Biology subjects

Xia, T.

Publications and source records attributed to Xia, T..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

DynPerturb:Dynamic Perturbation Modeling for Spatiotemporal Single-Cell Systems

Perturbation responses in multicellular systems are inherently spatiotemporal, governed by temporal dynamics, perturbation intensity, and tissue context. However, existing approaches rarely capture explicit spatiotemporal perturbation dynamics, limiting the ability to reconstruct how perturbations propagate over time, spread across space, and induce feedback at multiple biological scales. Here we introduce DynPerturb, a dynamic inference framework for single-cell and spatial transcriptomics that systematically models the "when-how strong-where" dimensions of perturbations. DynPerturb incorporates temporal encoding, spatial adjacency, and memory modules to capture nonlinear signal propagation and cross-regional feedback with mechanistic interpretability. Across temporal benchmarking datasets, DynPerturb consistently outperformed existing methods in regulatory link prediction, demonstrated zero-shot generalization to unseen nodes, and preserved accuracy under varying sparsity conditions and temporal resolutions. Applied to representative biological systems, DynPerturb uncovers dynamic regulatory rewiring in kidney disease, highlighting a mid-age sensitivity window and potential reversibility with early intervention; delineated a temporal fate-switching boundary between megakaryocyte-erythroid progenitors (MEPs) and granulocyte-monocyte progenitors (GMPs) during hematopoiesis; and revealed spatiotemporal interchamber signaling in murine heart development, where Igf2 perturbations induced spatial feedback shaping chamber proportioning. Collectively, these results establish DynPerturb as a unified and versatile system for dissecting dynamic perturbation effects and guiding precision intervention strategies in multicellular systems.

bioinformatics↗

bioRxiv2025

Various Optimization Strategies for the Isolation of Mitochondria from Sprague-Dawley Rat Liver Tissue

Mitochondria are key organelles that establish a very large free energy drop for the ATP hydrolysis reaction in the cytoplasm of cells, which provides energy required for the cell maintenance of homeostasis. They are commonly isolated from living tissue as it makes them easier to interrogate at the biochemical level. Thus, isolated mitochondria of sufficient quality are desired to improve translational impact. To gain a greater insight into the isolation process and optimize our current isolation protocol for quality, we implemented various fine-tuning modifications to our standard homogenization, centrifugation, and purification steps with isolated rat hepatocyte mitochondria. The following modifications we tested were: i) different homogenization speeds (10,000, 14,000, and 18,000 rpm) with varying time intervals (10, 20, and 30 sec), ii) addition of an additional purification spin, and iii) use of density gradients to further purify the isolated mitochondria from non-mitochondrial contaminants. Mitochondrial quality was approximated using the well-established respiratory control ratio (RCR). The data reveal that our original protocol yields isolated mitochondria with acceptable quality and the optimization attempts produced similar or worse mitochondrial isolates. In addition, an extra purification spin decreased RCR values and therefore is not recommended. We note that the use of density gradients did not improve the RCR, but it did remove presumed peroxisomal contamination. While this protocol can be further enhanced using additional metrics, the data indicate that our current isolation protocol is sufficiently effective since additional modifications did not yield major improvements.

biochemistry↗

bioRxiv2025

Mid-Infrared Photothermal Imaging of Fatty Acid Desaturation Reaction in Cancer Cells

Direct visualization of metabolic conversions within living systems is essential for understanding metabolic activities yet challenging due to the absence of reaction-specific reporters and the limited sensitivity of current imaging modalities. Herein, we report an approach to monitor fatty acids (FAs) desaturation, primarily catalyzed by stearoyl-CoA desaturase, in cancer cells using deuterium (D)-labeled palmitic acid (PA-d31) as the reaction-specific reporter and mid-infrared photothermal (MIP) microscopy as the bond-selective imaging modality. The desaturation of PA-d31 produced a peak at 2246 cm-1 in the cell-silent region, corresponding to the stretching vibration of unsaturated C-D bonds (D-C=C-D) in unsaturated fatty acids. Penalized least squares fitting was employed to remove water background for enhancing the visibility of this peak. Our study revealed heterogeneous spatial distributions of both saturated FAs and their desaturated metabolites within lipid droplet pools in cancer cells. Furthermore, we observed an increase in fatty acid unsaturation level in OVCAR5 cells under cisplatin-induced stress. By directly visualizing fatty acid desaturation, this study offers new insights into fatty acid metabolism and opens avenues for evaluating new therapeutic strategies targeting fatty acid metabolism.

biochemistry↗

bioRxiv2025

UniCell: Towards a Unified Solution for Cell Annotation, Nomenclature Harmonization, Atlas Construction in Single-Cell Transcriptomics

Standardizing cell type annotations across single-cell RNA-seq datasets remains a major challenge due to inconsistencies in nomenclature, variation in annotation granularity, and the presence of rare or previously unseen populations. We present UniCell, a hierarchical annotation framework that combines Cell Ontology structure with transcriptomic data for scalable, interpretable, and ontology-aware cell identity inference. UniCell leverages a multi-task architecture that jointly optimizes local and global classifiers, yielding coherent predictions across multiple levels of the ontology-defined hierarchy. When benchmarked across 20 human and mouse datasets, UniCell consistently outperformed state-of-the-art tools, including CellTypist, scANVI, OnClass, and SingleR, in annotation performance, and sensitivity to low-abundance populations. In disease settings, UniCell effectively identified previously unseen cell types through confidence-guided novelty detection. Applied to 45 human and 23 mouse tissue atlases, UniCell enabled cross-dataset and cross-species harmonization by embedding cells into a unified latent space aligned with Cell Ontology structure. Moreover, when used to supervise single-cell foundation models, UniCell substantially improved downstream annotation accuracy, rare cell detection, and hierarchical consistency. Together, these results establish UniCell as a generalizable framework that supports high-resolution annotation, nomenclature standardization, and atlas-level integration, providing a scalable and biologically grounded solution for single-cell transcriptomic analysis across diverse biological systems.

bioinformatics↗

bioRxiv2025

High-resolution reconstruction of cell-type specific transcriptional regulatory processes from bulk sequencing samples

Biological systems exhibit remarkable heterogeneity, characterized by intricate interplay among diverse cell types. Resolving the regulatory processes of specific cell types is crucial for delineating developmental mechanisms and disease etiologies. While single-cell sequencing methods such as scRNA-seq and scATAC-seq have revolutionized our understanding of individual cellular functions, adapting bulk genome-wide assays to achieve single-cell resolution of other genomic features remains a significant technical challenge. Here, we introduce Deep-learning-based DEconvolution of Tissue profiles with Accurate Interpretation of Locus-specific Signals (DeepDETAILS), a novel quasi-supervised framework to reconstruct cell-type-specific genomic signals with base-pair precision. DeepDETAILS core innovation lies in its ability to perform cross-modality deconvolution using scATAC-seq reference libraries for other bulk datasets, benefiting from the affordability and availability of scATAC-seq data. DeepDETAILS enables high-resolution mapping of genomic signals across diverse cell types, with great versatility for various omics datasets, including nascent transcript sequencing (such as PRO-cap and PRO-seq) and ChIP-seq for chromatin modifications. Our results demonstrate that DeepDETAILS significantly outperformed traditional statistical deconvolution methods. Using DeepDETAILS, we developed a comprehensive compendium of high-resolution nascent transcription and histone modification signals across 39 diverse human tissues and 86 distinct cell types. Furthermore, we applied our compendium to fine-map risk variants associated with Primary Sclerosing Cholangitis (PSC), a progressive cholestatic liver disorder, and revealed a potential etiology of the disease. Our tool and compendium provide invaluable insights into cellular complexity, opening new avenues for studying biological processes in various contexts.

bioinformatics↗

bioRxiv2025

How sleeping minds decide: state-specific reconfigurations of lexical decision-making

Decision-making is a core cognitive function that enables adaptive behavior across diverse contexts. While extensively studied in wakefulness, its persistence and reconfiguration across sleep states remain poorly understood. Here, we use computational modeling to examine lexical decision-making across wakefulness, N1 sleep, and lucid REM sleep in both healthy participants (HP) and participants with narcolepsy (NP). Using facial electromyography (EMG) to capture real-time behavioral responses to spoken words and pseudowords during sleep, we quantify how decision-making strategies adapt under different sleep and consciousness states. Our findings reveal two key insights. First, decision-making mechanisms are dynamically reconfigured across sleep states. In N1 sleep, the advantage for word (vs. pseudoword) judgments is supported by faster sensory encoding and motor preparation, combined with efficient evidence accumulation. In contrast, in lucid REM sleep, the word advantage is driven exclusively by enhanced evidence accumulation, while sensory encoding and motor preparation remain unchanged. Second, cross-state comparisons reveal distinct patterns of preservation and impairment. In N1 sleep, word judgment remains largely intact, whereas pseudoword judgment is significantly impaired, characterized by prolonged stimulus encoding, delayed motor preparation, and reduced evidence accumulation. In contrast, lucid REM sleep is marked by a global reduction in processing efficiency, reflected in slower evidence accumulation and elevated decision thresholds for both words and pseudowords. These results demonstrate that rather than being uniformly degraded, decision-making is dynamically reconfigured across sleep stages, reflecting adaptive neurocognitive strategies that sustain cognition in altered states of consciousness. By identifying state-specific computational mechanisms, this study provides new insights into the brains resilience and flexibility under changing cognitive and physiological conditions.

neuroscience↗

bioRxiv2025

RegFormer: A Single-Cell Foundation Model Powered by Gene Regulatory Hierarchies

Single-cell RNA sequencing (scRNA-seq) enables high-resolution profiling of cellular diversity, but current computational models often fail to incorporate regulatory priors, handle data sparsity, or efficiently process long gene sequences. Here, we present RegFormer, a foundation model that integrates gene regulatory networks (GRNs) with Mamba-based state-space modeling, overcoming the scalability and context-length limitations of Transformer architectures. RegFormer encodes each gene through dual embeddings, a value embedding for quantitative expression and a token embedding for regulatory identity, organized within a GRN-guided gene order to capture both expression dynamics and hierarchical regulation. Pretrained on 26 million human single cells spanning 45 tissues and diverse biological contexts, RegFormer achieves superior scalability and biological fidelity. Across comprehensive benchmarks, it consistently outperforms state-of-the-art single-cell foundation models (scGPT, Geneformer, scFoundation, and scBERT), delivering higher clustering accuracy, improved batch integration, and more precise cell type annotation. RegFormer also reconstructs biologically coherent GRNs, accurately models transcriptional responses to genetic perturbations, and enhances drug response prediction across cancer cell lines. By combining regulatory priors with efficient long-sequence Mamba modeling, RegFormer establishes a biologically grounded and scalable framework for single-cell representation learning, enabling deeper mechanistic insight into gene regulation and cellular state transitions.

bioinformatics↗

bioRxiv2024

BioLLM: A Standardized Framework for Integrating and Benchmarking Single-Cell Foundation Models

The application and evaluation of single cell foundational models (scFMs) present significant challenges stemming from the heterogeneity of architectural frameworks and coding standards. To address these issues, we introduce BioLLM, a framework facilitating the integration and application of scFMs in single-cell RNA sequencing data analysis. BioLLM provides a universal interface, bridging diverse scFMs into a seamless ecosystem. By mitigating architectural disparities and coding conventions, it empowers researchers with streamlined access to scFMs. With standardized APIs and comprehensive documentation, BioLLM streamlines model switching and comparative analyses, while incorporating best practices for consistent model evaluation. Our comprehensive evaluation of scFMs revealed distinct strengths and limitations, highlighting scGPTs robust performance across all tasks, both in zero-shot and fine-tuning scenarios. Geneformer and scFoundation also demonstrated strong capabilities in gene-level tasks, benefiting from effective pretraining strategies. In contrast, scBERT underperformed relative to other models, likely attributable to its considerably smaller parameter count and the limited size of the training dataset. Ultimately, BioLLM aims to empower the scientific community to leverage the full potential of foundational models, advancing our understanding of complex biological systems through enhanced single-cell analysis.

bioinformatics↗

bioRxiv2024

Three-dimension transcriptomics maps of whole mouse embryo during organogenesis

Understanding mammalian development heavily relies on classical animal models like the house mouse (Mus musculus). Advanced spatial transcriptomics has enabled biologists to break new ground in studies of molecular dynamics and cellular patterning during embryonic development with spatiotemporal resolution. To construct a comprehensive developmental trajectory, current three-dimensional (3D) spatial transcriptomic profiling leverages mouse embryos from gastrulation (E5.5) to organogenesis (E13.5) in continuity. However, a crucial phase for early organogenesis between E9.5 and E11.5 was deficient. To unveil the mystery of this stage, we present the 3D transcriptomics of mouse embryos at E9.5 and E11.5, with a widely applicable reconstruction workflow that bypasses sophisticated bioinformatic calculations. As our 3D atlas is generated at single-cell resolution, we demonstrate how organogenetic processes can be interpreted at different levels of granularity, from local cellular interactions to whole embryonic regionalization. We release the open-access database MOSTA3D (Mouse Organogenesis Spatiotemporal Transcriptomic Atlas in Three-dimensional) and hope a broader community will contribute to extending this framework from conception to senility in the near future.

developmental biology↗

bioRxiv2024

Slow-wave sleep and REM sleep differentially contribute to memory representational transformation

Memory transforms over time, gradually becoming less idiosyncratic and more gist-like. While sleep contributes to memory transformation, how different sleep stages and EEG activity influence memory transformation is far from clear. Applying representational similarity analysis to electroencephalogram (EEG) recordings, we examined memory representational transformation at both the idiosyncratic "item-level" and the generic "category-level". Our findings revealed that after an overnight sleep, item-level neural representations for post-sleep remembered items were abolished. In contrast, category-level representations remained prominent, but they became distinctive from pre-sleep. Across participants, more rapid eye movement (REM) sleep relative to slow-wave sleep (SWS) was associated with reduced item-level neural representational strength, increased category-level representational strength, as well as the decreased item-level representational similarity between pre-sleep learning and post-sleep retrieval sessions. Moreover, the theta and beta EEG power during REM sleep, and delta power during SWS differentially supported these representational transformations. These findings suggest that post-learning REM sleep and SWS play differential roles in supporting overnight memory transformation.

neuroscience↗

bioRxiv2024

Empowering Integrative and Collaborative Exploration of Single-Cell and Spatial Multimodal Data with SGS

Recent advancements in single-cell and spatial omics technologies have revolutionized our ability to capture multiple modalities data at a genome-wide scale within individual cells. However, visualizing these large-scale, high-dimensional, and complex datasets poses significant challenges. Here, we present SGS, a user-friendly, collaborative and versatile browser for visualizing single-cell and spatial multiomics data. SGS incorporates a novel genome browser framework, flexible visualization modules and a multi-panel adaptive communication mechanism to enable the synchronous visualization of diverse datasets. Notably, SGS empowers users with advanced capabilities for comparative visualization, through features like scCompare, scMultiView, and dual-chromosome module. Additionally, by adopting the "workstation concept", SGS enables data fast visualization and collaborative exploration. We showcase the potential of SGS in the comparative visualization and coordinated exploration of mutlimodal data with two examples. SGS is publicly available at https://sgs.bioinfotoolkits.net/home.

bioinformatics↗

bioRxiv2024

Spatiotemporally resolved transcriptome atlas of developing mouse placenta reveals mechanisms of embryonic lethality.

Proper placentation is fundamental to the growth and viability of the maturing embryo. However, a thorough grasp of the spatial organization of cell types, their interactions, and gene expression patterns along the maternal-fetal function during placental development remains incomplete. Here, we utilized Stereo-seq to construct the spatiotemporal transcriptomic atlas of the mouse placenta (STAMP) spanning embryonic (E) days 9.5 to 18.5 at single cell resolution (https://db.cngb.org/stomics/stamp/). This resource delineates spatially resolved cellular dynamics and gene expression patterns across placental and maternal compartments. We identified distinct glycogen trophoblast cell (GC) subclusters, mapped their developmental trajectories, and uncovered transcriptional transitions accompanying their migration from the junctional zone to the maternal decidua from E12.5 onward. In a defective placentation model with perinatal lethality, GCs abnormally persisted at E18.5 with excessive glycogen and reduced degradation metabolites in placenta and fetal liver, indicating impaired glycogen breakdown. Maternal glucose supplementation restored glucose levels and rescued fetal survival, underscoring GC-mediated metabolic support as critical for viability. Together, this study provides a comprehensive spatiotemporal placental atlas and demonstrates that GC-mediated glycogen metabolism is essential for sustaining fetal viability.

developmental biology↗

bioRxiv2024

CD81+ fibroblasts, a unique subpopulation with accelerated cellular senescence, exaggerate inflammation and activate neutrophils via C3/C3aR1 axis in periodontitis

Periodontitis, a prevalent inflammatory disease worldwide, poses a significant economic burden on society and the country. Previous research has established a connection between cellular senescence and periodontitis. However, the role and mechanism of cell senescence in the progression of periodontitis have not been thoroughly investigated. This study aimed to explore the involvement of cellular senescence in the pathogenesis of periodontitis and determine the underlying mechanisms. Our findings demonstrated that senescent cells accumulated during the progress of periodontitis. Moreover, several scRNA-seq analysis suggested that gingival fibroblasts were the main cell population undergoing cellular senescence during periodontitis, which helps mitigate tissue damage and bone loss. Furthermore, we identified a high expression of CD81 in the senescent gingival fibroblast population. These cells were found to actively contribute to inflammation through their potent pro-inflammatory metabolic activity and secretion of SASP-related factors. Additionally, they recruited neutrophils via the C3/C3aR1 pathway, indirectly sustaining the inflammatory response. Senolytics via Navitoclax successfully alleviated inflammation and bone loss in periodontitis and administration of metformin could alleviate alleviated inflammation and bone loss in periodontitis through inhibiting cellular senescence. These results provide valuable insights into the cellular and molecular basis of periodontitis-induced tissue damage, highlighting the significance of fibroblast senescence. In conclusion, our study sheds light on the relationship between CD81 and cellular senescence, suggesting its potential as a therapeutic target for periodontitis.

immunology↗

bioRxiv2024

Hydrophobic bulk conservation and genetic code determine the pattern of prokaryotic genome organization

Prokaryotic genomes in general exhibit marked organizational asymmetry, and differ substantially in GC-skew, AT-skew, and GS-bias (gene strand bias). Despite enigmatic origins, these organizational features and genomic GC-content have been described to be widely associated with each other recently, providing an opportunity to probe the evolutionary mechanisms. By analyzing sequence data of 4,012 leading strands from all available representative genomes, we found that the unusual nucleotide usage of coding genes under low GC-content contributed to a shift between two organizational patterns in the context of mutational biases. Analysis of artificially established neutral and natural models further suggests that in genes both GC-skew and AT-skew increase with decreasing GC-content, mostly contributed by synonymous substitutions because of genetic code, and by nonsynonymous substitutions because of hydrophobic bulk conservation in amino acid usage, respectively. This novel mechanistic framework in our study highlights the importance of evolutionary processes "operating at lower levels of organization" in the microbial world.

evolutionary biology↗

bioRxiv2023

Potential clinical value and influence of conductivity in conductive cardiac patches for reducing post-MI arrythmia risks

Conductive cardiac patches can help to restore electric signal conduction of the diseased myocardium after myocardial infarction (MI). However, none of the conductive cardiac patches reported in literature has entered clinical trials. Bench-to-bedside translation of conductive patches has long been hindered by the lack of knowledge of the optimal patch conductivity and deep understanding of the potential clinical benefits and risks in patients. Here, we first prepared conductive cardiac patches with conductivities covering 5 orders of magnitude (10-3-101 S/cm). Disagreeing with the mainstream opinion that patch conductivity close to native myocardium (10-3-10-2 S/cm) is most favorable, our results showed that patches with conductivity two orders of magnitudes higher than native myocardium (10-1-100 S/cm) are most effective in restoring cardiac conduction and lowering inducibility quotient. Conduction velocity (CV) is the essence of the observed results. Rat experiments showed that the low-conductivity patch could increase the CV of infarcted myocardium, but did not fully compensate the mismatch in the CVs of infarct and healthy myocardium. Moderate-conductivity patches could increase myocardial CV to the same level of healthy myocardium, while high-conductivity patches further increased myocardial CV, causing a reversed mismatch. The relationship between patch conductivity and improved CVs in myocardium can be explained by monodomain model theory. Based on the theory, 3D finite element simulation of a MI patient heart predicted that a suitable, patch-improved myocardial CV could reduce the number of reentrants, and stabilize the remaining reentry circuits in the myocardium of the MI patient, which indicated its clinical value.

bioengineering↗

bioRxiv2023

Modulating social learning-induced evaluation updating during human sleep

People often change their evaluations upon learning about their peers evaluations, i.e., social learning. Given sleeps vital role in consolidating daytime experiences, sleep may facilitate social learning and thereby further changing peoples evaluations. Combining a social learning task and the sleep-based targeted memory reactivation technique, we asked whether social learning-induced evaluation changes can be modulated during sleep. After participants indicated their initial evaluation for snacks, they learned about their peers evaluation while hearing the snacks spoken names. During the post-learning non-rapid-eye-movement sleep, we re-played half of the snack names (i.e., cued snack) to reactivate the associated peers evaluations. Upon waking up, we found that the social learning-induced evaluation changes further enlarged for both cued and uncued snacks. Examining sleep electroencephalogram (EEG) activity revealed that cue-elicited delta-theta EEG power and the overnight N2 sleep spindle density predicted post-sleep evaluation changes for cued but not for uncued snacks. Our findings suggested that sleep-mediated memory reactivation processes could strengthen social learning-induced evaluation changes.

neuroscience↗

bioRxiv2023

ST-GEARS: Advancing 3D Downstream Research through Accurate Spatial Information Recovery

Three-dimensional Spatial Transcriptomics has revolutionized our understanding of tissue regionalization, organogenesis, and development. However, to reconstruct single sections back to their in situ three-dimensional morphology, existing approaches either only adopt gene expression information to guide reconstruction or overlook shape correction against experiment-induced section distortions. This leads to significant discrepancies between reconstruction results and the actual in vivo locations of cells, imposing unreliable spatial profiles to downstream analysis. To address these challenges, we propose ST-GEARS (Spatial Transcriptomics GEospatial profile recovery system through AnchoRS), which solves optimized anchors between in situ closest spots utilizing expression and structural similarity across sections and recovers in vivo spatial information under the guidance of anchors. By employing innovative Distributive Constraints into the Optimization scheme, it retrieves anchors with higher precision compared to existing methods. Taking these anchors as reference points, ST-GEARS first rigidly aligns sections, then introduces and infers Elastic Fields to counteract distortions. ST-GEARS denoises the fields using context information by Gaussian Denoising. Utilizing the denoised fields, it eliminates distortions and eventually recovers original spatial profile through innovative and mathematically proved Bi-sectional Fields Application. Studying ST-GEARS on both bi-sectional registration and complete tissue reconstruction across sectional distances and sequencing platforms, we observed its outstanding performance in spatial information recovery across tissue, cell, and gene levels compared to current approaches. Through this recovery, ST-GEARS provides precise and well-explainable gears between in vivo situations and 3D in vitro analysis, powerfully fueling the potential of biological discoveries.

bioinformatics↗

bioRxiv2023

Aversive memories can be weakened during human sleepvia the reactivation of positive interfering memories

Recollecting painful or traumatic experiences can be deeply troubling. Sleep may offer an opportunity to reduce such suffering. We developed a procedure to weaken older aversive memories by reactivating newer positive memories during sleep. Participants viewed 48 nonsense words each paired with a unique aversive image, followed by an overnight sleep. In the next evening, participants learned associations between half of the words and additional positive images, creating interference. During the following non-rapid-eye-movement sleep, auditory memory cues were unobtrusively delivered. Upon waking, presenting cues associated with both aversive and positive images during sleep, as opposed to not presenting cues, weakened aversive memory recall while increasing positive memory intrusions. Substantiating these memory benefits, computational modeling revealed that cueing facilitated evidence accumulation toward positive affect judgments. Moreover, cue-elicited theta brain rhythms during sleep predominantly predicted the recall of positive memories. A noninvasive sleep intervention can thus modify aversive recollection and affective responses. Significance StatementNot all memories are welcome: recollection and involuntary intrusions of aversive memories can be debilitating. Prior research shows that reactivating positive memories benefits psychological health, and memory reactivation can be externally triggered during human sleep. Here, we introduced a novel sleep-based memory updating procedure to weaken older aversive memories by reactivating newer positive memories. We found that such reactivation not only weakened the recall of aversive memories, but also increased involuntary intrusion of positive memories. Moreover, reactivation enhanced positive affect judgments. Our findings open new avenues for weakening aversive or traumatic memories during human sleep.

neuroscience↗