Biology subjects

You, L.

Publications and source records attributed to You, L..

At least 19 recordsLinked to original sources

bioRxiv2026

Overcoming segmentation bottlenecks in brain tissue for spatial proteomic inference at population scale

In spatial proteomics across large cohorts, downstream estimates from tens of thousands of images depend on accurate cell segmentation. Yet pretrained segmentation models struggle with morphologically complex cell bodies in human brain tissue, and how domain adaptation affects downstream discovery remains unclear. Here, we introduce SPInfer, a workflow for spatial proteomic inference developed using 5,832 manually annotated cell bodies from 120 immunofluorescence images of the human prefrontal cortex. After benchmarking eight segmentation methods spanning classical, convolutional, and foundation models, we adapted Cellpose-SAM, raising cell segmentation F1 from 0.37 to 0.84 and recovering cell compositions concordant with matched single-nucleus RNA-seq and spatial transcriptomics. Across 14,765 images from 69 postmortem donors, identical Alzheimer's disease pathology models found no significant associations after false discovery rate correction using zero-shot masks, versus four using fine-tuned masks; three linked higher OLIG2-positive fractions to greater neurofibrillary tangle and plaque burden. Using the fine-tuned masks, a graph attention network over nuclear neighborhoods classified four cell types without marker information, and differential expression analysis linked cell morphology to gene expression. This provides a pipeline for fine-tuning foundation models to segment complex tissues, enabling biological discovery in spatial proteomics at the population scale.

bioinformatics↗

bioRxiv2026

Behavior-coupled single-cell transcriptomics reveals trafficking protein-dependent regulation of tumor cell migration

Aggressive cellular migration underlies cancer invasion and metastatic dissemination, yet the molecular programs defining highly migratory tumor cell states remain poorly characterized. In head and neck squamous cell carcinoma (HNSCC), metastatic progression is associated with poor clinical outcome, motivating method developments that link migratory behavior to underlying mechanisms. Here, we present behavior-coupled single-cell sequencing (BC-seq), a live-cell imaging-guided transcriptomic platform that enables real-time identification, isolation, and profiling of individual cells based on migratory behavior. Applying BC-seq to 10 HPV-negative oral squamous cell carcinoma (OSCC), the most common HNSCC subtype, we distinguished fast-migrating cells (FMCs) from slow-migrating cells (SMCs) and identified transcriptional programs associated with aggressive migration. Mechanistically, BC-seq uncovered cholesterol-regulated signaling pathways and associated trafficking proteins involving SREBP2, STARD4, Cavin1/3, and VEGFR2 that promotes actin cytoskeletal remodeling and cellular motility. Disruption of STARD4-mediated cholesterol trafficking or Cavin1/3-dependent VEGFR2 nuclear translocation significantly impaired cell migration, linking cholesterol metabolism to migratory machinery. We next examined whether molecular programs associated with aggressive migration are reflected in clinical outcomes. A 15-gene expression signature derived from FMCs predicted distant metastasis with 83.6% accuracy in an independent OSCC cohort, outperforming conventional clinicopathological risk stratification. Together, our study establishes BC-seq as an approach for integrating dynamic cellular behavior with single-cell transcriptomics to uncover mechanisms of aggressive migration and clinically relevant molecular signatures.

cancer biology↗

bioRxiv2026

A public-data reproducibility benchmark of a 2.5-D U-Net with a slice-level topology term for type-B aortic dissection segmentation

Background: Type-B aortic dissection requires separation of the true and false lumina in heterogeneous, anisotropic CTA. A topology term may reduce fragmentation, but its reproducibility must be tested under a fixed case split and independent native-grid evaluation. Methods: We used the public Graz type-B aortic dissection CTA collection (40 cases; CC BY 4.0) and froze a 28/6/6 case split for training, development, and held-out testing. A compact 2.5-D U-Net trained with cross-entropy plus multiclass Dice was compared with the identical model augmented by a fixed-weight (0.5) differentiable slice-level soft clDice term. Two training seeds were run for each arm. All 40 cases passed the NRRD audit; predictions for six held-out cases were restored to their native grids and scored with case-level 3-D voxel-grid Dice, voxel-grid morphological clDice, 6-connected component counts, physical-volume error, and 2,000-resample case-level paired bootstrap intervals. Results: With seed 20261007, topology-minus-baseline differences were -0.028 (95% CI -0.053 to -0.004) for true-lumen Dice, +0.032 (-0.006 to +0.096) for true-lumen voxel-grid clDice, +0.071 (+0.037 to +0.103) for false-lumen Dice, and +0.073 (+0.048 to +0.103) for false-lumen voxel-grid clDice. With seed 20261008, the corresponding differences were -0.040 (-0.065 to -0.018), +0.013 (-0.034 to +0.071), -0.099 (-0.187 to -0.004), and -0.058 (-0.124 to -0.001). Descriptive pooled means over 12 case rows per arm were true-lumen Dice 0.577 for topology versus 0.611 for baseline and false-lumen Dice 0.409 versus 0.422; these pooled rows are not independent cases. Conclusions: Under this fixed six-case test split, the corrected slice-level topology term did not produce a consistent benefit. True-lumen Dice was lower in both seeds, while false-lumen effects reversed direction between seeds; topology-clDice changes were small and/or interval-crossing. The benchmark supports an auditable failure-analysis workflow but does not support a general topology benefit, clinical use, external generalization, or a panvascular decision system.

bioinformatics↗

bioRxiv2026

Panvascular-JEPA: Unseen-Structure Evaluation of Factorized Predictive States for Hemodynamic Networks

Predictive representations for vascular simulation should be tested on unseen network structures and compared with the governing equations before being interpreted as general-purpose world models. We constructed a new, dimensioned but uncalibrated nodal resistance-inertance-compliance benchmark containing all 38 connected, non-isomorphic six-node graphs with five to seven edges. Twenty-four structures supplied training, validation, and new-background testing; 14 structures were wholly withheld. Across 648 parameter backgrounds and two prescribed action programs, 1,296 trajectories contained 96 forecast steps over three cycles. Four graph predictors received equal six candidate optimization searches, followed by five fixed-seed fits. Selected configurations and analysis code were locked before test trajectories were generated. Comparators included known-equation numerical integrators, an inertia-free reduction, and a five-coefficient model fitted to training trajectories. Unseen-structure mean standardized RMSE was 8.46 x 109 for four-factor OPF, 329 for single factor OPF, and 1.83 x 1031 for supervised graph prediction, compared with 1.09 x 10-4 for foursubstep known-equation integration and 6.77 x 104 for the fitted mechanism. The primary OPF-4 minus OPF-1 difference was 8.46 x 109, with a paired stratified-graph 95% percentile interval of [2.23 x 105, 2.54 x 1010]. Neural errors were strongly dependent on training seed and network structure, with extreme forecasts dominating the means. All four neural means exceeded persistence; relative performance within this comparison did not establish a usable forecasting model. The study supplies a distinct numerical test of structure transfer, factor-count control, and recursive stability. Its conclusions concern the implemented models, optimization budget, and finite graph catalogue. It establishes neither patient-specific performance nor anatomical transfer, biological remodeling, device efficacy, or clinical readiness.

bioengineering↗

bioRxiv2026

AC-Diff: Anatomy-Contrast Disentangled Diffusion for Multi-Vendor Liver MRI Harmonization

Magnetic resonance imaging (MRI) exhibits substantial scanner and protocol dependent appearance variations, making harmonization challenging without distorting patient specific anatomy, particularly in abdominal imaging. We propose AC-Diff, an anatomy--contrast disentangled diffusion framework that formulates MRI harmonization as a factor-specific generative intervention. Using aligned multi-contrast supervision and cross-patient factor swapping, AC-Diff learns to separate anatomical structure from acquisition-dependent contrast. Unlike conventional image or latent diffusion, AC-Diff restricts stochastic generation to the contrast subspace while the source anatomy bypasses diffusion and is directly reused during reconstruction. Experiments on an in-house multi-vendor cohort and the external Duke Liver MRI dataset demonstrate improved target-domain alignment with strong structural preservation. In downstream liver segmentation, AC-Diff improves Dice from 0.942 for the original inputs to 0.954 for the harmonized images. These results support contrast-specific latent generation as a promising approach to anatomy-preserving MRI harmonization.

bioengineering↗

bioRxiv2026

PanVasc Research for AI assisted evidence analysis in panvascular intervention

Panvascular intervention research requires evidence workflows that preserve source identity, outcome definitions and observation windows. We developed PanVasc Research, an executable research framework, and evaluated a fixed local Qwen3-4B model using two complementary tasks. Fifty ClinicalTrials.gov records from five vascular query strata generated 200 source-fidelity tests with intact evidence or controlled removal of the requested source, primary outcome or timeframe, plus 50 clean controls. A separate 100-statement sample from the official NLI4CT test set assessed clinical-trial entailment and evidence selection against existing expert labels. Generic and checklist prompts used identical evidence and maximum generation budgets; each output was also evaluated with an input-only deterministic contract. Registry exact accuracy was 51/200 (25.5%) with the generic prompt and 77/200 (38.5%) with the checklist; intact-case agreement was 50/50 and 48/50. The rule baseline recovered 200/200 tuples. NLI label accuracy was 48/100 (48.0%) and 50/100 (50.0%), respectively. The source contract retained 35 and 31 incorrect NLI labels in the two arms. Registry references establish fidelity to a registration snapshot, while NLI4CT concerns breast-cancer trials and does not validate vascular expertise. The framework also retains provenance-recorded literature retrieval, structured research planning and local numerical analysis. These experiments support a bounded assessment of source handling and semantic failure, rather than a new foundation model or autonomous scientific discovery. A proposed endpoint representation identifies the additional domain annotation and independent validation required for a panvascular research model.

bioinformatics↗

bioRxiv2026

Gene circuit-driven amplified selection enables evolution of fast-growing Escherichia coli

The laboratory Escherichia coli K-12 strain has doubled no faster than [~]20 minutes for decades. This plateau could reflect a biophysical limit or simply the way batch culture selects on growth rate. Here we show it can be broken through amplified selection with a Red Queen gene circuit, which takes advantage of growth rate heterogeneity in monoclonal populations to selectively suppress slow-growing cells and creates a tunable mapping from intrinsic growth rate to survival. After 70 days ([~]1,000 generations) of amplified selection in MG1655+FHr and subsequent removal of the circuit, a top evolved clone (RQ70) reached a maximum specific growth rate of 2.61 h-{superscript 1} in shake-flask culture. This corresponds to a doubling time of 15.9 minutes, to our knowledge the shortest reported for E. coli K-12, against 18.1 minutes for evolved controls and 20.3 minutes for the ancestor. The gain came at the cost of a [~]3-fold increase in lag time, indicating that the 20-minute plateau is a multi-trait optimum under conventional batch selection rather than an absolute constraint. We argue that synthetic gene circuits can therefore reshape the evolutionary process itself, pushing performance beyond apparent physiological limits.

synthetic biology↗

bioRxiv2026

Piezo1 links cytoskeletal remodeling to differential YAP and β-catenin signaling in response to mechanical cues

Mechanical cues from the extracellular matrix regulate cancer cell behavior, but how these inputs are translated into distinct nuclear signaling responses remains incompletely understood. This study examined whether Piezo1, a mechanosensitive ion channel, contributes to cytoskeletal remodeling and differential regulation of YAP and {beta}-catenin localization in breast cancer cells exposed to defined mechanical cues. MDA-MB-231 breast cancer cells were cultured on substrates of defined stiffness and analyzed after Piezo1 knockdown or pharmacological modulation of Piezo1, Src signaling, myosin II activity, and actin polymerization. Nuclear localization of YAP and {beta}-catenin was assessed by immunofluorescence imaging, cytoskeletal organization was evaluated using filamentous and globular actin staining, protein phosphorylation was analyzed by capillary electrophoresis-based immunoblotting, and cell migration was assessed using a wound-healing assay. Piezo1 knockdown reduced YAP nuclear localization and increased {beta}-catenin nuclear localization, while Piezo1 activation partially reversed these localization changes. Piezo1 knockdown also disrupted filamentous actin organization, and pharmacological disruption of actin polymerization produced similar effects on YAP and {beta}-catenin localization. Piezo1 knockdown selectively reduced YAP tyrosine phosphorylation without altering canonical Hippo-associated YAP serine phosphorylation, and inhibition of Src signaling produced effects similar to Piezo1 knockdown. Functionally, Piezo1 knockdown impaired stiffness-dependent cell migration. These findings support a role for Piezo1 in linking extracellular mechanical cues to cytoskeletal organization and differential regulation of YAP and {beta}-catenin localization in breast cancer cells. This work provides a framework for understanding how mechanosensitive ion-channel signaling may contribute to context-dependent nuclear signaling responses during cancer cell mechanotransduction.

cell biology↗

bioRxiv2026

Anatomy-Guided 3D Graph Networks for Couinaud Segmentation in Tumor Affected Livers

Image-based liver Couinaud segmentation is designed to automatically provide the locations of suspicious objects in liver CT/MR images. Once achieved, the physicians will be guided to the target slice and area where the suspicious node is located. However, conventional algorithms trained primarily on healthy liver images often fail to generalize to Hepatocellular Carcinoma (HCC) cases due to pathological structural distortions. In this work, we propose a robust two-stage framework that integrates a 3D Unet with a 3D Anatomical Structure-Guided Graph Convolutional Network (3D GCN). This two-stage strategy effectively isolates the liver volume to eliminate structural noise from neighboring organs, such as the spleen, allowing the framework to focus exclusively on the complex 3D anatomical relationships among the eight segments. To ensure the topological consistency required for global spatial reasoning, we implement a standardized preprocessing pipeline that normalizes liver-only volumes to exactly 50 frames along the z-axis. By combining a lightweight 3D UNet backbone with the 3D GCN for refined boundary reasoning, our model demonstrates superior generalization performance on unseen clinical datasets, achieving a mean Dice score of 0.828 in blind testing. By releasing our code and pretrained weights, we aim to provide the first publicly available deep learning resource for robust Couinaud segmentation.

bioinformatics↗

bioRxiv2026

Microscale spatial fragmentation promotes bacterial survival under antibiotics

Most bacteria inhabit fragmented microscale environments rather than well-mixed systems in which antibiotic efficacy is typically defined. Here, we show that microscale spatial fragmentation alone can deterministically promote bacterial survival under antibiotic exposure. A coarse-grained theoretical model predicts that, at fixed bulk cell density and antibiotic concentration, fragmentation generates refuges through two coupled mechanisms: slower growth, which reduces susceptibility to growth-dependent antibiotics, and a finite antibiotic-per-cell constraint that lowers the effective intracellular antibiotic concentration. Using a microdroplet platform spanning five orders of magnitude in volume, we experimentally confirm these predictions for {beta}-lactams; we further demonstrate that the same fragmentation-mediated protection extends to antibiotics with distinct modes of action. Spatial fragmentation thus creates robust, non-genetic antibiotic refuges governed by physical constraints rather than resistance or collective protection, with implications for bacterial persistence in natural and clinical environments and for antibiotic design.

microbiology↗

bioRxiv2025

Simulation-templated photorealistic prediction of bacterial patterns

Pattern formation underlies biological organization across scales, but predicting experimentally observed patterns remains difficult because mechanistic models and data-driven generative models fail in complementary ways. Coarse-grained mechanistic models can encode causal constraints and global morphology, yet they omit fine-scale features such as texture, color gradients, and stochastic replicate-to-replicate variation. In contrast, generative image models can produce realistic images but are not inherently grounded in the biophysical rules that shape real patterns. Here, we introduce a simulation-conditioned generative framework that uses mechanistic simulations as spatial priors for generating biologically realistic pattern data. As a concrete test, we use a synthetic-to-real inverse task to show that these generated patterns preserve information needed for inference on real experimental images, beyond merely reproducing plausible visual appearance. Using branching colony expansion of Pseudomonas aeruginosa as a model system, we combine a coarse-grained PDE model with latent representations from a foundation image model and a conditional diffusion model. The resulting framework preserves the global structures imposed by simulation while restoring experimentally observed fine-scale morphology and stochastic variability. A model trained exclusively on simulation-conditioned synthetic patterns transfers without fine-tuning to real experimental patterns, enabling inference of initial seeding configurations from experimental colony morphology. Together, these results establish simulation-conditioned generative modeling as a strategy for converting coarse mechanistic models into scientifically structured synthetic data, enabling inference tasks on real biological patterns where experimental data are scarce.

systems biology↗

bioRxiv2025

scProtoTransformer: Scalable Reference Mapping Across Molecules, Cells and Donors

The rapid accumulation of single-cell data has made it possible to comprehensively characterize biological systems at molecular, cellular, and donor levels. However, scalable reference mapping across different resolutions remains a major challenge in current research. Here, we propose scProtoTransformer, a prototype-based Transformer architecture designed to achieve scalable reference mapping across molecular, cell, and donor levels. scProtoTransformer introduces a knowledge-guided prototype tokenizer that projects gene expression into biologically interpretable pathway prototypes, effectively reducing numerical batch effects while preserving biological semantic patterns. Furthermore, by leveraging knowledge distilled from the foundation model and a dynamic supervised fine-tuning strategy, scProtoTransformer achieves robust biological representations with reduced pre-training requirements. Benchmark experiments across molecular, cell, and donor-level reference mapping demonstrate that scProtoTransformer delivers competitive or even superior performance compared with state-of-the-art approaches, while providing interpretability through biologically prototypes. Together, these results establish scProtoTransformer as a unified framework for scalable reference mapping, laying the foundation for systematic understanding from genes to individuals.

bioinformatics↗

bioRxiv2025

A foundation model for microbial growth dynamics

Microbial growth dynamics contain rich information about microbial populations, which support applications from antibiotic testing to microbiome engineering. However, the high dimensionality of growth data and the scarcity of large, task-specific datasets have limited generalizable modeling analysis across systems. Here, we develop a foundation model for microbial growth dynamics. It is a large-scale, self-supervised representation model trained on [~]370,000 experimental and simulated growth curves spanning diverse microbial species, environmental conditions, and community contexts. The model learns lower-dimensional latent embeddings that capture essential dynamical features of raw growth data and enable accurate reconstruction of these data. The concise representations enhance predictive performance in diverse downstream applications. Using these embedding, we achieve few-shot learning for antibiotic classification and concentration prediction, accurate forecasting of simulated and experimental communities, and inference of total abundance from relative-abundance data. By extracting transferable representations from heterogeneous datasets, our model provides a general framework for analyzing and predicting microbial community dynamics from limited measurements.

synthetic biology↗

bioRxiv2025

Dynamical memory underlies prolonged plasmid persistence after transient antibiotic treatment

Plasmids play critical roles in spreading and maintaining antimicrobial resistance (AMR). They often exhibit prolonged persistence upon antibiotic treatment, even when they impose substantial burden on their hosts. This persistence has been primarily attributed to rapid horizontal transfer or low plasmid cost. However, these mechanisms cannot account for the slow decay of burdensome plasmids with poor mobility. Here, we show that the decoupling of time scales between slow segregation loss and fast growth competition leads to a slow-down in plasmid abundance decay at high initial plasmid abundance, reminiscent of the ghost effect from nonlinear dynamical systems. Integrating theory, simulations, and quantitative experiments across clonal populations and multi-species bacterial communities, we demonstrate that a transient antibiotic pulse can eliminate plasmid-free cells and create a ghost state that extends plasmid persistence from days to months. Our research reveals a generalizable mechanism for the prolonged ecological memory of antibiotic exposure and underscores the need for proactive strategies to curb the spread of AMR.

microbiology↗

bioRxiv2025

Integrating theory and machine learning to reveal determinants of plasmid copy number

Plasmids are extrachromosomal mobile genetic elements whose copy numbers (PCNs) critically influence microbial evolution, antibiotic resistance and pathogenicity. Despite their importance and immense diversity, the ecological, evolutionary and molecular factors determining PCN remain poorly understood. Here, we present a theoretical model to explain the empirical power-law relationship between plasmid size and copy number, one of the fundamental quantitative principles governing PCN control. However, this relationship alone has limited predictive power. To improve PCN prediction, we introduce a data-driven approach incorporating diverse features. Trained on >10,000 plasmids, our machine learning model achieves significantly enhanced accuracy, with plasmid-encoded protein domains emerging as key predictors. Applying this framework, we conduct the first comprehensive analysis of PCN distributions across hundreds of thousands of metagenomic plasmids (IMG/PR database) and tens of thousands of clinical isolates, uncovering niche specific taxonomic PCN hotspots and ecological adaptations. These results provide critical insights into plasmid ecology, ARG surveillance and shed lights on the gut plasmidome, a "dark matter" in human microbiome.

microbiology↗

bioRxiv2025

Emergence of population-level feedback control by transposon-plasmid coevolution

The origins of adaptive functions remain poorly understood, despite considerable interest1-5. Here, we report the de novo evolution of population-level feedback control in clonal Escherichia coli strains containing high-copy plasmids, enabling these strains to express green fluorescent protein (GFP) in response to tetracycline. Selection maintains tetA+ and tetA- plasmids within single cells, due to negative feedback from the toxic effects of a tetA-gfp tetracycline resistance transposon. At high plasmid copy numbers, the intracellular equilibrium of tetA+ and tetA- plasmids robustly responds to tetracycline through rapid evolutionary dynamics6-8. Theory predicts that the GFP response to tetracycline is determined by the covariance between GFP expression and bacterial fitness. Our findings show that small mutational changes can result in large qualitative changes to population-level behavior. Here, the evolution of polymorphic intracellular populations of mobile genetic elements allows host populations to dynamically respond to antibiotic in the environment.

systems biology↗

bioRxiv2025

Role of tankyrase scaffolding in the β-catenin destruction complex and WNT signaling

Aberrant WNT/{beta}-catenin signaling drives tumorigenesis and metastasis in cancer. Although enzymatic inhibitors of tankyrase (TNKS) effectively block AXIN degradation and stabilize the {beta}-catenin destruction complex (DC), they have demonstrated limited efficacy in various cancer models. Here we demonstrate that, unexpectedly, the induction of AXIN puncta represents a major barrier to achieving therapeutic efficacy. Mechanistically, catalytic inhibition of TNKS prevents TNKS turnover and drives its accumulation in the DC, wherein the scaffolding function of TNKS induces AXIN puncta formation, rigidifies the DC, and impedes {beta}-catenin turnover. Chemically induced degradation of TNKS overcomes this limitation by stabilizing AXIN without puncta formation, providing a deeper suppression of the WNT/{beta}-catenin pathway activity and the proliferation of colorectal cancer cells harboring dysfunctional APC mutations. Collectively, these findings provide an explanation for the unsatisfactory outcomes of drugging the WNT/{beta}-catenin signaling pathway by TNKS inhibitors and highlight TNKS degradation as a promising approach to treat WNT/{beta}-catenin-driven cancers.

cancer biology↗

bioRxiv2025

Uncovering genetic architecture of the heart via genetic association studies of unsupervised deep learning derived endophenotypes

Recent genome-wide association studies (GWAS) have effectively linked genetic variants to quantitative traits derived from time-series cardiac magnetic resonance imaging, revealing insights into cardiac morphology and function. Deep learning approach generally requires extensive supervised training on manually annotated data. In this study, we developed a novel framework using a 3D U-architecture autoencoder (cineMAE) to learn deep image phenotypes from cardiac magnetic resonance (CMR) imaging for genetic discovery, focusing on long-axis two-chamber and four-chamber views. We trained a masked autoencoder to develop Unsupervised Derived Image Phenotypes for heart (Heart-UDIPs). These representations were found to be informative to indicate various heart-specific phenotypes (e.g., left ventricular hypertrophy) and diseases (e.g., hypertrophic cardiomyopathy). GWAS on Heart UDIP identified 323 lead SNP and 628 SNP-prioritized genes, which exceeded previous methods. The genes identified by method described herein, exhibited significant associations with cardiac function and showed substantial enrichment in pathways related to cardiac disorders. These results underscore the utility of our Heart-UDIP approach in enhancing the discovery potential for genetic associations, without the need for clinically defined phenotypes or manual annotations.

bioinformatics↗