Biology subjects

Tang, S.

Publications and source records attributed to Tang, S..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Structure and inhibition of the sperm TMEM95-FIMP complex in mammalian fertilization

TMEM95 is a sperm acrosomal membrane protein essential for mammalian fertilization. How TMEM95 facilitates sperm-egg interaction has largely remained unknown. Analogous sperm fertilization proteins function as complexes, leading us to hypothesize that TMEM95 may have a binding partner on sperm. Here, we surveyed interactions between TMEM95 and individual proteins in a curated library of testis-expressed proteins using AlphaFold3. We identify FIMP, a fertilization-essential acrosomal membrane protein, as a high-confidence interaction partner of TMEM95. These two proteins form a high-affinity complex through their ectodomains. Using single-particle cryo-EM, we determine the structure of the human TMEM95-FIMP ectodomain complex at high resolution. An aromatic motif of FIMP binds to a conserved surface of TMEM95, and amino acid substitutions within this motif ablate the TMEM95-binding activity of FIMP. We isolate an anti-TMEM95 antibody, termed 3A02, that binds to human and murine TMEM95 and disrupts the interaction between TMEM95 and FIMP. By determining the cryo-EM structure of human TMEM95 bound to the 3A02 fragment antigen-binding region, we find that 3A02 recognizes the FIMP-binding site on TMEM95. 3A02 inhibits fusion of murine sperm with eggs, independent of antibody size, suggesting that the TMEM95-FIMP interface is critical for sperm-egg interaction. Together, these results establish the human sperm TMEM95-FIMP complex and suggest that a FIMP-mediated interaction of TMEM95 facilitates membrane fusion during mammalian fertilization.

biochemistry↗

bioRxiv2026

Microbiome-mediated lauric acid turnover contributes to colonization resistance against Ralstonia in the tomato rhizosphere

Host-associated lipids shape plant-microbiome interactions, but their microbiome-mediated turnover and role in pathogen exclusion are not fully understood. In tomato, lauric acid (C12:0) occurs across root-associated compartments, and its abundance at the root-soil interface is linked to soil-dependent differences in rhizosphere community composition and Ralstonia solanacearum colonization. Lauric acid promotes Ralstonia motility and virulence in vitro while inhibiting some beneficial lauric-acid-sensitive taxa, especially Gram-positive antagonists. In disease-suppressive soils, rhizosphere communities enriched in lauric-acid-degrading taxa are associated with lower local lauric acid levels and reduced pathogen colonization, suggesting community-dependent buffering of the rhizosphere lauric acid pool alongside other protective microbiome functions. Soil perturbation, microbiome transplantation, gnotobiotic SynCom reconstruction, and isotope tracing provide convergent evidence that microbiome-mediated lauric acid turnover contributes to colonization resistance and helps explain how rhizosphere lipid chemistry influences pathogen invasion outcomes in tomato, revealing an ecologically grounded protective mechanism for pathogen management.

microbiology↗

bioRxiv2026

Coordinated synthesis of double-stranded DNA by a dual reverse transcriptase immune system

Defense-associated reverse transcriptase (DRT) systems mediate antiviral immunity through distinct modes of cDNA synthesis: Class 1 DRTs catalyze untemplated synthesis, whereas Class 2 DRTs polymerize noncoding RNA-templated products. However, how these distinct modes drive defense remains unclear. Here, we report that DRT3 immunity arises when Class 1 and Class 2 RT activities cooperate to produce self-complementary double-stranded DNA (dsDNA). DRT3a uses a 5'-ACACAC-3' RNA template to synthesize poly-(dTdG) repeats, whereas DRT3b synthesizes poly-(dCdA) repeats without any nucleic acid template. Cryo-electron microscopy reveals that DRT3b forms a hexamer and uses active-site-adjacent residues as deoxyadenosine and deoxycytidine gates to enforce alternating nucleotide addition, representing a unique example of amino acid-templated DNA polymerization. DRT3 is toxic in cells lacking RecBCD, implicating host recombination machinery in limiting dsDNA accumulation, and the phage-encoded RecBCD inhibitor Gam triggers DRT3-mediated abortive infection. These findings reveal how two polymerases with distinct templating strategies generate complementary DNA for defense.

microbiology↗

bioRxiv2026

The cytoplasmic lattice in mammalian eggs sequesters ubiquitination machinery and tubulin in reserve

The cytoplasmic lattice (CPL) in mammalian eggs is essential for early embryonic development, but its molecular components, structural organization, and functional capacity have remained elusive. Here, using cryo-electron microscopy, we show that the CPL filament in mouse eggs contains repeating units with a periodicity of [~]37 nm, and determine its high-resolution, native structure and complete subunit composition. The CPL architecture organizes maternal-effect proteins, ubiquitination machinery, and tubulin into a highly structured reservoir. Maternal-effect proteins form the scaffold of the CPL to sequester a UBE2D-UHRF1 E2-E3 ubiquitination module and three distinct FBXW E3 ubiquitin ligases, notably all in activity-excluded states. The CPL further contains an {beta}-tubulin heterodimer in a GTP-bound state with a calcium ion coordinated to -tubulin, suggesting microtubule assembly-competent tubulin held in reserve. The CPL structure is capped at each end by a terminal unit that lacks a PADI6 dimer, a scaffold component, thereby preventing further oligomerization. Interactions between maternal-effect proteins in adjacent CPL units promote the assembly of a three-dimensional lattice in the egg cytoplasm. Taken together, our work defines how CPL assembly and architecture prime mammalian eggs for ubiquitin-mediated protein degradation and cytoskeletal remodeling during the egg-to-embryo transition.

biophysics↗

bioRxiv2026

Clarified an rDNA Gene Unit Pattern with (CTTT)n and (CT)n Microsatellites Aggregation Ahead of and Behind the Gene in Human Genome

BackgroundConventional models of human ribosomal DNA (rDNA) array organization have historically depended on transcription-centric boundaries, partitioning the unit into a [~]13 kb rDNA transcription region and a monolithic [~]31 kb intergenic spacer (IGS). While our previous identification of Duplication Segment Units (DSUs) mapped these arrays based on an intuitive analysis of the microsatellite density landscape of the complete reference human genome, our present deep mining of this landscape has revealed a more accurate rDNA Gene Unit Pattern. Methods & ResultsIn this study, we conducted a deep mining analysis of our previously established microsatellite density landscape of the T2T-CHM13 assembly, focusing specifically on nucleolar organizing regions (NORs). We suggest a more accurate rDNA Gene Unit Pattern containing a (CTTT)n microsatellite aggregation ahead of the rDNA gene and a (CT)n microsatellite aggregation behind the gene, rather than a pattern featuring an IGS region inserted between two rDNA genes. ConclusionsA correct rDNA gene pattern of the human genome probably includes a (CTTT)n microsatellite aggregation ahead of the gene and a (CT)n microsatellite aggregation behind it, which possibly constitute cis- and trans-regulating regions; the (CTTT)n and (CT)n microsatellite aggregations may provide two different local stable DNA structures for regulatory protein binding.

genetics↗

bioRxiv2026

CodonRL: Multi-Objective Codon Sequence Optimization Using Demonstration-Guided Reinforcement Learning

Optimizing synonymous codon sequences to improve translation efficiency, RNA stability, and compositional properties is challenging because the search space grows exponentially with protein length and objectives interact through long range RNA structure. Dynamic programming-based methods can provide strong solutions for fixed objective combinations but are difficult to extend to additional constraints. Deep generative models require large-scale, high-quality mRNA sequence datasets for training, limiting applicability when such data are scarce. Reinforcement learning naturally handles sequential decision-making but faces challenges in codon optimization due to delayed rewards, large action spaces, and expensive structural evaluation. We present CodonRL, a reinforcement learning framework that learns a structural prior for mRNA design from efficient folding feedback and demonstration-guided replay, and then enables user-controlled multi-objective trade-offs during inference. CodonRL uses LinearFold for fast intermediate reward computation during training and ViennaRNA for final evaluation, warms up learning with expert sequences to accelerate convergence for global structure objectives, and introduces milestone-based intermediate rewards to address delayed feedback in long range optimization. On a benchmark of 55 human proteins, CodonRL outperforms GEMORNA, a state-of-the-art codon optimization method, across multiple metrics, achieving 9.5% higher codon adaptation index (CAI), 25.4 kcal/mol more favorable minimum free energy (MFE), and 3.4% lower uridine content on average, while improving codon stabilization coefficient (CSC) in over 90% of benchmark proteins under matched constraints. These gains translate into designs that are predicted to be more efficiently translated, more structurally stable, and less immunogenic, while supporting continuous objective reweighting at inference time.

bioinformatics↗

bioRxiv2026

Orion Microbiome Database: mapping constellations in the microbiome universe

BackgroundPublicly available microbiome sequencing data have grown rapidly in scale and diversity, but secondary use remains limited by fragmentation across repositories, inconsistent annotations, heterogeneous processing pipelines, and incomplete metadata. These barriers complicate cross-study comparisons and restrict the reproducibility of large-scale microbiome research. ObjectiveThe Orion Microbiome Database was developed to overcome these challenges by providing a standardized, accessible, and reproducible resource for the integration and analysis of whole-genome shotgun (WGS) microbiome datasets. MethodsOrion aggregates raw metagenomic datasets from public repositories and harmonizes them into computed microbial profiles using automated and transparent analysis pipelines. Metadata are curated to ensure consistency and enable systematic cross-cohort comparisons. The platform provides interactive tools for data exploration, visualization, and downstream analysis, reducing redundancies in data preprocessing and lowering the barrier for both novice and expert users. Beyond research applications, Orion also serves as an educational resource, enabling microbiome data exploration in classroom settings. ConclusionBy unifying fragmented sequencing resources into a reproducible framework, Orion accelerates microbiome discovery, supports scalable cross-study investigations, and fosters integration of microbiome science into teaching and training. This database has the potential to serve as a model for future large-scale, community-driven platforms in microbiome research.

microbiology↗

bioRxiv2026

Notch mediates non-regenerative alveolar repair after staphylococcal lung injury

Severe lung infection causes dysfunction of the lungs air-blood barrier, leading to respiratory failure. In alveoli of lungs infected with Staphylococcus aureus (SA) - either alone or after respiratory viral infection - the SA toxin, alpha hemolysin (Hla) causes epithelial barrier protein degradation and airspace edema formation. How the barrier repairs is not clear. We used confocal imaging of intact, perfused, SA-infected lungs to define barrier repair mechanisms in live alveoli. Though we expected to find the non-surviving alveolar epithelium was regenerated, we found, instead, the surviving alveolar epithelium spontaneously regained barrier function. Thus, SA stimulated Notch protein cleavage in the alveolar epithelium in an Hla- and ADAM10-dependent manner. Subsequent exposure of the Notch transmembrane domain catalyzed epithelial junctional protein recovery to reseal the barrier and restore barrier integrity. While disrupting Notch cleavage in the alveolar epithelium prolonged SA-induced lung injury, augmenting it accelerated lung repair. We interpret that barrier repair in the surviving alveolar epithelium resulted from Notch-mediated junctional protein reassembly. These findings show, for the first time, that the alveolar epithelium is a resilient tissue that harbors robust endogenous repair mechanisms. We propose strategies that leverage Notchs non-regenerative, barrier-strengthening capability may promote lung repair after staphylococcal lung infection.

molecular biology↗

bioRxiv2026

Human fallopian tube-on-a-chip for preclinical testing of non-hormonal contraceptives with living human sperm

The fallopian tube serves as a sperm reservoir, and it is the site where the oocytes become fertilized. Here, we describe development of an organ-on-a-chip microfluidic model of the fallopian tube (FT Chip) lined by primary human epithelial cells and stromal fibroblasts derived from the FT ampulla. Abundant tissue folds lined by hormone-responsive, epithelial cells resembling those seen in vivo formed on-chip, but not in epithelial organoids cultured in gel cultures. Comparative time-resolved analysis of human sperm versus oocyte-sized microparticles introduced into the epithelial channel in the presence of estradiol revealed that sperm movement was significantly reduced, while the oocyte-sized particles increased, relative to movements in acellular chips. When the non-hormonal contraceptive TDI-11861 was administered to the chip, dose-dependent inhibition of human sperm motility was detected. Thus, this FT Chip may offer a human preclinical tool to study FT physiology and assess the efficacy and mechanism of action of contraceptives.

cell biology↗

bioRxiv2026

EpiTADformer: A Transformer-Based Model for High-Resolution TAD Boundary Detection Using Epigenomic Signal Embeddings

The human genome is partitioned at different levels of 3D genome organization, with topologically associating domains (TADs) being among the most well-known and biologically important structures. TAD boundary disruption is associated with a wide range of diseases such as cancer, neurological and developmental disorders. Numerous methods have been developed to detect TAD boundaries from chromatin contact maps obtained with Hi-C technology. However, these methods are largely limited by the resolution of Hi-C data, typically 1 Kb to 100 Kb. In contrast, functional DNA loci, collectively referred to as epigenomic data, are profiled at a much higher resolution (100-200 bp for a typical ChIP-seq experiment). To improve the resolution of boundary detection, we hypothesize that the patterns of epigenomic signals associated with regions in proximity to TAD boundaries can serve as embeddings for these genomic regions, defining region similarity. These embeddings, along with their positional relationships, can be effectively modeled using deep learning to achieve more precise boundary prediction. We present EpiTADformer, a transformer-based model that takes as input transcriptional and histone modification signals of neighboring regions centered around TAD boundaries. We demonstrate that EpiTADformer outperforms feedforward neural network, convolutional neural network (CNN), and bidirectional long short-term memory (BiLSTM) network architectures. These results suggest the positional information of epigenomic signals surrounding TAD boundaries provides a strong predictive signal, enabling improved performance of the transformer model. Our findings highlight the potential of epigenomic signals to serve as region embeddings for refining the epigenomic language of TAD domains and 3D genome organization.

bioinformatics↗

bioRxiv2026

PeptiVerse: A Unified Platform for Therapeutic Peptide Property Prediction

Therapeutic peptides combine the advantages of small molecules and antibodies, offering target flexibility and low immunogenicity, yet their successful translation requires careful evaluation of multiple developability properties beyond binding alone. As chemically modified peptides become increasingly common in drug design, no unified platform currently supports systematic property assessment across both canonical sequences and SMILES-based representations. Leveraging the generalizability of large foundational models trained on protein and chemical data, we introduce PeptiVerse, a universal therapeutic peptide property prediction platform. PeptiVerse accepts either amino acid sequences or chemically modified peptide SMILES, delivers state-of-the-art performance across diverse property prediction tasks, and provides both a web interface and open-source implementation for rapid, accessible, and scalable pep-tide developability analysis. By unifying property prediction across representations, PeptiVerse directly supports early-stage peptide therapeutic development campaigns and property-aware generative design workflows.

bioinformatics↗

bioRxiv2026

Use of Metagenomic Reference Materials to Compare Lysis and Extraction Methods Against a Known Input

Inconsistent cellular lysis is a major source of inaccuracy in microbial analysis; there is a need to develop objective and standardized evaluation of lysis efficiency across diverse methods. Using a mock microbial community as a ground truth, we built a bioinformatic and experimental toolkit for this purpose, comparing the performance of chemical, thermal, enzymatic, and various physical lysis protocols. We evaluated over 150 lysis conditions and observed the highest performance from mechanically- and materially-optimized bead-based approaches. To extend these findings to a practical setting, we demonstrated that the Firmicutes to Bacteroides ratio, a frequently used compositional metric in gut microbiome analysis is biased heavily by insufficient lysis, indicating the importance of benchmarking ongoing methodological development.

microbiology↗

bioRxiv2025

SIMAIS: A swarm intelligence inspired biosensor for rapid protein detection

Swarm intelligence, the collective behaviour of animals such as birds, fish, and ants, arises from simple local interactions and enables survival, efficient foraging, and decision-making without central control. Inspired by this principle, we aimed to harness swarm intelligence for biosensing applications. Conventional assays often struggle with limited sensitivity, long processing times, and the need for specialised instruments, restricting their use in decentralised healthcare. Here we developed a swarm intelligence of microbead-inspired, artificial intelligence (AI)-assisted, and smartphone-based (SIMAIS) biosensing platform that can transform invisible molecular recognition into visible, macroscale patterns. Millions of antibody-coated magnetic microbeads (MBs) organise into magnetically induced structures upon binding protein biomarkers, with the resulting patterns correlating directly to biomarker concentration. Using interferon-gamma (IFN-{gamma}) as a model biomarker, we demonstrate that this approach delivers results within 10 minutes and achieves a tenfold increase in sensitivity compared to conventional lateral flow assays. Robust performance in human serum and urine confirms clinical applicability. By mimicking natural swarm behaviour, this platform introduces a new biosensing paradigm that bridges biology and engineering. Its versatility, speed, and ease of use highlight its potential for decentralised diagnostics, early disease detection, and broader applications in point-of-care healthcare.

biophysics↗

bioRxiv2025

Antiviral reverse transcriptases reveal the evolutionary origin of telomerase

Defense-associated reverse transcriptases (DRTs) employ diverse and distinctive mechanisms of cDNA synthesis to protect bacteria against viral infection. However, much of DRT family diversity remains unstudied. Here we identify a new antiviral defense system, DRT10, that associates with a non-coding RNA (ncRNA) to catalyze processive, protein-primed synthesis of tandem-repeat DNA. Repeat addition is dictated by sequence and structural features of the ncRNA that have direct parallels in the RNA component of telomerase. Remarkably, a phylogenetic analysis of RTs across domains of life reveals an unexpected link between DRT10 and telomerase that is further supported by structural comparisons and mechanistic evidence. These findings expand the repertoire of reverse transcription mechanisms in antiviral defense and point to a bacterial origin for telomerase. One-Sentence SummaryInsights from antiviral defense systems reveal an unexpected bacterial origin for the mechanism of chromosome maintenance by eukaryotic telomerase.

molecular biology↗

bioRxiv2025

Fork Reversal Safeguards Epigenetic Inheritance during DNA Replication Under Stress

During DNA replication, epigenetic information carried by histone modifications is faithfully propagated and re-established on sister chromatids, ensuring cell identity. Chromatin reassembly is tightly coupled to DNA replication, therefore whether and how perturbations to DNA replication affects the fidelity of epigenetic inheritance remain poorly understood. In this study, we reveal a critical role for replication fork reversal in maintaining the transmission of epigenetic information under replication stress. We identify that cells defective in fork reversal exhibit reduced nucleosome density at replication forks, accompanied by the loss of parental histones during their transfer onto nascent DNA. Mechanistically, we demonstrate that PrimPol activation leads to single-stranded DNA gaps in fork reversal deficient cells, and that subsequent PARylation (poly ADP-ribosylation) and DNA-protein crosslinking on these gaps evicts nucleosomes. Our findings demonstrate that replication fork reversal, a widespread physiological process, is not only essential for preserving genome integrity but also for safeguarding epigenetic stability.

biochemistry↗

bioRxiv2025

ARCADE: Controllable Codon Design from Foundation Models via Activation Engineering

Codon sequence design is crucial for generating mRNA sequences with desired functional properties for tasks such as developing mRNA vaccines or gene editing therapies. Yet existing methods lack flexibility and controllability to adapt to various design objectives. We propose a novel machine learning-based framework, ARCADE, that enables flexible and controllable multi-objective codon design. Leveraging inherent knowledge from pretrained genomic language models, ARCADE extends activation engineering, a technique originally developed for controllable text generation, beyond discrete feature manipulation such as concepts and styles, to steering continuous-valued biological metrics. Specifically, we derive biologically meaningful semantic steering vectors in the models activation space, which directly control properties such as the Codon Adaptation Index, Minimum Free Energy, and GC content. Experimental results demonstrate the flexibility of ARCADE in designing codon sequences with multiple objectives, underscoring its potential for advancing programmable biological sequence design. Our implementation is available at https://github.com/Kingsford-Group/arcade.

bioinformatics↗

bioRxiv2025

Multi-omics reveals lipid metabolism, mitochondrial and extracellular matrix dysregulation of diabetic cardiomyopathy in the human heart

Diabetic cardiomyopathy (DbCM) is a condition characterized by myocardial dysfunction in diabetes. Although the clinical recognition of DbCM is increasing, its underlying molecular mechanisms remain poorly understood. We aimed to conduct a comprehensive multi-omics analysis of heart tissue from cardiomyopathy patients (both with and without diabetes),alongside tissue from healthy donors, by integrating transcriptomic, 4D-DIA proteomic, and full-spectrum widely targeted metabolomic data. Differential analyses revealed a significant metabolic pattern in patients with DbCM, characterized by increasing utilization of triglyceride-derived fatty acids. Correlation analyses highlighted impaired BNIP3L-mediated mitophagy as a potential contributor to the disruption of fatty acid metabolism and mitochondrial dysfunction in DbCM. Protein interaction and fibrotic trichrome analyses showed that patients with DbCM had impaired cardiac matrix repair compared to patients with cardiomyopathy without diabetes. These findings provide a system-level understanding of DbCM and reveal molecular signatures driving its progression, offering insights for targeted therapeutic strategies against diabetic heart disease.

bioinformatics↗

bioRxiv2025

Neural Correlates of Trial Outcome Monitoring during Long-term Learning in Primate Posterior Parietal Cortex

Monitoring behavioral outcomes is crucial for optimizing behavior and learning new tasks. However, how the brain monitors outcome information during long-term learning remains largely unknown. Using two-photon calcium imaging in behaving monkeys, we tracked neuronal activity in area 7a of the posterior parietal cortex (PPC), an important associative cortex that has not been typically implicated in associative learning (AL), while the macaque monkeys learned new visual-motor associations over multiple days. We found robust neuronal representation of salient behavioral outcomes (correct vs. incorrect) which closely correlated with the monkeys learning behavior. Furthermore, outcome representation significantly reorganized when monkeys transitioned to learn novel associations, followed by gradual evolution over subsequent learning days, a process constrained by the functional connectivity among neurons within local network. These suggest a substantive role for primate PPC in long-term AL through monitoring trial outcome, and indicate a principle for long-term learning: network connectivity constrains the evolution of neural encodings.

neuroscience↗