduline tits were more likely to desert the clutch when male mating opportunities were high. In contrast, females consistently provided care, regardless of variation in female mating opportunities. C_LIO_LIWe also found that fitness costs on the current brood resulting from offspring desertion differed across populations. This desertion cost on brood fitness matters more than achieving a high reproductive reward in driving the emergence of uniparental care, contrary to the theory that uniparental care only occurs when single parents can efficiently raise nestlings alone. C_LIO_LIOur study demonstrates that similar parental care patterns across populations could result from different underlying factor interactions and highlights the consistent sex-specific responses to local conditions in parental care. C_LI

ecology↗

bioRxiv2025

Wearable sweat nicotine biosensor based on a nicotineoxidoreductase and its natural electron acceptor

The advent of wearable biosensors is empowering clinical and lifestyle decision making. Nicotine is a drug of significant negative impact on public health given the prevalence of smoking and vaping. Yet noninvasive, continuous monitoring of nicotine is challenging due to lack of specific and sensitive biosensing elements. Here, we report a wearable highly sensitive nicotine biosensor and demonstrate on-body deployment of the sensor in a first-in-human study. The biosensor comprises nicotine oxidoreductase (NicA2) with its natural cytochrome c electron acceptor (CycN) as mediator: both proteins play a key role in the nicotine-degrading bacterium, Pseudomonas putida S16. The biosensor detects nicotine over four-orders of magnitude (0.1-100 M) with nanomolar sensitivity (LOD [~]33.6 nM), performs in the physiological pH range (pH 6-9), and is highly selective against common interferants and the major human nicotine metabolite, cotinine. Incorporation of the biosensor in a custom wearable device enables real-time measurement of nicotine from locally induced sweat on volunteer subjects and demonstrates higher analytical accuracy than gold-standard mass spectrometry.

bioengineering↗

bioRxiv2025

Bronsted-basic small molecules activate GTP hydrolysis in Ras Q61 mutants

The RAS oncogenes (KRAS, HRAS, NRAS) are among the most frequently mutated genes in human cancer, affecting over three million patients annually. Therapeutic development has largely focused on inhibitors for KRAS codon 12 mutations (G12C/D/V/S/R) which are key drivers in lung, colorectal, and pancreatic cancers. In contrast, mutant-selective inhibitors for Q61 variants remain elusive. A common mechanistic feature of G12 and Q61 mutants is the impaired hydrolysis of GTP, which traps Ras in its active, signaling-competent state. We envisioned that an alternative therapeutic strategy - reactivation of GTP hydrolysis - could address this shared oncogenic mechanism. Here we report small molecules that accelerate GTP hydrolysis in K-Ras Q61 mutants. These compounds compensate for the loss of the catalytic residue Gln61 by introducing a general base into the active site, selectively enhancing hydrolysis of K-Ras Q61X (X = H, L, K, R) mutants by up to 20-fold while sparing the wild-type protein. In mutant cancer cell lines, these compounds reduce GTP-bound Ras levels and suppress downstream signaling. We show that the chemical design principles are generalizable to other Ras isoforms. This work establishes a mechanistic foundation for small-molecule "GTPase activators" and offers a new paradigm for targeting Ras-driven cancers.

cancer biology↗

bioRxiv2025

Circuit-specific gene editing for precision modulation of neuronal activity with CRISPR-rabies virus

CRISPR gene editing has revolutionized our ability to study and manipulate specific genes, enabling novel insights into gene function and potential therapies for brain disorders. Recent advances in cell-type-specific regulatory elements and viral delivery systems have made precise in vivo gene editing possible. However, neurons with similar molecular profiles can belong to different circuits, complicating efforts to manipulate circuit function and behavior. To address this, we developed CRISPR-rabies virus (CRV), which leverages the trans-synaptic spread of rabies virus to enable gene editing within anatomically defined neural circuits. By pairing CRV with cell type-specific Cas9 expression, we achieved targeted gene modifications in specific circuits. We demonstrate that CRV can modulate sodium and potassium channel expression in parvalbumin interneurons, thereby effectively regulating synaptic transmission of pyramidal neurons in the CA3 region of the hippocampus. Its compatibility with 3'-capture single-cell RNA-seq allows simultaneous circuit perturbation and molecular profiling. In summary, CRV allows precise circuit-level gene modulation, providing a platform for studying gene function in neural circuits and developing novel gene therapies for brain disorders.

neuroscience↗

bioRxiv2025

Models for the architecture of the human inner kinetochore CCAN complex on centromeric α-satellite CENP-A nucleosome arrays

Human kinetochores assemble onto centromeric DNA defined by tandem copies of thousands of the 171 bp -satellite repeat sequence. The centromere-specific CENP-A nucleosome (CENP-ANuc) recruits the inner kinetochore constitutive centromere associated network (CCAN) complex. A previous cryo-EM structure of CCAN bound to a CENP-ANuc reconstituted with a 171 bp -satellite repeat sequence showed how extranucleosomal linker DNA threads through an internal tunnel in the CCAN complex. To understand the higher-order architecture of the inner kinetochore assembled onto -satellite repeat arrays, we have determined cryo-EM structures of CCAN with longer DNA sequences. These include free DNA and single and dimeric -satellite repeats with CENP-A nucleosomes. We show from the structures of CCAN bound to both free DNA and monomeric CENP-ANuc that CCAN engages 65-70 bp of DNA comprising 30-35 bp of an upstream -satellite repeat. This upstream DNA interacts with the histone-fold domain subunits of the CENP-TWSX module in a manner resembling how DNA is wrapped in nucleosomes. A complex of CCAN assembled onto a dimeric -satellite repeat with two CENP-A nucleosomes showed that CCAN can only be accommodated on the linker DNA by unwrapping DNA from the CENP-TWSX module together with 25 bp of DNA from the upstream nucleosome. We discuss the implications of these results for models of CCAN assembly on arrays of -satellite chromatin containing CENP-ANuc.

molecular biology↗

bioRxiv2025

Fine-tuned Protein Language Model Identifies Antigen-specific B Cell Receptors from Immune Repertoires

Scalable identification of antigen-specific antibodies from whole immune repertoire V(D)J sequences is a central challenge in biomedical engineering. We show that protein language models (PLMs) fine-tuned on antibody heavy-chain sequences can directly predict antigen specificity from unselected immune repertoires. We assessed our model, Antigen Specificity Predictor (ASPred), against SARS-CoV-2, influenza, and HIV-AIDS antigens, observing comparable predictive performance. In the whole immune repertoire V(D)J sequences of mice immunized with the SARS-CoV-2 spike proteins receptor-binding domain (RBD), ASPred identified antibody sequences specific to RBD. Several candidate sequences were validated, including one as a heavy chain-only nanobody with 20.7 nM dissociation constant. Molecular dynamics simulations supported the predicted interactions at coarse-grained and atomic levels. Benchmarking against Barcode-Enabled Antigen Mapping (BEAM) of B cell receptor sequence data had highly significant overlaps with ASPred predictions, suggesting scalability. The predicted SARS-CoV-2 binders differed substantially from training sequences, demonstrating generalization beyond sequence memorization. Together, we establish that heavy chain antibody sequences encode sufficient information for PLMs to infer specificity, offering a scalable framework for antibody discovery with broad applications.

bioinformatics↗

bioRxiv2025

Characterization of Cancer-Reactive T Cells and Neoantigen-specific T Cell Receptors

Not all tumor-infiltrating T cells are cancer-reactive (CR). Several studies have reported gene expression signatures associated with CR-T cells. To integrate these findings, we developed a computational workflow, CAT (Cancer-Associated T cells), which harmonizes existing CR-T cell signatures and applies them to identify CR-T cells across an atlas of one million T cells. Our findings reveal that the abundance of CR-T cells varies across cancer types and that baseline levels of CR-T cells predict patients responses to immunotherapy. In parallel, we established a high-throughput computational platform, Neo-TCR, for systematic screening of neoantigen-specific TCRs and their cognate neoantigens. The efficacy of Neo-TCR is validated by cross-validation studies and replications in two independent datasets. Together, our findings suggest a new direction for developing biomarkers for cancer detection and monitoring: integrating CR gene expression signatures with neoantigen-specific TCRs.

cancer biology↗

bioRxiv2025

Consensus Pituitary Atlas, a scalable resource for annotation, novel marker discovery and analyses in pituitary gland research

Previous single-cell profiling studies of the pituitary gland have yielded minimally reproducible insights largely due to their low statistical power and methodological inconsistencies. To address this problem, we generated a uniformly pre-processed Consensus Pituitary Atlas (CPA) using all existing mouse pituitary single-cell datasets (267 biological replicates, >1.1 million high-quality cells). The CPA revealed novel cell typing and lineage markers, including low-expression transcripts that previous analyses could not detect. The scale of the CPA enabled the development of machine learning models to automate and standardize cell type annotation and doublet identification for future studies. Leveraging the curated metadata, we identified sex-biased and age-dependent gene expression patterns at cell type resolution. To identify drivers of cell fates, first we determined consensus cell communication patterns. Secondly, we used RNA-sequencing and chromatin accessibility data to identify transcription factors associated with cell fates across modalities. The epitome platform acts as an interface with the CPA, allowing streamlined user-friendly analyses. HighlightsO_LIUniform processing of 267 mouse pituitary single-cell datasets (>1.1M cells) C_LIO_LIThe statistical power enabled cell type, sex- and age-specific marker discovery C_LIO_LIMachine learning models facilitate doublet detection and cell typing in new datasets C_LIO_LIepitome platform provides programming-free data access and visualizations C_LI

bioinformatics↗

bioRxiv2025

LRP8 is a Receptor for Yellow Fever Virus

Yellow fever virus (YFV), an arbovirus causing substantial human morbidity and mortality, was the first human virus discovered over a century ago. The live-attenuated 17D vaccine is among the most successful vaccines in medicine. Despite the importance of YFV, its receptor has remained unknown. Here, we performed a barcoded, genome-wide human ORF library screen and identified LRP8 (also named APOER2) as a receptor for YFV. We show that LRP8 expression specifically boosts YFV infection in cell lines by promoting entry. AAV-mediated expression of human LRP8 in mouse liver aggravates infection and pathology. LRP8 knockdown abolishes YFV infection in brain cells, primary human hepatocytes, and notably in mosquitoes. Biochemically, LRP8 directly interacts with YFV particles via the viral envelope protein. This function of LRP8 is conserved across species, particularly in mosquitoes and primates. A soluble LRP8 decoy protein can block YFV infection in vitro and in mice, providing a potential therapeutic or prophylactic strategy. Our findings provide groundwork for understanding YFV entry, tropism, and pathogenesis, and may enable development of novel therapeutics to treat YFV infection.

microbiology↗

bioRxiv2025

PIWIL3-piRNA Pathway Is Essential for Rabbit Oogenesis and Embryogenesis via Broad Regulation of the Transcriptome and Proteome

Female infertility represents a significant challenge in reproductive medicine. Although it is known to be primarily due to oogenic and early embryonic failures, the molecular mechanisms underlying these failures remain elusive. The Piwi-piRNA pathway is crucial for gametogenesis in diverse organisms. Yet, its role in mammalian female fertility is unclear. This is partly because most studies are done in mice, which contain only PIWIL1, PIWIL2, and PIWIL4 that are essential for spermatogenesis but not for oogenesis. PIWIL3 emerges in higher mammals and is highly expressed in human oocytes. Interestingly, female PIWIL3-knockout golden hamsters exhibit only reduced fertility without detectable defects in oogenesis. This has left the function of PIWIL3 in higher mammals, including humans, unexplored. Here, we discovered that PIWIL3 in the rabbit (Oryctolagus cuniculus) shares high homology with human PIWIL3. It is the predominant PIWI protein in oocytes in both species. Using CRISPR-cas9-mediated knockout, we demonstrated that rabbit PIWIL3 is essential for female fertility. Its loss leads to severe defects in oogenesis. Moreover, embryos lacking maternal PIWIL3 arrest developmentally at the 8-cell stage. Mechanistically, rabbit PIWIL3 binds [~]18-nucleotide piRNAs, mirroring the behavior of human PIWIL3, and is critical for piRNA biogenesis. Moreover, it regulates transcriptomic and proteomic landscapes and silences a broad array of transposons during late oogenesis yet activates another set of transposons during early embryogenesis. These findings establish PIWIL3 as a pivotal dual regulator of gene expression and transposon activity, which is essential for oogenesis and early embryogenesis in non-rodent mammals, potentially including humans.

developmental biology↗

bioRxiv2025

Lymphatics unload VEGFR3 in response to the endothelial stress upon VEGFR2 insufficiency

Vascular endothelial growth factor receptor 2 (VEGFR2) is a key target for regulating the endothelial cell lineage and angiogenesis. It is also expressed by lymphatic endothelial cells (LEC) while its participation in the process of lymphangiogenesis remains inadequately characterized. We show in this study that VEGFR2 is highly expressed in dermal initial lymphatic vessels and valves. The induced deletion of pan-endothelial Vegfr2 at the neonatal stage produced a potent suppression of dermal lymphatic growth, including the thinner lymphatic diameter, the decrease of LEC number and lymphatic valves. Mechanistically, the VEGFR2 insufficiency led to a dramatic decrease of lymphatic VEGFR3, a key regulator mediating signals for lymphatic growth and remodeling. The RNA sequencing analysis revealed that GO terms enriched for the downregulated genes included biological processes related to the EC development while pathways related to the hematopoiesis and immune responses were upregulated in the skin of Vegfr2 mutants compared with the littermate controls. This was further confirmed by the RNA-seq analysis of dermal tissues 48 hours after the endothelial Vegfr2 deletion. Consistently, an inhibitory effect on the dermal lymphatic growth was also observed by targeting Vegfr2 in PROX1+ cells, manifesting a similarly altered transcriptome signature. The alteration of lymphatic gene expression was further validated by the siRNA mediated knockdown of Vegfr2 in primary lymphatic endothelial cells, showing a transcriptional trend of LEC to hematopoietic transition. Findings from this study imply that VEGFR2 is required for the maintenance of endothelial identity and its insufficiency may trigger the alteration of lymphatic transcriptional programs to diminish the VEGFR3-mediated lymphangiogenesis.

developmental biology↗

bioRxiv2025

DNA methylome responses to biotic and abiotic stress in Arabidopsis thaliana: A multi-study analysis

Plants have evolved complex physiological and morphological strategies to cope with biotic and abiotic stress. At the molecular level, these responses are partly mediated by epigenetic mechanisms such as DNA methylation. In Arabidopsis thaliana, numerous studies have examined stress-induced changes in DNA methylation, aiming to elucidate their roles in stress acclimation. However, comparing these studies remains difficult due to variations in analytical methods and focus. To address this, we conducted an integrative analysis of 13 studies (16 datasets) using a unified bioinformatic pipeline to identify common and stress-specific methylation responses. Our results show that the number and distribution of differentially methylated regions (DMRs) vary by stress type and methylation context CG, CHG, and CHH. CG DMRs were primarily located in gene bodies, while CHG and CHH DMRs were enriched in transposable elements (TEs). Gene Ontology analysis revealed consistently enriched stress-related terms across conditions, indicating shared epigenetic regulation. As TEs have been proposed to act as stress sensors and regulators of nearby gene expression, we examined gene-proximal TEs in more detail. We found significant enrichment of stress-responsive methylation changes in LINE/L1, RathE1_ons, and DNA/HAT elements. While stress-induced methylation is often considered transient, many affected CG DMRs overlapped with loci known to accumulate stable epimutations over generations, suggesting that stress may contribute to long-term heritable epigenetic variation. These findings enhance our understanding of epigenetic regulation in plant stress responses, highlighting both shared and stress-specific methylation dynamics. Significance statementWe combined 16 Arabidopsis thaliana Whole-Genome Bisulfite Sequencing datasets to reveal common and stress-specific DNA methylation responses. Although global methylation levels remained stable, specific DMRs varied according to the type of stress and genomics context. Gene Ontology analysis revealed shared stress pathways. Gene-proximal transposable elements (TEs) (e.g. LINE/L1, RathE1_ons, DNA/HAT) were enriched for stress-responsive methylation changes. Notably, a subset of stress-induced CG DMRs overlap with epimutation-prone loci, suggesting an environmental route toward stable, heritable epigenetic variation.

plant biology↗

bioRxiv2025

LabOS: The AI Co-Scientist That Sees and Works With Humans

Modern science advances fastest when thought meets action. LabOS represents the first AI co-scientist that unites computational reasoning with physical experimentation through multimodal perception, self-evolving agents, and XR-enabled, embodied human-AI collaboration. To empower agentic AI with embodied intelligence, we built LabSuperVision, the first vision-language-model (VLM) benchmark spanning biomedical and materials science laboratories, designed to evaluate scientific perception and reasoning. This endowed LabOS with visual reasoning capabilities that surpass those of general-purpose models on laboratory-specific tasks. By connecting multi-model AI agents, smart glasses, and human-AI-robot collaboration, LabOS allows AI to see what scientists see, understand experimental context, and participate in real-time execution through immersive XR interaction and robotic autonomation. Across applications-from cancer immunotherapy target discovery to cell engineering-LabOS shows that AI can move beyond computational design to participation, turning the laboratory into an intelligent, collaborative environment where human and machine discovery evolve together.

bioengineering↗