Biology subjects

Xie, S.

Publications and source records attributed to Xie, S..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Targeting FGFR signaling overcomes therapeutic resistance and immune evasion in oncogenic PIK3CA-driven serous-like endometrial cancer

Serous endometrial cancer (SEC) is an aggressive subtype of endometrial cancer (EC) with poor prognosis and limited treatment options. Here, we developed a clinically relevant, immunocompetent serous-like mouse model incorporating oncogenic PIK3CA mutation, Trp53 loss, and MYC overexpression. Using this model together with human EC cell lines, patient-derived organoids (PDOs), xenografts, and patient datasets, we investigated mechanisms underlying resistance to PI3K-targeted therapy. Single-cell profiling reveals that FGFR1/2 upregulation associates with intrinsic resistance, whereas FGFR3 characterizes acquired resistance. Dual FGFR and PI3K inhibition produced superior tumor control compared with either agent alone. Mechanistically, FGFR signaling promotes immune evasion by downregulating MHC-I/HLA-mediated antigen presentation and enriching M2-type tumor-associated macrophages. FGFR inhibition reversed these changes and synergized with anti-PD-1 therapy to enhance antitumor immune responses and establish durable immune memory. Collectively, these findings identify FGFR signaling as a key driver of therapeutic resistance and immune escape in SEC and support FGFR-targeted combination strategies.

cancer biology↗

bioRxiv2026

Using user-centered design to better understand challengesfaced during genetic analyses by novice genomicresearchers

The lack of user-centered design principles in the current landscape of commonly-used bioinformatics software tools poses challenges for novice genomics researchers (NGRs) entering the genomics ecosystem. Comparing the usability of one analysis software to that of another is a non-trivial task and requires evaluation criteria that incorporates perspectives from both existing literature and a diverse, underrepresented user base of NGRs. To better characterize these barriers, we utilized a two-pronged approach consisting of a literature review of existing bioinformatics tools and semi-structured interviews of the needs of NGRs. From both knowledge sources, the key attributes that resulted in poor adoption and sustained use of most bioinformatics tools included poor documentation, lack of readily-accessible informational content, challenges with installation and dependency coordination, and inconsistent error messages/progress indicators. Combining the findings from the literature review and the insights gained by interviewing the NGRs, an evaluation rubric was created that can be utilized to grade existing and future bioinformatics tools. This rubric acts as a summary of key components needed for software tools to cater to the diverse needs of both NGRs and experienced users. Due to the rapidly evolving nature of genomics research, it becomes increasingly important to critically evaluate existing tools and develop new ones that will help build a strong foundation for future exploration.

bioinformatics↗

bioRxiv2026

NanoSimFormer: An end-to-end Transformer-based simulator for nanopore sequencing signal data

Nanopore sequencing has achieved a new standard of accuracy with the advent of R10.4.1 flow cell and high-performance Transformer-based basecalling models. However, existing signal simulators often fail to capture the complex, non-linear dynamics of nanopore current signals, relying on static pore models or lacking optimization objectives linked to basecalling, resulting in synthetic signals with substantially lower accuracy and fidelity than experimental data. To address this, we introduce NanoSimFormer, an end-to-end Transformer-based signal simulator that integrates basecaller guidance during training to generate high-fidelity nanopore signals explicitly optimized for accurate calling. Rigorous evaluation across diverse human, bacterial, and fungal R10.4.1 DNA sequencing datasets demonstrates that NanoSimFormer consistently outperformed competing methods (seq2squiggle and Squigulator), achieving median read accuracies exceeding 99% and Q-scores above 22.8, closely matching experimental baselines. NanoSimFormer faithfully recapitulated experimental variant calling performance on the human HG002 sample, achieving F1-scores of 0.9967 for SNPs and 0.8295 for small indels, and notably minimized false-positive errors in homopolymer and short tandem repeat (STR) regions where other simulators struggled. Furthermore, NanoSimFormer-derived reads enabled high-quality de novo bacterial assembly with consensus error rates below one mismatch per 100 kbp, comparable to experimental assemblies, and preserved fungal mock community structures with high correlation to experimental abundance profiles in metagenomic benchmarks. With tunable parameters for amplitude noise and event duration variance, NanoSimFormer enables the simulation of datasets spanning a wide range of data qualities. Together, these results establish NanoSimFormer as a robust tool for benchmarking and algorithm development in the latest nanopore sequencing era.

bioinformatics↗

bioRxiv2026

Identification and inhibition of the Cyclin D Rb-docking interface that drives cell division

The animal cell division cycle is initiated by the cyclin-dependent kinases CDK4 and CDK6 in complex with D-type cyclins. Cyclin D-CDK4/6 complex formation is promoted by the assembly factors p21 and p27, which bind both subunits. p27 binds the hydrophobic patch on cyclin D that is similar to the patch used by other cell cycle cyclins to dock their substrates. This raised the question as to how cyclin D could find its substrates if its hydrophobic patch were already occupied? Here, we show that D-type cyclins use their A2 helix to dock the retinoblastoma protein Rb, a key substrate regulating cell cycle progression. The specific interface of cyclin Ds A2 helix is unique among cyclins and its mutation slows proliferation. Taken together, our work identifies a cyclin D-substrate docking mechanism that can be targeted by novel cancer therapeutics.

biochemistry↗

bioRxiv2026

DNA sequence quantitatively encodes CTCF-binding affinity at genome scale

CTCF is a central architectural protein that shapes 3D genome organization through sequence-specific DNA binding, but how DNA sequence quantitatively determines CTCF-binding strength remains poorly understood. Progress has been hampered by the lack of large, high-quality measurements of binding affinity. Here, we experimentally determine in vitro CTCF-binding affinity for 276,765 DNA sequences derived from the human genome, generating a comprehensive quantitative landscape of CTCF-DNA interactions. Leveraging these data, we develop DeepCTCF, a deep learning model that predicts CTCF-binding strength directly from DNA sequence and enables quantitative interpretation of CTCF motif grammar. Using this framework, we systematically dissect how specific sequence features modulate CTCF-binding affinity and generate quantitative predictions for disease-associated variants that alter CTCF binding. Together, this study defines general principles by which DNA sequence encodes CTCF-binding affinity and provides a quantitative framework for interpreting regulatory sequence variation. TeaserMapping how DNA encodes CTCF binding reveals quantitative rules across over 1{per thousand} of the human genome.

genomics↗

bioRxiv2025

Polus: a Transformer-based Soft-decision Codec Enhancement Platform for DNA Storage

DNA storage offers exceptional information density and archival longevity, but is constrained by the complex, heterogeneous errors inherent to synthesis, storage, and sequencing. Conventional error-correction schemes often rely on excessive logical redundancy to mitigate these biochemical imperfections, thereby compromising storage efficiency. Here, we introduce Polus, a deep-learning-enabled platform that bridges the gap between biochemical constraints and digital reliability through soft-decision decoding. At its core is SeqFormer, a Transformer-based channel model that synergizes sequence context with quality signals to characterize platform-specific error profiles, generating calibrated per-base confidence scores. This mechanism transforms uncertain biochemical noise into informative "soft" erasures. In in silico benchmarks, Polus significantly enhances mainstream codecs: it reduces the sequencing coverage required for DNA Fountain by 38.9% --increasing effective physical density by [~]80%--and eliminates persistent indel-induced errors in the Yin-Yang codec. Furthermore, it enables a targeted resequencing strategy that achieves full recovery with 99.9% less overhead than brute-force deepening. To formalize these gains and address the lack of systematic benchmarking in the field, Polus establishes a standardized nine-metric evaluation framework that rigorously quantifies the trade-offs between reliability, density, and cost. This work provides a reproducible, quantitative foundation for next-generation, context-aware DNA storage systems.

bioinformatics↗

bioRxiv2025

Multiscale characterisation of the human claustrum from histology to MRI

Though linked to an unusually broad array of functions, the human claustrums complex morphology has hindered in vivo study, resulting in a small MRI literature marked by implausibly large discrepancies in reported characteristics. We constructed the first three-dimensional histological "gold standard" claustrum model, and systematically evaluated in vivo 7-Tesla MRI datasets against it and downsampled derivatives. MRI showed resolution-dependent differences rather than contrast limitations, transforming the claustrums intricate sheet into an artefactually thickened ribbon. However, submillimetre MRI reliably recovered the dorsal "core" that contains most claustral volume and density and houses major corticoclaustral connectivity. At 0.5mm resolution, extension into the temporal lobe, including irregular ventral "puddles", was partially recovered, with uncertainty reflecting boundary imprecision rather than anatomical loss. Our results refute the view that the claustrum is inaccessible in the living human brain, define practical measurement limits, and provide a foundation for future functional investigations.

neuroscience↗

bioRxiv2025

Evolutionary landscape of oral microbiome over 100,000 years

The human oral microbiome represents an intimate and enduring partnership between host and microbial communities, yet its evolutionary trajectory across deep time remains largely uncovered. Here we reconstruct the global history of the oral microbiome over 102,400 years using a global collection of 1854 oral samples spanning 61 countries. We reveal that major cultural transitions of the Neolithic Revolution, industrialization, and recent medical advances have driven a directional, globally coordinated co-evolution of oral microbial ecology, function, and population structure. Network analysis identifies two universal microbial modules of dental calculus modules: the commensal-enriched DCM1 module and the pathogen-enriched DCM2, both exhibiting progressive ecological polarization. DCM2 surged around 4,000 years ago with agricultural intensification but subsequently declined in modern European populations while persisting at high levels in African and some American groups, a geographic divide mirroring socioeconomic inequities. Species-level analyses demonstrate widespread directional changes, with obligate anaerobes declining as aerotolerant taxa expand, reflecting a carbohydrate-associated functional remodeling toward starch metabolism. Phylogenomic reconstruction reveals contrasting evolutionary trajectories. Pauljensenia mediterranea experienced near-extinction following a severe bottleneck, while Actinomyces israelii underwent a selective sweep with rapid population expansion 1,000 years ago. Critically, the oral resistome has accelerated in the antibiotic era, resulting in [~]33-50-fold surge of highly mobile, clinically relevant antimicrobial resistance genes in the last century, with Streptococcus acting as a central hub of dissemination. Collectively, these results position the oral microbiome as an active participant that has co-evolved with human subsistence, urbanization and medical practice, with direct implications for global oral health, antimicrobial resistance and interventions aimed at restoring resilient microbial states.

microbiology↗

bioRxiv2025

A cytosolic function of DNMT1 controls neuronal morphogenesis via microtubule regulation

Proteins traditionally confined to a single cellular compartment are increasingly recognized to exert non-canonical functions in alternative domains. The DNA methyltransferase 1 (DNMT1), classically defined as the maintenance methyltransferase that preserves DNA methylation patterns during replication, exemplifies this versatility. Beyond its canonical role, DNMT1 is highly expressed in the developing and adult brain, where it contributes to transcriptional regulation in postmitotic neurons. Notably, cytoplasmic DNMT1 localization has been observed in neural cells, and emerging evidence links DNMT1 to mitochondrial function with implications for neurodegenerative disease, whereby the underlying functional mechanisms remain to be fully elucidated. Here, we identify a previously unrecognized cytosolic function of DNMT1 in developing cortical excitatory neurons. Through genetic perturbation, proteomics, and high-resolution imaging, we show that DNMT1 regulates dendritic and axonal branching independently of its catalytic activity and nuclear localization. Instead, DNMT1 operates as a cytosolic scaffold interacting with the polarity regulator DOCK7 to modulate Rac1-STMN1 signaling, microtubule dynamics, and organelle trafficking. These findings expand the conceptual framework of DNMT1 from a genome guardian to a dual-compartment regulator that coordinates cytoskeletal remodeling and mitochondrial positioning. Beyond advancing our understanding of neuronal morphogenesis, this work provides mechanistic insight into how DNMT1 mutations may lead to neurodegenerative diseases.

neuroscience↗

bioRxiv2025

Paw switching with lateralized cholinergic modulation

Motor preferences, such as handedness, reflect fundamental asymmetries in brain function and behavior across vertebrate species, including humans and rodents1. Although individual hand or paw preferences are typically stable, they can be reshaped through experience or training, underscoring the plasticity of lateralized motor circuits2. However, the neural mechanisms that enable this behavioral switching remain unclear. Here we show that experience-driven shifts in paw preference in mice arise from asymmetric cholinergic modulation between the two hemispheres. Using a behavioral paradigm designed to induce paw switching, we found that a subset of animals developed a stable change in preference following training. Selective activation or inhibition of cholinergic neurons on one side of the brain was sufficient to bias this switching behavior. During lateralized movements, an endogenous imbalance in cholinergic activity emerges between the hemispheres, suggesting that interhemispheric cholinergic asymmetry enables motor flexibility. These findings identify a previously unrecognized mechanism by which cholinergic signaling dynamically regulates lateralized motor control, providing insight into how hemispheric imbalances may shape persistent individual motor preferences.

neuroscience↗

bioRxiv2025

A Fkh1/2 binding site array in the WHI5 promoter drives sub-scaling transcription

Cells typically regulate their size within a relatively tight range by coupling growth to the cell division cycle using a dedicated set of molecular mechanisms. In budding yeast, cells are born with a similar amount of the G1/S inhibitor protein Whi5 that is then diluted by growth throughout G1. As cells grow, Whi5 concentration decreases and cells become more likely to enter the cell cycle. Cells are born in G1 with similar amounts of Whi5 because of the size-independent (sub-scaling) expression of WHI5 mRNA during S/G2/M phases and the equal partitioning of Whi5 protein at division. While the latter is known to be achieved by association with chromatin before anaphase, the mechanism for the former is poorly understood. Through systematic mutations of the WHI5 promoter, we discovered that WHI5s core promoter region located -126 to -75 base pairs upstream of the start codon is responsible for sub-scaling expression. This sequence contains a repeating array of binding sites for the transcription factors Fkh1 and Fkh2. Mutation of any of these sites, deletion of either FKH1 or FKH2, or preventing Fkh1 or Fkh2 dimerization weakens the sub-scaling of WHI5 transcription. Taken together with structural predictions and a mathematical model of cooperative Fkh-DNA binding, we conclude that WHI5s sub-scaling transcription is regulated by a Fkh1/2 heteropolymer binding an array of sites in its core promoter.

cell biology↗

bioRxiv2025

Programmable Aptamer-Embedded Circular RNAs for Targeted Antigen-Presenting Cells Immunotherapy

The targeted delivery of mRNA to specific cell types in vivo remains a major unmet need in mRNA-based therapeutics. Conventional strategies rely on nanocarriers engineered with various modifications. However, these exogenous materials could cause adverse reactions, and their complex manufacturing processes often hinder clinical application. By leveraging the innate programmability of RNA sequences, we herein report a carrier-free, scalable, and programmable aptamer-embedded circular RNAs (Apt-circRNAs) design that confers targeting capability through the strategic incorporation of different aptamers at varying densities, while retaining circular RNA stability and protein production capability. Experimental results show that antigen-loaded Apt-circRNAs enable targeted delivery to antigen-presenting cells (APCs), drive endogenous antigen expression, activate antigen-specific T cells, and elicit potent immune responses that mediate clearance of both early-and late-stage tumors in mouse models. In humanized models of colorectal cancer, the combination of Apt-circRNA and PD-1 blockade produced synergistic antitumor effects and significantly reshaped the tumor microenvironment. In a first-in-human (FIH) clinical trial, Apt-circRNA vaccine encoding KRAS G12D/G12V neoantigens exhibited a favorable safety profile and induced robust immune activation, T cell mobilization, and neoantigen-specific immune responses. With this study, we have established a new paradigm in carrier-free, APC-targeted RNA vaccine design, offering a safe and effective platform for next-generation cancer immunotherapy.

bioengineering↗

bioRxiv2025

Gene-sized DNA insertion at genomic safe harbors in human cells using a site-directed transposase

Achieving precise and efficient integration of gene-sized DNA sequences into the human genome remains a major obstacle to gene therapy. Existing approaches depend on double-strand DNA breaks, which can lead to unintended genome alterations. Many monogenic diseases arise from diverse patient-specific mutations, making individualized correction impractical and underscoring the need for universal full-gene replacement strategies. We developed INsertion by Targeted Anchoring and Conditional Transposition (INTACT) to enable targeted insertion at genomic safe harbor loci. We engineered a mammalian transposase with mutations in its DNA-binding domain to reduce off-target integration. Site specificity was then restored by linking programmable sequence-specific DNA-binding proteins to the transposase. Systematic optimization of INTACT revealed key determinants of precision, including non-covalent linkage between the transposase and DNA-binding protein, strict spacing between the binding site and the TTAA insertion sequence, and linkage of the DNA-binding protein to an internal position within the transposase. On-target insertion was achieved across multiple loci, with optimized INTACT averaging 1.2 targeted insertions per cell. An off-target assay confirmed that DNA-binding domain mutations substantially reduced unwanted integration events to near-background levels. Our site-directed transposase enables precise, efficient genomic insertion of >4kb DNA without double-strand breaks, offering a powerful new tool for genome engineering.

molecular biology↗

bioRxiv2025

To jump or not to jump: Comparing effects of phenotypic plasticity on the visual responses and escape behavior of locusts and grasshoppers

Locusts exhibit remarkable phenotypic plasticity changing their appearance and behavior from solitary grasshoppers to gregarious locusts when population density increases. These changes include morphological differences in the size and shape of brain regions, but little is known about plasticity within individual neurons and alterations in behavior not directly related to aggregation or swarming. We examined looming escape behavior and the properties of a well-studied collision-detection neuron in gregarious and solitarious animals of three closely related species, the desert locust (Schistocerca gregaria), the Central American locust (S. piceifrons) and the American bird grasshopper (S. americana). For this neuron, the lobula giant movement detector (LGMD), we examined dendritic morphology, membrane properties, gene expression, and looming responses. Gregarious animals reliably jumped in response to looming stimuli, but surprisingly solitarious desert locusts did not produce escape jumps. These solitarious animals also had smaller LGMD dendrites. This is the first study done on three different species of grasshoppers to observe the effects of phenotypic plasticity on the jump escape behavior, physiology and transcriptomics of these animals. Surprisingly, there were little differences in these properties between the two phases except for behavior. For all the three species, gregarious animals jumped more than solitarious animals, but no significant differences were found between the two phases of animals in the electrophysiological and transcriptomics studies. Our results suggest that phase change impacts mainly the motor system and that the physiological properties of motor neurons need to be characterized to understand fully the variation in jump escape behavior across phases. New & Noteworthy (74 words): Swarming is observed in some grasshopper species, called locusts. We compared jump escape behavior between gregarious and solitarious grasshoppers and locusts, as well as LGMD responses to looming stimuli, and analyzed the morphological differences in this neuron. This study provides insights into the effects of phase change on the visual system of locusts and grasshoppers as it relates to looming-evoked jump escape behavior. In this context, our results suggest that phenotypic plasticity mainly impacts the motor system.

neuroscience↗

bioRxiv2025

Deep Generative Models for Discrete Genotype Simulation

Deep generative models open new avenues for simulating realistic genomic data while preserving privacy and addressing data accessibility constraints. While previous studies have primarily focused on generating gene expression or haplotype data, this study explores generating genotype data in both unconditioned and phenotype-conditioned settings, which is inherently more challenging due to the discrete nature of genotype data. In this work, we developed and evaluated commonly used generative models, including Variational Autoencoders (VAEs), Diffusion Models, and Generative Adversarial Networks (GANs), and proposed adaptation tailored to discrete genotype data. We conducted extensive experiments on large-scale datasets, including all chromosomes from cow and multiple chromosomes from human. Model performance was assessed using a well-established set of metrics drawn from both deep learning and quantitative genetics literature. Our results show that these models can effectively capture genetic patterns and preserve genotype-phenotype association. Our findings provide a comprehensive comparison of these models and offer practical guidelines for future research in genotype simulation. We have made our code publicly available at https://github.com/SihanXXX/DiscreteGenoGen.

bioinformatics↗

bioRxiv2025

Serotonin shapes the temporal window for associative fear learning

Fear learning is a critical adaptive mechanism that enables the association of an environmental cue (the conditioned stimulus, CS) with a potential threat (the unconditioned stimulus, US), even when these events are temporally discontiguous. When associations are instead formed between unrelated cues and threats, the maladaptive learning process can give rise to neurological disorders. Therefore, maintaining accurate temporal boundaries for forming such associations is crucial for distinguishing real threats from irrelevant stimuli. Despite its importance, the mechanisms defining this precise time window remain poorly understood. Here, we report that serotonin plays a crucial role in modulating the associable interval in trace fear conditioning. Specifically, we show that serotonergic projections from the dorsal raphe nuclei (DRN) to the CA1 region of the ventral hippocampus (vCA1) dynamically regulate the association of discontiguous CS and US. Mechanistically, we demonstrate that sustained serotonin release in the vCA1 maintains the ensemble activity of 5-HT2C receptor-expressing pyramidal neurons, facilitating the association between temporally separated events. In contrast, when the trace interval exceeds the associable window, transient serotonin release fails to drive vCA1 neuronal ensemble activity, thus preventing maladaptive learning. These findings highlight a key role for serotonin in regulating the temporal precision of fear learning and reveal the DRN-vCA1 serotonergic pathway as a potential therapeutic target for related disorders.

neuroscience↗

bioRxiv2025

O-mannose glycosylations influence E-cadherin functional interactions

Cadherins are plasma membrane proteins that play critical roles in maintaining cell-cell adhesion and modulating cell signaling during development. Their functions are mediated by extracellular cadherin (EC) domains, which facilitate adhesive interactions and enable the formation of cis- and trans assemblies at adherens junctions and desmosomes. EC domains adopt a characteristic immunoglobulin-like fold composed of seven {beta}-strands (A-G) and are modified by N -linked and O -linked glycosylations, including O -linked mannose monosaccharides (O -Man) on conserved serine and threonine residues of B- and G-strands. O -Man glycosylations on EC domains are catalyzed by the TMTC1-4 enzymes, with different TMTC enzymes modifying B- or G-strands. Given the site-specific deposition of O -Man glycans by dedicated enzymes and the central role of EC domains in cadherins functions, we hypothesized that these PTMs may fine-tune cellular adhesion and otherwise contribute to diverse physical interactions that involve cadherins. To test these hypotheses, we assayed for changes in protein-protein interactions formed with epithelial (E)-cadherin in model cells where O -Man were genetically ablated. Herein, we report O -Man-dependent E-cadherin (CDH1) protein interactions, revealed by affinity proteomics, and we orthogonally validate an altered association between CDH1 and CDH3 (P-cadherin). We show different interactomic changes associated with O -Man ablation on B- vs. G-strands, highlighting the importance of these PTMs in CDH1-associated interaction. These findings provide new insights into how O -Man regulates CDH1-dependent protein complexes.

cell biology↗

bioRxiv2025

Quantifying spatiotemporal decoupling of GDGT-temperature relationships in a deep alpine lake

Glycerol dialkyl glycerol tetraethers (GDGTs), membrane lipids produced by archaea and some bacteria, are widely used in paleoclimate reconstructions due to their empirical relationship with temperature. However, their application in lakes is complicated by uncertainties in source attribution and environmental controls on their distribution. To address these constraints, we analyzed both branched and isoprenoid GDGTs (brGDGTs and isoGDGTs) in settling particles collected over 19 months using sediment traps at four depths (10, 15, 25 and 35 m) in Lake Lugu, a deep, stratified alpine lake in southwestern China. GDGT fluxes showed synchronous spatiotemporal variations across depths, with higher values during winter mixing than summer stratification, suggesting in situ production enhanced by nutrient upwelling during lake overturn. Correlations between GDGT distributions and high-resolution water temperature profiles revealed strong temperature sensitivity in isoGDGTs, particularly the Ring Index (RI), with peak correlations linked to mean temperatures [~]20 days prior to trap recovery, indicating a clear temporal lag in GDGT-temperature relationship. Moreover, stronger correlations with temperatures at overlying depths, implying vertical transport of isoGDGTs and a dominant autochthonous origin from the upper water column. In contrast, brGDGTs displayed weak or non-significant temperature dependence, likely reflecting distinct microbial sources or other controlling factors. These findings underscore the utility of isoGDGT-based proxies, particularly RI, while highlighting the importance of accounting for spatiotemporal offsets in GDGTs production when reconstructing paleotemperatures in deep, stratified lakes.

ecology↗