Biology subjects

Chen, C.

Publications and source records attributed to Chen, C..

At least 199 records · Page 11Linked to original sources

bioRxiv2025

MuSHIN: A multi-way SMILES-based hypergraph inference network for metabolic model reconstruction

Genome-scale metabolic models (GEMs) are indispensable tools for probing cellular metabolism, enabling predictions of metabolic fluxes, guiding strain optimization, and advancing biomedical research. However, their predictive capacity is often compromised by incomplete reaction networks, stemming from gaps in biochemical knowledge, annotation inaccuracies, and insufficient experimental validations. Here we present MuSHIN (Multi-way SMILES-based Hypergraph Interface Network), a novel deep hypergraph learning method that integrates network topology with biochemical domain knowledge to predict missing reactions in GEMs. Evaluated on 926 high- and intermediate-quality GEMs with artificially removed reactions, MuSHIN significantly outperforms state-of-the-art methods, achieving up to a 17% improvement across multiple metrics and maintaining robust recovery even under severe network sparsity. Furthermore, MuSHIN substantially enhances phenotypic predictions in 24 draft GEMs associated with fermentation by resolving critical metabolic gaps, as validated against experimental measurements. Together, these findings highlight MuSHINs potential to advance GEM reconstruction and accelerate discoveries in systems biology, metabolic engineering, and precision medicine.

bioinformatics↗

bioRxiv2025

DREAM-GNN: Dual-route embedding-aware graph neural networks for drug repositioning

Drug repositioning presents a compelling strategy to accelerate therapeutic development by uncovering new indications for existing compounds. However, current computational methods are often limited in their ability to integrate heterogeneous biomedical data and model the intricate, multi-scale relationships underlying drug-disease associations, while large-scale experimental validation remains prohibitively resource-intensive. Here we present DREAM-GNN (Dual-Route Embedding-Aware Model for Graph Neural Networks), a multi-view deep graph learning framework that incorporates biomedical domain knowledge with two complementary graphs capturing both topological structure and feature similarity to enable accurate and biologically meaningful prediction of drug-disease associations. Extensive experiments on benchmark datasets demonstrate that DREAM-GNN significantly outperforms current state-of-the-art methods in recovering artificially removed repositioning candidates. These results establish DREAM-GNN as a robust and generalizable computational framework with broad potential to streamline drug discovery and advance precision medicine.

bioinformatics↗

bioRxiv2025

Genetic Network Shaping Kenyon Cell Identity and Function in Drosophila Mushroom Bodies

Revealing the molecular mechanisms underlying neuronal specification and acquisition of specific functions is key to understanding how the nervous system is constructed. In the Drosophila brain, Kenyon cells (KCs) are sequentially generated to assemble the backbone of the mushroom body (MB). Broad-complex, tramtrack and bric-{square}-brac zinc finger transcription factors (BTBzf TFs) specify early-born KCs, whereas the essential TFs for specifying late-born KCs remain unidentified. Here, we report that Pipsqueak domain-containing TF Eip93F promotes the identity of late-born KCs by reciprocally regulating gene expression in main KC types. Moreover, Eip93F not only regulates the expression of calcium channel Ca-1T in late-born KCs to functionally control animal behavior, but it also forms a genetic network with BTBzf TFs to specify the identities of main KC types. Our study provides crucial information linking KC-type diversification to unique function acquisition in the adult MB.

developmental biology↗

bioRxiv2025

Transformer-based Deep Learning for Glycan Structure Inference from Tandem Mass Spectrometry

Glycans play critical roles in diverse biological processes, but their structural analysis by tandem mass spectrometry (MS/MS) remains a major challenge due to their branched structure and stereochemistry. Traditional computational methods, such as database searching, are constrained by the scope of existing libraries and can be computationally intensive. While recent deep learning models have advanced the field, they often struggle to capture the complex, long-range dependencies within MS/MS spectra required for accurate inference. To address these challenges, we present GlycoBERT and GlycoBART, novel transformer-based models for glycan structure prediction from MS/MS data. GlycoBERT, a sequence classifier, achieves 95.1% structural accuracy on test data, surpassing the current state-of-the-art deep learning model, CandyCrunch. However, classification-based methods are inherently limited to predicting structures present in the training data. To overcome this, we developed GlycoBART, a generative sequence-to-sequence model capable of de novo glycan inference. On independent validation datasets, both models demonstrate robust performance. Critically, when applied to an MS/MS dataset from human embryonic kidney cells, GlycoBART generated two de novo glycan structures absent from the training set, one of which is a novel structure not catalogued in major glycan databases. Together, GlycoBERT and GlycoBART establish a new benchmark for glycan analysis, offering a powerful framework that enables more accurate and comprehensive exploration of glycan diversity and discovery.

bioinformatics↗

bioRxiv2025

Evolution of antibody cross-reactivity to influenza H5N1 neuraminidase from an N2-specific germline

The ongoing spread of highly pathogenic avian influenza H5N1 clade 2.3.4.4b virus in animals and its occasional spillover to humans have raised concerns about a potential H5N1 pandemic. Although recent studies have shown that pre-existing human antibodies can recognize H5N1 neuraminidase, the molecular basis of how this cross-reactivity develops remains poorly understood. In this study, we used a phage display antibody library derived from 245 healthy donors to isolate an antibody, HB420, that cross-reacts with neuraminidases of human H3N2 and avian H5N1 clade 2.3.4.4b viruses and confers protection in vivo. Cryo-EM analysis reveals that HB420 targets the neuraminidase active site by mimicking sialic acid binding through a single Asp residue. Furthermore, the inferred germline of HB420 is N2-specific but acquires cross-reactivity to H5N1 neuraminidase through somatic hypermutation. Overall, our findings provide insights into how neuraminidase antibody evolves breadth, which has important implications for the development of broadly protective influenza vaccines.

immunology↗

bioRxiv2025

ZCCHC4 Promotes Translation of Replication-dependent Histone mRNAs by Recruiting Cytoplasmic eIF3 complex

Chromatin instability is a major driving force in tumorigenesis, so the correct chromatin structure is essential for maintaining genome stability during the various stages of the cell cycle and DNA template processes. However, there is limited research on how histones are regulated and whether insufficient histone supply leads to chromatin instability, thereby affecting the process of tumor generation. Our study demonstrates that ZCCHC4 exhibits dual subcellular localization in both the nucleus and cytoplasm. Notably, in the cytoplasm, ZCCHC4 facilitates the translation of replication-dependent (RD) histone mRNAs, potentially through direct interaction with components of the translation initiation machinery. Depletion of ZCCHC4 results in a global reduction in protein translation, particularly affecting the translation of replication-dependent histones. This translational impairment prolongs the S phase of the cell cycle, leading to insufficient histone supply, chromatin relaxation, and genomic instability. Moreover, ZCCHC4 knockout markedly inhibits cell proliferation, colony-forming ability, and the growth of xenograft tumors. In conclusion, this study demonstrates that ZCCHC4 plays a pivotal role in regulating global protein translation by interacting with the eIF3 complex, which is essential for efficient mRNA translation in cancer cells and for sustaining the supply of replication-dependent histones during the S phase.

cancer biology↗

bioRxiv2025

Bidirectional regulation of glycoprotein nonmetastatic melanoma protein B by β-glucocerebrosidase deficiency in GBA1 isogenic dopaminergic neurons from a patient with Gaucher disease and parkinsonism

Variants in GBA1 are common genetic risk factors for several synucleinopathies. The increased risk has been attributed to the toxic effects of misfolded glucocerebrosidase (GCase) (gain-of-function), and the accumulation of lipid substrates due to reduced enzyme activity (loss-of-function). To delineate GBA1 pathogenicity, an iPSC line was generated from a patient with both type 1 Gaucher disease (GBA1: N370S/N370S; p.N409S/p.N409S) and Parkinson disease (PD). From this line, we created a reverted wild-type (WT) line and a GBA1 knockout (KO) line to eliminate misfolded GCase and intensify lipid accumulation. N370S/N370S and KO dopaminergic neurons (DANs) exhibited decreasing GCase levels and progressive accumulation of lipid substrates compared to WT DANs. Notably, the expression of GPNMB, whose levels correlate with PD risk, was upregulated by the mild lipid accumulation in N370S/N370S DANs but disrupted in KO DANs. These findings refine the loss-of-function mechanism by associating PD risk levels of GPNMB with lipid levels specific to GBA1 risk variants.

neuroscience↗

bioRxiv2025

Super-giga and tiny orchid genomes illuminate evolution of Orchidaceae

Orchidaceae (orchids) is commonly known as one of the largest families of seed plants, and grow in an extensive range of habitats worldwide. In the present study, we generated chromosome-level reference genomes for two orchids using a combination of PacBio, Illumina, and Hi-C sequencing, Cypripedium singchii has the largest genome and chromosomes among the sequenced species so far, with a genome size of 43.19 Gb (1C) with ten chromosomes, and Apostasia fujianica has the smallest known genome and chromosomes in Orchidaceae, with a genome size of 340.90 Mb (1C) with 35 chromosomes. We predicted a total of 32,412 and 21,724 protein-coding genes for C. singchii and A. fujianica, respectively. The overall BUSCO score was 85.01% for C. singchii and 91.80% in A. fujianica. Based on protein-coding sequences from 55 conserved single-copy families across 21 plant species, we constructed a high-confidence phylogenetic tree and estimated the divergence times. The high-quality genomes of super-giga and tiny orchids offer key insight for future evolutionary researches.

genomics↗

bioRxiv2025

Complex I Governs Iron Levels in Bat Mitochondria to Couple Respiration with Lipid Metabolism

The evolution of mitochondrial respiration has been implicated in the remarkable adaptations of bats, including powered flight, longevity and viral tolerance. However, the mechanisms coordinating mitochondrial respiration with lipid metabolism--a critical energy source for flight--remain poorly understood. Here, we discovered that complex I inhibition triggered iron overload into bat mitochondria, inducing a burst of reactive oxygen species (ROS). This iron translocation relied on the VDAC-MCU axis and mitochondrial membrane potential. Intriguingly, disrupting VDAC oligomerization enhanced iron uptake in a complex I-dependent manner. Mitochondrial iron translocation perturbed fatty acid oxidation and upregulated linoleic acid species, which promoted the reversible accumulation of lipid droplets, suggesting metabolic plasticity crucial for bat energetics. Our findings establish complex I as a pivotal regulatory hub that directly couples respiratory efficiency to lipid metabolism by governing mitochondrial iron levels. This mechanism may represent a key evolutionary adaptation supporting the unique metabolic demands of bat physiology. HIGHLIGHTSComplex I inhibition induces unique mitochondrial ROS burst in bat cells. Mitochondrial iron overload is the primary trigger of complex I-linked ROS burst. VDAC-MCU axis couples membrane potential to drive mitochondrial iron influx. Mitochondrial iron translocation remodels lipid metabolism in bat cells.

cell biology↗

bioRxiv2025

Single-cell spatial map of cis-regulatory elements for disease-related genes in the macaque cortex

Single-cell spatial transcriptomes have demonstrated molecular and cellular diversity in the brain1, 2, 3, 4, 5, 6, 7, 8, but gene regulatory mechanisms underlying transcriptomic profiles and disease pathogenesis remain largely unknown in primates. Here we performed single-nucleus Assay for Transposase-Accessible Chromatin followed by sequencing (snATAC-seq) for [~]1.6 million cells from 142 cortical regions of two male cynomolgus monkeys (Macaca fascicularis), and identified distinct chromatin accessibility profiles of cis-regulatory elements (CREs) for various cell types. By integrative analysis with large-scale spatial transcriptome data, we found that these CREs showed laminar and regional preferences, with their regional accessibility exhibiting striking dependence on the regions hierarchical level. Cross-species comparison of snATAC-seq data revealed human/macaque-enriched layer-4 glutamatergic neurons and LAMP5/LHX6-expressing GABAergic neurons as well as human/macaque-biased CREs for genes related to neurodevelopment and psychiatric diseases. Importantly, risk single-nucleotide polymorphisms for many brain disorders strongly associated with human/macaque-biased CREs in glutamatergic neuronal types and those for Alzhemers disease strongly associated with CREs exclusively in microglia. Our results provided the basis for understanding the spatial gene regulatory mechanisms underlying cellular diversity and disease pathogenesis in the primate cortex.

neuroscience↗

bioRxiv2025

Sequential transcriptional gates in the thalamo-cortical circuit coordinate memory stabilization

The molecular mechanisms that enable memories to persist over long time-scales from days to weeks and months are still poorly understood. To develop insights we created a behavioral task where, by varying the frequency of learned associations, mice formed multiple memories but only consolidated some, while forgetting others, over the span of weeks. We then monitored circuit-specific molecular programs that diverge between consolidated and forgotten memories. We identified multiple distinct waves of transcription, i.e., cellular macrostates, specifically in the thalamo-cortical circuit, that defined memory persistence. Notably, a small set of transcriptional regulators orchestrated broad molecular programs that enabled entry into these macrostates. Targeted CRISPR-knockout studies revealed that while these transcriptional regulators had no effects on memory formation, they had prominent, causal, and strikingly time-dependent roles in memory stabilization. In particular, the calmodulin-dependent transcription factor Camta1 was required for initial memory maintenance over days, while Tcf4 and the histone methyl-transferase Ash1l were required later to maintain memory over weeks. These results identify a critical Camta1-Tcf4-Ash1l thalamo-cortical transcriptional cascade required for memory stabilization, and puts forth a model where the sequential, multi-step, recruitment of circuit-specific transcriptional programs enable memory maintenance over progressively longer time-scales.

neuroscience↗

bioRxiv2025

TypeSINE: Genome-Wide Detection of SINE Retrotransposon Polymorphisms Reveals Functional Variants Linked to Body Size Variation in Pigs

The genotyping of Short Interspersed Element (SINE) retrotransposon insertion polymorphisms (RIPs) from large-scale sequencing data remains technically challenging. To overcome this limitation, we developed TypeSINE to genotype SINE-RIPs from short-read sequencing data. Analyzing 362 porcine genomes, we identified 749,835 SINE-RIPs (>85% accuracy), including 65,917 common variants (5-95% frequency). These showed uniform genomic distribution ([~]22/Mb), with 40% in introns. Population analyses based on SINE-RIPs revealed independent domestication of Asian and European pigs from local wild populations, followed by introgression. GWAS detected body size-associated regions using SINE-RIPs. We created pigRIPdb (82,227 curated SINE-RIPs) with browsing and visualization tools. The methods cross-species applicability is supported by conserved SINE evolutionary patterns, enabling polymorphism discovery of SINEs in livestock and humans. TypeSINE represents an efficient, scalable solution for genome-wide SINE-RIP analysis, advancing population genomics research. TeaserAn integrated SINE-RIP genotyping platform with companion database enables large-scale analysis of SINE evolution and population genetics in swine

bioinformatics↗

bioRxiv2025

Source tracking of larval bacterial community of Pacific white shrimp across the developmental cycle: ecological insights for microbial management and pathogen prevention in larviculture

Health of postlarvae (as products of larviculture) of Pacific white shrimp (Penaeus vannamei) is critical to subsequent culture. Larval microbiota establishment has potential long-term impact on host health, but bacterial sources of larval shrimps remain largely unexplored. We investigated the larviculture cycles of two shrimp strains by utilizing 16S rRNA gene sequencing alongside mainstream microbial source-tracking tools. Our goal was to evaluate the contributions of environmental elements, feeds, and internal succession to larval bacterial community. Regardless of shrimp strains or tools, the mouth opening marked a pivotal shift in bacterial source composition of larvae, with a notable contribution from the rearing water at the beginning of mouth opening, followed by enhanced internal succession, emphasizing the critical role of early microbiota establishment. Notably, the dominance of external sources at zoea II highlights the necessity for ensuring microbial safety due to the prevalence of zoea-II syndrome at this vital stage. Source tracking of specific bacterial groups suggests regulatory timing and routes of Roseobacteraceae (with potential probiotics) and frozen Artemia as a major source of vibrios (with opportunistic pathogens). The determinism-dominated assembly of bacteria, sourced either internally or externally, highlights strong host selection. This process exhibits overall similarities in taxonomic preferences for internal succession, while differences in taxonomic preferences from external sources reflect the influence of the genetic backgrounds of shrimp strains. Our work presents the first systematic assessment on bacterial sources of P. vannamei larvae across the complete developmental cycle, providing new insights that facilitate microbial management and pathogen prevention in larviculture.

microbiology↗

bioRxiv2025

PEARL: Integrative multi-omics classification and omics feature discovery via deep graph learning

Integrating multi-omics data provides valuable insights into biological processes by capturing information across multiple molecular layers, enabling a comprehensive understanding of complex diseases and driving advancements in precision medicine. However, existing computational methods for multi-omics integration face significant challenges, such as low reliability and poor generalizability, due to the high dimensionality and low sample size nature of omics data. To address these challenges, we present PEARL (Pearson-Enhanced spectrAl gRaph convoLutional networks), a novel deep graph learning method for biomedical classification and functional important omics features identification. PEARL leverages a simple yet effective learning architecture to achieve superior and robust performance in high-dimensional, low-sample-size multi-omics settings. Our results demonstrate that PEARL significantly outperforms existing state-of-the-art methods on both synthetic and real biomedical datasets. Furthermore, applied to Alzheimers disease (AD) brain multi-omics data, features prioritized by PEARL lead to functionally important genes that demonstrate significant enrichment in AD-related pathways. These findings highlight PEARLs practical utility in biomedical research and its potential to enhance biological interpretability in multi-omics studies.

bioinformatics↗

bioRxiv2025

Rab11FIP2 controls NLRP3 inflammasome activation through Rab11b

Membrane trafficking through the trans-Golgi network has recently been shown to guide activation of the NLRP3 inflammasome. The GTPases Rab11a and Rab11b, and their effector molecule Rab11-FIP2, are regulators of endosome trafficking and retrograde transport. Rab11-FIP2 binds phosphatidylinositol species including PI4P, enriched in the trans-Golgi network and peripheral endosomes following NLRP3 inflammasome activation. We here demonstrate that Rab11-FIP2 and Rab11b, but not Rab11a, control caspase-1 mediated cleavage of pro-IL-1{beta} and GSDMD, and pyroptotic cell death in human macrophages. Rab11-FIP2 also controlled LPS stimulated IKK{beta} activation by TAK1 and IKK{beta} mediated NLRP3 translocation to the trans-Golgi network. Furthermore, we show that NLRP3 bound Rab11-FIP2 via its KMKK motif and that Rab11-FIP2 interacts with NLRP3 via its N-terminal C2-domain. The formation of PI4P positive endosomes and ASC-specks were also controlled by Rab11-FIP2. Collectively our results demonstrate that Rab11-FIP2 and Rab11b control NLRP3 inflammasome activation on early endosomes in human macrophages.

immunology↗

bioRxiv2025

Multiple-omics analyses of the Alviniconcha holobiont reveal multi-faceted adaptations to deep-sea hydrothermal vents

Deep-sea hydrothermal vents are extreme environments with constantly fluctuating physicochemical conditions - but dense animal aggregations thrive primarily through symbiosis with chemoautotrophic bacteria to exploit the unusual chemistry. Alviniconcha snails, which harbor symbionts in their enlarged gill at an intracellular-extracellular intermediate state, are a prime example. Here, we present chromosome-level genomes of two Alviniconcha species (A. adamantis and A. marisindica) to investigate the adaptations of this holobiont. Significant expansion of solute carrier families enhances nutrient transport between the two parties. Alviniconcha lacks complete methionine biosynthesis pathways, compensated by symbiont provisioning, highlighting host-symbiont metabolic complementarity. High myoglobin expression in gills, contradicting prior reports of hemoglobin, suggests myoglobin-mediated oxygen storage to mitigate fluctuating environmental oxygen. Spatial transcriptomics further delineated gills functional zones on the gill filament responsible for symbiont digestion via phagocytosis in bacteriocytes, oxygen transport in secretory zones, and ciliary water flow regulation. Our findings elucidate molecular and physiological adaptations underpinning the Alviniconcha holobionts success in dynamic and treacherous vent ecosystems.

microbiology↗

bioRxiv2025

PIN1 Drives Cellular Plasticity and Immune Modulation in Chronic Pancreatitis

Background and AimsChronic pancreatitis (CP) is characterized by inflammation, fibrosis, and acinar-to-ductal metaplasia (ADM). PIN1, known to drive oncogenic signaling and cellular plasticity in cancer, has an unexplored role in CP. This study investigates PIN1s expression and function in CP pathogenesis using human tissues and mouse models. MethodsPIN1 expression was assessed in human CP tissue microarrays (TMAs) via immunohistochemistry (IHC) and cyclic immunofluorescence (CyCIF). Acute and chronic pancreatitis were induced in wild-type (WT) and PIN1 knockout (PIN1KO) mice using caerulein. Disease progression was monitored histologically, and immune profiling was conducted using flow cytometry. Pharmacological inhibition was performed using a small molecule PIN1 inhibitor-Sulfopin, and effects were evaluated by histology, qPCR, and cytokine analysis. Single-cell RNA sequencing (scRNA-seq) was performed on pancreatic tissues to perform pathway analysis and intercellular communication. ResultsPIN1 expression was elevated in human CP tissues, correlating with disease severity and ADM. In mice, both acute and chronic pancreatitis increased PIN1 expression, but only in our chronic PIN1KO mice displayed reduced pancreatic injury, fibrosis, ADM, and modulated immune infiltration. Pharmacological PIN1 inhibition mimicked the protective effects of genetic knockout, dampening inflammatory pathways. scRNA-seq revealed that PIN1 inhibition altered the intercellular communication networks between epithelial, immune, and stromal cells. ConclusionPIN1 drives cellular plasticity, immune modulation, and disease progression in CP. Targeting PIN1 may offer a therapeutic strategy to mitigate CP. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=184 HEIGHT=200 SRC="FIGDIR/small/653850v1_ufig1.gif" ALT="Figure 1"> View larger version (36K): org.highwire.dtl.DTLVardef@67f9fcorg.highwire.dtl.DTLVardef@4d4c1forg.highwire.dtl.DTLVardef@c099d7org.highwire.dtl.DTLVardef@b41aa5_HPS_FORMAT_FIGEXP M_FIG C_FIG Created in BioRender. Shah, V. (2025) https://BioRender.com/undefined

cell biology↗

bioRxiv2025

Multiscale patterning of a model apical extracellular matrix revealed by systematic endogenous protein tagging

Barrier epithelia are shielded from the external environment by their apical extracellular matrices (aECMs). The molecular complexity of aECMs has challenged understanding of their organization in vivo. To define the molecular architecture of a model aECM we generated a toolkit of 102 fluorescently tagged aECM components using gene editing in C. elegans, focusing on proteins secreted by the epidermis to form the collagen-rich cuticle. We developed efficient pipelines for modular protein tagging and rapid fluorophore swapping. Most tagged collagens were functional and exhibited exquisitely specific patterning across stages, cell types, and matrix substructures. We define multiple reference markers for key substructures including the little-understood cortical layer, as well as the helical crossed fiber arrays that function as a hydrostatic skeleton to maintain organismal shape. We further tagged >30 members of key aECM protein classes including proteases, protease inhibitors, and lipid transporters. Our standardized markers will allow dissection of the mechanistic basis of aECM spatiotemporal patterning in vivo. HighlightsO_LIFirst large-scale protein tagging resource for the apical extracellular matrix C_LIO_LIOptimization of CRISPR methods for protein tagging including color swaps C_LIO_LITagged proteins are functional and exhibit a high degree of stage-, cell- and compartment specificity C_LIO_LIReference localization patterns for multiple aECM compartments and markers for newly defined compartments C_LI

developmental biology↗