Biology subjects

Xu, F.

Publications and source records attributed to Xu, F..

At least 19 recordsLinked to original sources

bioRxiv2026

Bayesian in situ point spread function modeling enables high-fidelity three-dimensional nanoscopy in deep tissue

Three-dimensional single-molecule localization microscopy enables nanometer-resolution mapping of biomolecular organization in native cellular environments, but extending this capability to deep tissue requires in situ point spread function (PSF) modeling to account for specimen-induced optical distortions. Existing in situ approaches rely on a deterministic mapping between each single-molecule detection and the underlying PSF model. In deep tissue, however, specimen-induced aberrations, optical scattering and elevated background degrade the information content of single-molecule observations, posing a fundamental barrier to reliable PSF estimation. Here we introduce Bayesian in situ PSF modeling (BIM), a computational framework that explicitly models uncertainty in emitter axial position and propagates it through PSF reconstruction. By using an uncertainty-aware posterior fusion strategy, BIM retrieves accurate specimen-matched PSF models directly from raw blinking data under native imaging conditions. We demonstrate BIM by resolving hollow mitochondrial membranes, amyloid-{beta} fibrils, dendrites and spines with high fidelity. When combined with adaptive optics, BIM improves high-fidelity 3D reconstruction in deep tissue at depths exceeding one hundred micrometers.

biophysics↗

bioRxiv2026

Matrix stiffness shapes immunometabolic states and response to PD-1 blockade in gastric cancer

The extracellular matrix of solid tumors exhibits spatially heterogeneous stiffness, yet how matrix stiffness shapes antitumor immunity in gastric cancer (GC) remains unclear. Here, we integrated anti-PD-1 clinical cohorts, GC specimens, single-cell and spatial transcriptomic analyses, a syngeneic mouse model, and tunable-stiffness hydrogels to clarify the immunological consequences of matrix stiffness. High matrix stiffness was associated with collagen accumulation and immune exclusion, while higher stiffness-associated transcriptional scores were related to poorer response to PD-1 blockade. Matrix softening markedly reshaped the immune microenvironment and enhanced the efficacy of PD-1 blockade. Mechanistically, stiff matrices promoted F-actin remodeling and YAP nuclear enrichment, driving LDHA-dependent glycolytic reprogramming in GC cells that further impaired the effector function of CD8 T cells. Intriguingly, matrix softening was accompanied by enhanced IFN-{gamma}-responsive PD-L1 expression in GC cells, consistent with adaptive immune resistance that may contribute to the limited antitumor effects of matrix softening alone. These findings suggest that matrix remodeling alleviates spatial and metabolic constraints on antitumor immunity, providing a preclinical rationale for combining matrix softening with PD-1 blockade in GC.

cancer biology↗

bioRxiv2026

Mesaconate-Based Polyester Nanoparticles Reveal How Metabolite Isomerism Shapes Polymer Behavior, Macrophage Reprogramming, and Systemic Immunomodulation

Metabolite-based polymer nanoparticles can transform endogenous immunometabolites into intrinsically bioactive materials, yet most platforms have focused on itaconate or soluble metabolite analogs. Mesaconate is a downstream itaconate-derived immunometabolite with distinct metabolic activity, but its potential as an engineerable material building block remains largely unexplored. Here, mesaconic acid was used as a monomer to construct a degradable, cargo-free polyester nanoparticle platform. Poly(itaconate-co-decanediol) (pITA) and poly(mesaconate-co-decanediol) (pMSA) were synthesized using 1,10-decanediol and formulated into nanoparticles. Varying pMSA synthesis conditions generated distinct molecular-weight development, thermal organization, and degradation-associated behavior. Despite comparable particle size, surface charge, and morphology, pITA- and pMSA-derived nanoparticles produced activation-state-dependent macrophage immunomodulation. Under inflammatory stimulation, both nanoparticles modulated activation-associated surface molecules, whereas their transcriptomic responses diverged markedly. pITA-NPs preferentially engaged redox-, metabolic-, and cellular stress-associated programs, while pMSA-NPs exhibited a distinct lysosome-associated, inflammatory, and remodeling transcriptional profile and remained distinguishable from soluble 4-octyl itaconate. In LPS-induced endotoxemia, pMSA-NP treatment broadly modulated systemic cytokine responses, including a pronounced reduction in circulating IL-6, and improved seven-day survival relative to vehicle control. Together, these findings establish mesaconate as a synthesis-tunable polymeric immunometabolite and highlight metabolite-isomer identity and nanoparticle material state as complementary design parameters for cargo-free immunometabolic biomaterials.

bioengineering↗

bioRxiv2026

Imaging biological tissue at extended depth using swept-source visible-light OCT

We present a swept-source visible-light optical coherence tomography (SS-vis-OCT) system with twice the imaging depth of prior SS-vis-OCT and improved sensitivity for imaging biological samples, achieving shot-noise-limited operation. Because second-harmonic generation is limited by the booster optical amplifier's output power, the visible-light SS power is capped at 450 W. Increasing it further would require a more powerful, but costlier, amplifier. Instead, we reduce the receiver noise floor by selecting a balanced detector with NEP approaching the thermal (Johnson-Nyquist) limit. Real-time numerical k-linearization increases the effective number of linear-in-wavenumber samples per sweep, further improving SNR and extending the imaging range. Together, these improvements yield a sensitivity of 89 dB and an imaging depth of 9 mm in air or 6.5 mm in tissue, enabling full anterior-segment imaging of ex vivo porcine eyes.

bioengineering↗

bioRxiv2026

Cooperative Peripheral CB1 Inhibition by Extracellular Antibody Binding and Orthosteric Antagonism

The cannabinoid receptor CB1 is an important therapeutic target for obesity and metabolic disorders, yet centrally acting CB1 antagonists are limited by neuropsychiatric side effects. Although the anti-CB1 antibody Nimacimab has advanced to Phase 2 clinical trials for metabolic disorders, the molecular mechanisms underlying peripheral CB1 inhibition remain poorly understood. Here, we report cryo-EM structures of human CB1 bound to Nimacimab alone and in complex with the orthosteric antagonists Rimonabant and JD5037. Unlike previously characterized inhibitory GPCR antibodies that block ligand access through steric occlusion or directly engage orthosteric sites, Nimacimab functions through extracellular allosteric inhibition by stabilizing an inactive-like extracellular architecture of CB1. Antibody binding restricts extracellular conformational dynamics and allosterically stabilizes inactive-state signaling microswitches. Importantly, extracellular antibody engagement is structurally compatible with orthosteric antagonism and cooperatively enhances CB1 inhibition. Structural analysis further reveals how JD5037 remodels the extracellular vestibule to achieve peripheral-selective antagonism. Together, these findings establish extracellular allosteric inhibition as a distinct mechanism of GPCR regulation and provide a structural framework for the development of peripherally selective CB1 therapeutics.

pharmacology and toxicology↗

bioRxiv2026

Architecture and Energy Transfer of the Bacterial Photosynthetic Unit

In phototrophic organisms, pigment-protein membrane complexes are densely packed to form photosynthetic units (PSUs) that capture solar energy and convert it into chemical energy. Although the structures of many individual photosynthetic complexes have been resolved, how they are arranged and interact with others within photosynthetic membranes to enable efficient excitation energy transfer (EET) remains poorly understood. Here, we report cryo-electron microscopy structures of PSU supercomplex assemblies from the phototrophic a-proteobacterium Rhodovulum viride, including an RC-LH1 core associated with one or two peripheral LH2 complexes and a curved LH2 tetramer. These membrane-derived assemblies define the relative positions and orientations of neighboring photosynthetic complexes and place their pigment arrays in proximity across antenna-antenna and antenna-core interfaces. Structure-based simulations identify potential EET pathways within the PSU assemblies and reveal rapid energy transfer across both LH2-LH2 and LH2-LH1 interfaces. Collectively, these findings provide insights into the assembly and structural modularity of bacterial PSUs and elucidate how the lateral organization of membrane protein complexes facilitates efficient energy transfer. This work extends structural studies of bacterial photosynthesis from individual complexes to their native higher-order assembly, providing a framework for understanding how photosynthetic supercomplex organization shapes energy migration and for guiding the design of artificial photosynthesis.

biochemistry↗

bioRxiv2026

Sex-dimorphic immune trajectories across the canine lifespan and early-life signatures of geroprotective interventions

Canine models are invaluable for translational geroscience, but widespread gonadectomy in companion dogs obscures natural sex-specific aging trajectories. Here, we characterize the age-associated hematologic and serum cytokine profiles of 80 intact laboratory Beagles spanning 1 to 11 years. Contrary to classical linear models of immunosenescence and inflammaging, we reveal that canine immune aging is highly dynamic and sexually dimorphic. While absolute leukocyte counts declined continuously, cytokine remodeling was sex-specific: intact males exhibited extensive age-dependent inflammatory restructuring, whereas females demonstrated restricted fluctuations. To establish baseline geroprotective signatures, we evaluated 24 young beagles following a 90-day intervention with rapamycin, canagliflozin, or dietary restriction. Rapamycin induced robust immune modulation, while canagliflozin prompted early, sex-biased weight loss and narrower cytokine responses. Together, our findings demonstrate that canine immune aging involves complex, sexually dimorphic remodeling. By utilizing an intact cohort to isolate natural dimorphic trajectories, we provide a crucial foundation for developing precision, sex-optimized geroprotectors.

physiology↗

bioRxiv2026

Non-invasive forecasting of skin cancer evolution through longitudinal hair sampling

The ability to longitudinally track clonal evolution non-invasively would transform cancer interception strategies, long before late-stage disease when most cancer genomes are analysed. Here, we demonstrate that repeated hair sampling from the same individual followed by exome sequencing enables tracking of somatic evolution in vivo over several months after chemically induced skin carcinogenesis. We found that hair follicles accumulate a higher mutation burden than spatially-matched skin and harbour mutations that spread into surrounding epidermis and persist throughout tumour progression. DNA-damaged follicles enter sustained quiescence that delays replication and repair, creating a reservoir for long-lived mutations. During premalignant progression, carcinogen-associated mutations become enriched as follicular clones expand into adjacent skin. Mutation tracking identified genes that may govern tumour predisposition and initiation, many of which are mutated at high incidence in human cutaneous squamous cell carcinoma cohorts. Hair follicles therefore provide a non-invasive readout to forecast the early development of skin cancer, enabling patient risk stratification.

cell biology↗

bioRxiv2026

Closed-loop optical optimization enables patterned retinal stimulation in vivo at cellular scales

Visual neuroscience requires precise spatiotemporal projection of optical stimulation onto the retina, especially in experimental mouse models. However, in vivo patterned stimulation in mice is profoundly hindered by the extreme optical power and severe anatomical aberrations of the eye. Consequently, visual stimulation relies mainly on unverifiable, open-loop approximations that often lack spatial precision. Here, we introduce a closed-loop, spatially modulated stimulation platform that overcomes these barriers. By integrating a digital micromirror device (DMD) with electronically tunable lenses (ETLs) and a real-time, fundus camera-guided focus optimization module, we directly verify the location of patterned stimuli on the retina while dynamically correcting for chromatic and geometric defocus. This platform delivers quantitatively verified static and dynamic patterned stimuli to the living retina with lateral resolutions as fine as 6.7 {micro}m. Guided by ray-tracing optical analysis, our work establishes a technological foundation that enables highly reproducible, cellular-scale interrogations of the visual pathway.

bioengineering↗

bioRxiv2026

Spaceland: Histology-Guided Reconstruction of High-Resolution Whole-Organ 3D Molecular Atlases from Sparse Spatial Transcriptomics

Reconstructing whole organs in three-dimensional molecular detail is a key step toward building virtual organs for modeling tissue organization, disease progression and drug perturbation responses. However, high-resolution whole-organ spatial transcriptomic profiling remains impractical, forcing a trade-off between reconstruction fidelity and sampling density. Here, we introduce Spaceland, a morphology-guided framework that reconstructs continuous, high-resolution 3D molecular landscapes from sparsely sampled spatial transcriptomic sections and serial H&E histology. Spaceland formulates this task as learning continuous gene-expression fields within a morphology-informed histological space. It constructs a dense 3D morphological scaffold by optical-flow interpolation of foundation model-derived H&E representations and decodes sparse spot-level transcriptomic measurements onto an 8 m histology-aligned grid. Across mouse olfactory bulb, mouse hemibrain and spatiotemporal planarian regeneration, Spaceland generalized across platforms, tissue scales and biological contexts. In mouse benchmarks, Spaceland bridged 400 m molecular gaps, resolved sub-spot organization, outperformed ST-based interpolation and 2D H&E-based prediction methods, and remained robust with 160-320 m H&E intervals. In planarian regeneration, it enabled time-resolved whole-organism analysis from only four Visium sections per stage, revealing dynamic neoblast-neural spatial remodeling. Together, Spaceland shifts 3D molecular atlas construction from exhaustive experimental sampling toward data-driven virtual tissue and organ modeling, providing a scalable route to whole-organ molecular reconstruction.

bioinformatics↗

bioRxiv2026

Persistent Post-Inflammatory Matrix Stiffening Defines a Premalignant Mechanical Niche in Ulcerative Colitis-Associated Neoplasia

Patients with long-standing ulcerative colitis (UC) remain at risk for colorectal neoplasia even after overt inflammatory activity improves, suggesting that repaired mucosa may retain residual tissue-level abnormalities. Whether extracellular matrix remodeling leaves a persistent mechanical cue that contributes to early neoplastic remodeling is unknown. Here we identify post-inflammatory matrix stiffening as a premalignant mechanical niche in UC-associated neoplasia. In human colonic biopsies, collagen remodeling and atomic-force-microscopy-based mucosal stiffness increased from non-UC controls to UC and UC-associated dysplasia. A stiffness-associated transcriptional program was also enriched in histologically non-dysplastic mucosa from patients with UC-associated neoplasia. In mouse models, colonic stiffness increased along the colitis-to-tumorigenesis axis and remained elevated during apparent recovery, when epithelial permeability, junctional protein loss, crypt proliferation and nuclear {beta}-catenin accumulation persisted despite reduced disease activity. Stiffness-controlled intestinal epithelial cultures showed that stiff substrates were sufficient to disrupt ZO-1 organization and enhance {beta}-catenin redistribution, particularly under TNF- stimulation. Pharmacological matrix normalization with {beta}-aminopropionitrile partially restored barrier organization and attenuated early epithelial remodeling in vivo. Single-cell profiling identified a stiffness-associated MMP7-positive/YAP-active epithelial state that was enriched in collagen-rich regions and reduced after matrix softening. Inhibition of YAP or MMP7 attenuated stiffness-associated junctional disruption and {beta}-catenin redistribution. These findings suggest that inflammatory recovery and mechanical recovery can be uncoupled, and that persistent mucosal stiffening provides a tissue-level mechanism by which chronically injured UC mucosa may remain biologically vulnerable to premalignant epithelial remodeling.

Cancer Biology↗

bioRxiv2026

Preconception e-nicotine impairs airway development and progenitor proliferation across generations in Drosophila Melanogaster

Electronic nicotine delivery devices are increasingly used worldwide, raising concern about the potential effects of nicotine exposure during fetal and early postnatal development. However, evidence regarding the effects of vaping during pregnancy remains limited, and the consequences of exposure before conception are largely unknown. Here, we investigated whether maternal nicotine vaping prior to conception alters airway development using the fruit fly Drosophila melanogaster. Offspring of nicotine-exposed mothers exhibited abnormal airway architecture, epithelial remodelling, and impaired terminal branching. These defects were accompanied by reduced proliferation of airway progenitor cells without increased apoptosis, resulting in decreased hypoxia tolerance, altered activity patterns, and reduced lifespan. Transcriptomic and functional analyses revealed suppression of innate immune signalling mediated by nuclear factor kappa-B (NF-{kappa}B). Notably, knockdown of its regulator Relish reproduced key phenotypes. Developmental abnormalities persisted across generations, with their severity progressively declining. Together, these findings identify maternal preconception nicotine exposure as a previously unrecognized determinant of airway development and provide insight into early origins of chronic airway disease. TeaserPreconception e-nicotine exposure shapes airway development, structural and physiological abnormalities across generations in Drosophila.

Developmental Biology↗

bioRxiv2026

Agent-Guided De Novo Design of Nanobody Binders Against a Novel Cancer Target

Therapeutic antibody discovery remains slow and resource-intensive, with traditional methods providing limited control over epitope selection. We present a workflow for de novo nanobody design applied to a novel Desmoplastic Small Round Cell Tumor target encompassing four stages: (1) epitope identification guided by our hotspot recommendation agent using physical chemistry-based structure and sequence analysis tools with two curated databases (IEDB, PFAM), (2) de novo nanobody generation using three independent methods (RFantibody, IgGM, mBER) across multiple predicted antigen structures and nanobody frameworks, (3) multi-metric scoring including structural metrics from folding models, and in silico binding affinity from our sequence-based predictor, (4) high-throughput yeast surface display (YSD) screening followed by surface plasmon resonance (SPR) characterization of the specific binders. We generated 288,000 nanobody designs spanning eight target epitope regions and three variable domains of heavy chain-only antibody (VHH) frameworks. Multi-objective Pareto filtering with our candidate selection agent yielded 100,000 candidates for YSD screening with fluorescence-activated cell sorting (FACS). Of 116 enriched candidates advanced to SPR characterization, 46/116 (39.7%) produced reliable kinetic fits with Rmax [≥] 30 RU, yielding KD values from 0.66 nM to 305 nM (median 31.7 nM). These results show that an agent-guided computational workflow can design nanomolar to sub-nanomolar nanobody binders against a novel target without experimental structure or prior antibody information.

bioinformatics↗

bioRxiv2026

Assessment of the household antibiotic resistance gene, virulence factor genes, and pathogen profiles from three global cities

This study assessed the prevalence of antibiotic resistance genes (ARGs) and virulence factor genes (VFs), DNA viruses, and medically-relevant pathogens in three major cities around the globe - Mysuru (India), Dubai (United Arab Emirates), and Tucson (Arizona, United States of America). Ten households were sampled in each city, at ten sites in the bathroom, kitchen, and living spaces. The alpha diversity of ARGs significantly differed between household locations in each country (ANOVA, p<0.05) and beta diversity (dissimilarity) analysis showed a significant association between the ARGs and the geographic locations (PERMANOVA, p<0.01). A set of ARGs were found in every location across the dataset (the core ARG profile) included 25 different genes. The alpha diversity of virulence factors differed across the household locations within the three cities (ANOVA, p<0.01). The beta diversity of VFs was not well explained by geographic location or location within the household (PERMANOVA, p=0.129 (geographic), p=0.127 (household)). There were 341 unique VFs found in the study, but only 5 core VFs across the dataset. Bacterial pathogens detected across the household included Escherichia coli, Acinetobacter baumanii, Klebsiella pneumoniae, and more. The DNA (and bacteriophage) virome varied between countries and was more diverse in Tucson and Dubai (top viral families included Poxviridae and Orthoherpesviridae - two families which contain human pathogens - and Steitzviridae, a family of bacteriophages) compared to Mysuru, where nearly all viruses were a part of the Muvirus genus (a bacteriophage which contributes to horizontal gene transfer by phage transduction). ImportanceThe diversity of the built environment microbiome is not well characterized globally. Household occupants interact with the built microbiome on a daily basis, and the built microbiome may have the potential to influence human health. The presence of pathogens in the built environment and the key genes contributing to microorganism pathogenicity were investigated in this study, as information on this is lacking on an international scale. The diversity of ARG and VFs across the globe, as well as the presence of medically relevant pathogens within the house that were found in this study highlights the need to explore further the factors which influence the household microbiome, virome, and resistome. This may aid in identify how the build microbiome may be shaped by humans and influence human health. Impact StatementThis research contributes to the understanding of the built microbiome, specifically how it varies within the house, within cities, and across the globe. This can aid in our understanding of microbial dynamics in environments with heavy human influence and help develop and improve hygiene habits and products which are mindful of the existing microbiome.

microbiology↗

bioRxiv2026

A Conserved Geometric Code: Extracellular Matrix Curvature Directs Cell Migration Strategy via Nuclear Mechanosensing

Cells navigate complex tissue microenvironments defined by intricate physical cues, yet how they interpret the three-dimensional geometry of the extracellular matrix (ECM) remains an open question. Current models often fail to account for the tortuous architectures found in physiological tissues. Here, we demonstrate that ECM curvature functions as a tissue-specific geometric code read by the cell nucleus. By mapping collagen architectures across cancers and tissues, we find unique curvature fingerprints preserved during metastasis. Using micro-engineered substrates, we show that high curvature imposes localized nuclear bending stress, triggering a Lamin A/C-cPLA2-Ca2+ mechanotransductive cascade. This sensor rewires the cytoskeleton from longitudinal stress fibers to a cortical actomyosin network, driving a sharp transition from fast mesenchymal migration to a slower, exploratory amoeboid phenotype. We term this "nuclear curvotaxis", establishing a physical principle linking static geometry to dynamic strategy, with implications for predicting metastatic risk, understanding immune exclusion, and designing bio-instructive scaffolds for tissue engineering.

biophysics↗

bioRxiv2026

Systemic Cysteine Elevation Sustains T-Cell Activation to Potentiate PD-1 Blockade

Resistance to immune checkpoint inhibition remains a major barrier in pancreatic cancer treatment. Here, we show that concurrent administration of probiotics restores sensitivity to anti-PD-1 therapy in pancreatic cancer mouse models. Mice treated with the combination of anti-PD-1 and probiotics demonstrate robust tumor control, accompanied by enrichment of microbial pathways governing cysteine biosynthesis, elevated serum cysteine levels, and increased T cell function. Serum cysteine levels, rather than intratumoral cysteine concentrations, inversely correlate with tumor burden. Functionally, cysteine directly promotes T cell survival, activation, and cytotoxicity while its restriction induces uncoupled transcriptional-translational stress and impairs T cell function. Oral cysteine supplementation synergizes with anti-PD-1 therapy in pancreatic cancer mice, reducing tumor burden and enhancing intratumoral T cell activation, phenocopying probiotics-mediated immune restoration. These findings suggest systemic cysteine availability as a tractable metabolic target to enhance cancer immunotherapy.

cancer biology↗

bioRxiv2026

Long-Read Transcriptome Sequencing and Functional Validation Reveals Novel and Oncogenic Gene Fusions in Fusion Panel-Negative Gliomas

Gliomas comprise a heterogeneous group of central nervous system tumors in which gene fusions (GFs) are significant oncogenic drivers and emerging diagnostic and therapeutic biomarkers. In cancer diagnosis, GF detection largely relies on targeted short-read sequencing fusion panels, such as the Childrens Hospital of Philadelphia (CHOP) Fusion Panel (FUSIP). While these panels are effective for detecting recurrent, well-characterized GFs, they are limited to predefined gene sets and cannot identify full-length transcripts. Here, we analyzed 49 high-and low-grade gliomas previously classified as fusion-negative by FUSIP using an untargeted whole-transcriptome RNA sequencing approach with Oxford Nanopore Technologies (ONT) long-read sequencing. This enabled transcriptome-wide fusion discovery of additional known and potentially novel oncogenic GFs beyond panel constraints. Long-read sequencing further allowed direct resolution of full-length fusion transcripts and their associated isoform structures. By integrating GF detection with isoform-level transcript analysis, we identified fusion-associated transcript isoforms with alternative splicing patterns that aligned near reported GF breakpoints, including ZNF254::GNAS and PTPRK::NOX3, which have not been reported in literature or existing fusion databases. To assess functional relevance, candidate GFs were evaluated using the Drosophila melanogaster model, with ventral nerve cord (VNC) morphology serving as a quantitative in vivo readout of fusion-induced disruption of glial regulation. VNC enlargement or elongation reflects abnormal glial growth or defects in brain tissue organization. Of the 15 candidate GFs subjected to experimental functional testing, 8 induced significant VNC abnormalities relative to wild-type controls, indicating fusion-specific disruption and oncogenic potential. Notably, CLDND1::WRN and DUSP22::APOE produced the most pronounced VNC phenotypes. Together, these findings demonstrate that untargeted transcriptome-wide GF discovery, coupled with long-read isoform-level analysis and in vivo functional validation, enables the identification and prioritization of potentially novel and clinically relevant GFs that are missed by standard targeted short-read fusion panels in glioma.

cancer biology↗

bioRxiv2026

DeepSRFusion: a point cloud deep learning framework for super-resolution particle fusion

Deciphering the spatial organization of macromolecular complexes in their native context is central to structural biology. Particle fusion in single-molecule localization microscopy offers a unique capability for high-resolution structural reconstruction in situ. However, existing methods face significant challenges from large rotational perturbations and sparse labeling, resulting in compromised accuracy and substantial computational cost. We present DeepSRFusion, a self-supervised pretraining framework for three-dimensional super-resolution particle fusion. By representing single-molecule point clouds as Gaussian Mixture Models, DeepSRFusion integrates data-driven feature learning with physical imaging constraints. A two-stage optimization strategy with dynamic template updating enhances robustness, and a novel Clustering Error metric quantifies fusion quality. Nanometer-scale validation on both simulated and experimental datasets demonstrates high reconstruction fidelity and structural consistency with cryo-electron microscopy and AlphaFold3. DeepSRFusion remains effective under challenging imaging conditions, including large 3D rotations, sparse labeling, high localization uncertainty, and limited particle numbers, while achieving over 100-fold speedups compared to current methods. It resolves fine structural features with a measured spatial resolution of 1.60 {+/-} 0.10 nm, sufficient to distinguish ~10 nm spaced protein pairs and visualize tilted internal substructures within macromolecular assemblies. DeepSRFusion provides a powerful tool for high-precision structural analysis in native cellular environments.

biophysics↗