Biology subjects

Wang, F.

Publications and source records attributed to Wang, F..

At least 163 records · Page 9Linked to original sources

bioRxiv2025

scTransMIL Bridges Patient-Level Phenotypes and Single-Cell Transcriptomics for Cancer Screening and Heterogeneity Inference

Single-cell sequencing technology has revolutionized cancer research by revealing unprecedented insights into tumor heterogeneity. However, reliably connecting patient-level cancer phenotypes with single-cell transcriptomic profiles remains challenging, due to technical constraints and labeling ambiguity. This further hinders the precise cancer screening and intensive study of tumor mechanism based on single-cell sequencing. To bridge this gap, we introduce scTransMIL, a scRNA-seq Transformer-based Multi-Instance Learning framework that learns whole-genome context to deliver comprehensive cancer insights at the sample, cell, and gene levels across three biological scales: (1) accurate patient-level cancer phenotype prediction, (2) precise single-cell disease scoring (validated on 4 million single cells), and (3) genome-wide biomarker discovery. Benchmark experiments demonstrate scTransMILs exceptional performance, including robust out-of-distribution generalization and clinically relevant prediction of metastatic tissue-of-origin, a crucial capability for identifying cancers of unknow primary. At single-cell resolution, scTransMIL identified both known and novel biomarkers for tumor B cells that conventional differential expression analysis failed to detect while maintaining consistent concordance with malignant cell annotations. scTransMILs adaptability is exemplified in acute myelocytic leukemia, where minimal fine-tuning with only a few patient sample labels enabled: (i) discovery of novel disease subtypes with distinct clinical outcomes, (ii) reconstruction of differentiation trajectories at the single-cell resolution, and (iii) identification of subtype-specific gene signatures. In summary, by systematically linking cellular and molecular profiles with clinical disease phenotypes, scTransMIL emerges as a transformative tool poised to advance both basic cancer research and precision oncology applications.

bioinformatics↗

bioRxiv2025

Meta-AlignNN: A meta-learning framework for stable brain-computer interface performance across subjects, time, and tasks

Practical application of brain-computer interfaces (BCIs) requires stable mapping between neuronal activity and behavior through various behavioral contexts and for different individuals. Due to neural activity instability, BCIs require frequent recalibration to maintain robust performance. Early approaches to addressing BCI stability issues mainly focused on tackling the challenge of neural activity changes over time. However, future BCI applications involve diversified scenarios and subjects, requiring solutions that address neural variability across time, subjects, and tasks. This study proposes a meta-learning-based algorithm to achieve BCI stability, named "Meta-AlignNN." By capitalizing on the consistency of neural population dynamics, it provides a unified solution for maintaining BCI stability, robustness, and scalability across subjects, time, and tasks. Tested over two years on four tasks with three monkeys, as well as on public datasets, the approach has achieved significantly excellent performance in both offline decoding and real-time brain-control, outperforming existing methods. Our findings provides a foundation for meeting the clinical demands of longterm, efficient, and stable usability across patients and tasks, offering a compelling solution for practical BCI applications.

bioengineering↗

bioRxiv2025

Wearable, High-Density, Time-Domain Diffuse Optical Tomography Array for Functional Neuroimaging

Noninvasive functional neuroimaging techniques, such as functional magnetic resonance imaging (fMRI) and electroencephalography (EEG), are essential tools for understanding brain activity and cognition for various neurological and mental health conditions. While fMRI offers high spatial resolution, its limited temporal resolution and costly large-form-factor restricts its accessibility and practicality for many applications. In contrast, EEG is more affordable and portable but has limited spatial resolution. In the present study, we overcome the limitations of existing neuroimaging technologies with the development of Micro-DOT, a functional near-infrared spectroscopy (fNIRS) system capable of high-density, time-domain diffuse optical tomography (HD-TD-DOT). Micro-DOT tackles the tradeoff between form factor and spatial resolution that has been a longstanding issue with existing fNIRS systems through the use of a unique hardware architecture that arrays HD-TD-DOT-capable electronics directly at the tissue surface. This is made possible with complementary-metal-oxide-semiconductor (CMOS) source-detector chiplets that contain all the electronics and optics necessary for HD-TD-DOT operation, and can be mounted on flexible polyimide packaging with a very minimal footprint. Pairing these hardware innovations with an advanced volumetric reconstruction software, Micro-DOT achieves in-plane spatial resolution, depth resolution, and localization accuracy comparable to fMRI, while maintaining the wearable form factor and portability of EEG, making it a viable stand-alone system for measuring subject-specific brain activation.

neuroscience↗

bioRxiv2025

The proofreading mechanism of the human leading strand DNA polymerase ϵ holoenzyme

Significance StatementReplicative DNA polymerases (Pol) function with a sliding clamp and their proofreading exonuclease provides high fidelity. Thus, study of the proofreading mechanism must both contain the Pol-clamp complex and generate the mismatch in situ to ensure the mismatched primer follows a physiological route from the pol to exo site. Despite numerous previous studies, none satisfy both criteria. This study on human Pol{varepsilon}-PCNA meets both criteria. Cryo-EM analysis captured proofreading intermediates that reveal an unexpected proofreading process. The primer is unwound by 6 nucleotides that scrunches to form out-of-register base pairs with the template. These findings provide new insights and calls for reevaluation of proofreading mechanisms performed without a sliding clamp or with an existing mismatch. The eukaryotic leading strand DNA polymerase epsilon is a dual function enzyme with a proofreading exonuclease site located 40 angstroms from the DNA synthesizing polymerase site. Errors in Pol epsilon proofreading can cause various mutations, including C to G transversions, the most prevalent mutation in cancers and genetic diseases. Pol epsilon interacts with all three subunits of the PCNA ring to assemble a functional holoenzyme. Despite previous studies on proofreading of several polymerases, how Pol epsilon, or any Pol complexed with its sliding clamp proofreads a mismatch generated in situ has been unknown. We show here by cryo-EM that a template/primer DNA substrate with a pre-existing mismatch cannot enter the exo site of Pol epsilon/PCNA holoenzyme, but a mismatch generated in the Pol site yields three proofreading intermediates of Pol epsilon/PCNA holoenzyme. These intermediates reveal how the mismatch is dislodged from the Pol site, how the DNA unwinds 6 base pairs and how the unpaired primer 3'-end is inserted into the exo site for cleavage. These results unexpectedly demonstrate that PCNA imposes strong steric constraints that extend unwinding and direct the trajectory of mismatched DNA, and that this trajectory is dramatically different than for Pol epsilon in the absence of PCNA. These findings suggest a physiologically relevant proofreading mechanism for the human Pol epsilon/PCNA holoenzyme.

biochemistry↗

bioRxiv2025

mcDETECT: Decoding 3D Spatial Synaptic Transcriptomes with Subcellular-Resolution Spatial Transcriptomics

Spatial transcriptomics (ST) is a powerful tool for studying the molecular basis of brain diseases. However, most current analyses focus only on nuclear or somatic expression, overlooking distally localized mRNAs, known as the "dark transcriptome." These transcripts, which make up nearly 40% of the brain transcriptome, are packaged into RNA granules and transported to polarized neuronal structures critical for neuronal function. Here, we present mcDETECT, a machine learning framework that leverages in situ ST (iST) data to detect RNA granules and resolve their RNA composition. Applying mcDETECT to simulated and real datasets from multiple iST platforms, we demonstrate robust granule detection and molecular subtyping. Granule-informed downstream analysis further revealed functionally distinct neuronal substates not captured by somatic expression alone. In an Alzheimers disease (AD) mouse model, mcDETECT uncovered alterations in RNA granule distributions and associated neuronal substates prior to measurable neuronal loss, highlighting potential therapeutic targets for early AD pathology.

bioinformatics↗

bioRxiv2025

Critical Environmental Limits: Assessing the Limitations of Core Temperature Inflection Point (CTIP) and Biophysical Modeling

The Core Temperature Inflection Point (CTIP) method and biophysical modeling are widely used to determine critical environmental limits (CEL), yet their validity under prolonged heat exposure remains untested. This study evaluated their predictive accuracy by exposing 36 healthy young adults (20 males, 16 females; age: 20.9-22.4 yr) to five counterbalanced 8-hour heat trials in a controlled chamber (36{degrees}C/74.5% RH, 40{degrees}C/55.0% RH, 44{degrees}C/29.2% RH, 47{degrees}C/35.6% RH, 50{degrees}C/24.5% RH). These conditions were selected based on prior CTIP and biophysical model predictions of CELs. Participants engaged in sedentary office tasks (1.29- 1.67 METs), wore standardized summer clothing (0.39-0.40 clo), and had ad libitum access to an electrolyte drink, with a 500-kcal sandwich provided at midday. Rectal temperature (Trec) was continuously monitored. Contrary to CTIP and biophysical model predictions, all five conditions remained compensable (Trec rise rate [≤] 0.1{degrees}C/h), with mean peak Trec staying well below heatstroke thresholds (38.2 {+/-} 0.4{degrees}C). At 44{degrees}C/29.2% RH, females had significantly lower Trec than males (p < 0.05), but steady-state Trec responses were similar between sexes (all p > 0.10). Collectively, CTIP and biophysical models substantially underestimated CELs, leading to overpredicted heat risk across all trials. These findings challenge the reliability of current predictive methods, suggesting human tolerance may exceed existing estimates. Refining these models is essential for improving heat risk assessment during real-world heatwaves and informing public and occupational health guidelines in a warming climate. NEW & NOTEWORTHYThis study demonstrates that healthy young adults maintained heat balance over 8 hours across five extreme heat conditions (36-50{degrees}C), despite the Core Temperature Inflection Point (CTIP) method and biophysical modeling predicting uncompensability. These findings challenge the reliability of these tools for assessing prolonged heat exposure, highlighting the need for refined models to accurately predict critical environmental limits in real-world scenarios, such as extreme heatwaves.

physiology↗

bioRxiv2025

Quantifying brain-wide cerebrospinal fluid flow dynamics using slow-flow-sensitized phase-contrast MRI

Cerebrospinal fluid (CSF) flow is a key component of the brains waste clearance system. However, our understanding of CSF flow in the human brain, particularly within the brain-wide subarachnoid space (SAS), is limited due to a lack of non-invasive tools for measuring slow flow. Here, we propose a CSF flowmetry technique using phase-contrast MRI combined with a slow-flow-sensitized acquisition. It achieves high sensitivity in measuring slow CSF flow (e.g., 100 m/s), and enables quantitative measurement of the velocity and direction with whole-brain coverage, spanning from ventricles to SAS. Our proof-of-concept results demonstrate repeatable flow measurements and show that cardiac pulsation induces coherent CSF flow changes within the SAS. Our data also suggest that cardiac pulsation has a stronger driving effect on brain-wide CSF flow compared to respiration. This technique provides a valuable tool for investigating CSF dynamics and pathways to advance a holistic understanding of brain-wide CSF flow. TeaserA novel MR technique enables noninvasive, quantitative mapping of slow cerebrospinal fluid flow in the subarachnoid space across the human brain.

bioengineering↗

bioRxiv2025

RNA Architecture Underlies Discontinuous Transcription and Evolution of Coronavirus

Coronaviruses employ discontinuous transcription to produce canonical subgenomic RNAs (sgRNAs) essential for gene expression, as well as non-canonical sgRNAs with unclear function. Nevertheless, the mechanisms regulating the balance between canonical and non-canonical sgRNAs throughout the viral replication cycle remain unknown. We conducted an integrated analysis on 234 transcriptome and 12 RNA-RNA interactome samples across various coronavirus genera. RNA-RNA interactions were identified between TRS-B and TRS-L flanking regions in the same genomic direction, correlating with the formation of canonical junctions. Non-canonical junctions frequently span short or long genomic distance, with short-range junctions generally mediated by reverse complementary stem-loops and coinciding with genomic deletion regions. Conserved long-range non-canonical sgRNAs were identified across different coronaviruses, and these sgRNAs harbor ORF10 or an evolving gene to suppress antiviral innate immune responses. This work establishes a structural framework that enables an understanding of the regulatory mechanism behind discontinuous transcription and the evolutionary pathway of coronaviruses. TeaserDeciphering coronavirus transcription and evolution: an integration analysis of transcriptome and RNA-RNA interactome. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=165 HEIGHT=200 SRC="FIGDIR/small/642953v1_ufig1.gif" ALT="Figure 1"> View larger version (45K): org.highwire.dtl.DTLVardef@1c58502org.highwire.dtl.DTLVardef@9f87d7org.highwire.dtl.DTLVardef@18f3492org.highwire.dtl.DTLVardef@c49362_HPS_FORMAT_FIGEXP M_FIG C_FIG

microbiology↗

bioRxiv2025

Thermal and mechanical modalities converge in the noxious range

SummaryThere are competing theories on how nociceptive input is encoded by sensory fibers. That sensory modalities are exclusively conveyed by distinct populations of fibers is supported by ablation studies, indicating that TRPV1+and MrgprD+afferents selectively encode for heat and mechanical input, respectively. However, interpreting ablation results can be clouded by compensatory plasticity. Furthermore, a population-level parametric analysis of afferent response profiles to natural stimuliin vivois still scarce. Using functional imagingin vivoin mice, we found that most TRPV1+neurons responded to noxious heat, but not cooling. While none of them responded to innocuous mechanical input, about half were also sensitive to noxious mechanical stimuli. In contrast, 80% of MrgprD+neurons responded to noxious mechanical stimuli. Yet, 70% were also sensitive to noxious thermal stimuli. Of innocuous mechanosensitive MrgprD+neurons, <15% responded to innocuous warming and none to innocuous cooling. Polymodality in the innocuous thermal and mechanical range also occurred in <10% of all primary afferents. Acute silencing of TRPV1+or MrgprD+afferents inhibited both thermal and mechanical nociception. In contrast, ablation did not reproduce a combined loss of sensitivity, which appeared to be due to compensatory central disinhibition. Our findings reveal that modality separation dominates in the innocuous regime, while polymodality predominates in the noxious range.

neuroscience↗

bioRxiv2025

Formaldehyde induces and promotes Alzheimer's disease pathologies in a 3D human neural cell culture system

Alzheimers disease (AD) arises from complex multilevel interactions between genetic, epigenetic, and environmental factors. Recent studies suggest that exposure to the environmental and occupational toxicant formaldehyde (FA) may play a significant role in AD development. However, the effects of FA exposure on A{beta} and tau pathologies in human neural cell 3D culture systems remain unexplored. To investigate FAs role in AD initiation, we differentiated 3D-cultured immortalized human neural progenitor ReN cells (ReNcell VM) into neurons and glial cells, followed by FA treatment. FA exposure for 12 weeks resulted in a dose-dependent increase in A{beta}40, A{beta}42, and phosphorylated tau levels. To further examine FAs role in AD progression, we established a 3D human neural cell culture AD model by transfecting ReN cells with AD-related mutant genes, including mutant APP and PSEN1, which recapitulate key AD pathological events. Our findings demonstrate that FA exposure significantly elevated A{beta}40, A{beta}42, and phosphorylated tau levels in this 3D-cultured AD model. These results suggest that FA exposure contributes to the initiation and progression of AD pathology in 3D-cultured human neural cells.

neuroscience↗

bioRxiv2025

Reading fluency development is predicted by cortical latency

The development of fluent literacy skills from childhood to adolescence is strongly constrained by the temporal dynamics of word recognition. Capturing the neural basis of these subtle timing changes in word recognition has remained challenging with EEG measures that lack reliability at the individual subject level. Here, we leverage phase information from Steady-State Visual Evoked Potentials (SSVEPs) to derive precise and reliable temporal dynamics of neural signatures underlying visual word form recognition at the individual level and examine their relationship to reading fluency and comprehension. Typically developing readers (N = 68), aged 8-15 years, viewed a stream of four-character stimulus strings presented at 3 Hz. Significant SSVEP signals emerged for nearly all participants. Signals at 3, 6, and 9 Hz harmonics exhibited a phase pattern consistent with a delay model, indicating a mean latency of approximately 170 milliseconds. Individual variations in latencies demonstrated (a) high internal consistency (R = .94); (b) stability across variations in letter string forms (familiar words, nonwords with familiar letters, nonwords with unfamiliar pseudo-characters); (c) a linear relationship with age; and most remarkably, (d) a predictive relationship with individual variation in reading fluency and reading comprehension. These results establish SSVEP visual word form latency as a promising approach for investigating the neural basis of reading development, paving the way for future translational applications in education and offering potential solutions to broader societal challenges in promoting population-level reading fluency.

neuroscience↗

bioRxiv2025

Transcription factors instruct DNA methylation patterns in plant reproductive tissues

DNA methylation is maintained by forming self-reinforcing connections with other repressive chromatin modifications, resulting in stably silenced genes and transposons. However, these mechanisms fail to explain how new methylation patterns are generated. In Arabidopsis, CLASSY3 (CLSY3) targets the RNA-directed DNA methylation (RdDM) machinery to different loci in reproductive tissues, generating distinct epigenomes via unknown mechanism(s). Here, we discovered that several different REPRODUCTIVE MERISTEM (REM) transcription factors are required for methylation at CLSY3 targets specific to male or female reproductive tissues. We designate these factors as REM INSTRUCT METHYLATION (RIMs) and demonstrate that disruption of their DNA binding domains, or the motifs they recognize, blocks RdDM. These findings not only reveal RIMs as the first sex-specific RdDM proteins but also establish a critical role for genetic information in targeting DNA methylation. This novel mode of targeting expands our understanding of how methylation is regulated to include inputs from both genetic and epigenetic information.

plant biology↗

bioRxiv2025

Neuromodulation of the endbulb of Held synapse in the cochlear nucleus

Synapses vary greatly in synaptic strength and plasticity, even within the same circuitry or set of pre- and postsynaptic neurons. Neuromodulation is a candidate mechanism to explain some of this variability. Neuromodulators such as monoamines can differentially regulate presynaptic function as well as neuronal excitability. Variability is found also for the large calyceal synapses of the auditory pathway that are endowed with high synaptic vesicle (SV) release probability (Pvr) and large postsynaptic currents enabling reliable and temporally precise transmission of auditory information. Here we investigated whether the calyceal endbulb of Held synapse formed by auditory nerve fibers onto bushy cells (BCs) in the anteroventral cochlear nucleus (AVCN) is modulated by norepinephrine (NE) and serotonin (5-HT). Using electron microscopy (EM) of the cochlear nucleus we found evidence for putative monoaminergic varicosities in both ventral and dorsal divisions. Immunostaining for vesicular 5-HT and NE transporters revealed NE-containing and 5-HT-containing varicosities in the AVCN, juxtaposed to both endbulbs and BCs. Furthermore, we detected immunofluorescence for 5-HT1B, 5-HT4, 5-H7 receptors (R) and 2C-adrenergic receptors (AR) in BCs. We used voltage-clamp recordings from mouse BCs in order to uncover potential presynaptic effects of neuromodulation, which revealed an increase in frequency of miniature excitatory postsynaptic currents (mEPSCs) upon application of NE but not 5-HT. Evoked synaptic transmission was unaffected by the application of either NE or 5-HT. Likewise, while studying the biophysical properties of the BCs, we did not observe effects of NE or 5-HT on low-voltage-activated K+ (K+LVA) and hyperpolarization-activated mixed cation (HCN) channels during application. In summary, we report evidence for the presence of monoaminergic innervation in the cochlear nucleus and for subtle functional NE-neuromodulation at the endbulb of Held synapse.

neuroscience↗

bioRxiv2025

Cryo-EM analysis of the Bacillus thuringiensis extrasporal matrix identifies F-ENA as a widespread family of endospore appendages across the Firmicutes phylum

For over 100 years, Bacillus thuringiensis (Bt) has been used as an agricultural biopesticide to control pests caused by insect species in the orders of Lepidoptera, Diptera and Coleoptera. Under nutrient starvation, Bt cells differentiate into spores and associated toxin crystals that can adopt biofilm-like aggregates. We reveal that such Bt spore/toxin biofilms are embedded in a fibrous extrasporal matrix (ESM), and using cryoID, we resolved the structure and molecular identity of an uncharacterized type of pili, referred to here as Fibrillar ENdospore Appendages or F-ENA. F-ENA are monomolecular protein polymers tethered to the exosporium of Bt and are decorated with a flexible tip fibrillum. Phylogenetic analysis reveals that F-ENA is widespread not only in the class Bacilli, but also in the class Clostridia, and the cryoEM structures of F-ENA filaments from Bacillus, Anaerovorax and Paenibaccilus reveal subunits with a generic head-neck domain structure, where the {beta}-barrel neck of variable length latch onto a preceding head domain through short N-terminal hook peptides. In Bacillus, two collagen-like proteins (CLP) respectively tether F-ENA to the exosporium (F-Anchor), or constitute the tip fibrillum at the distal terminus of F-ENA (F-BclA). Sedimentation assays point towards F-ENA involvement in spore-spore clustering, likely mediated via F-BclA contacts and F-ENA bundling through the antiparallel interlocking of the head-neck units.

molecular biology↗

bioRxiv2025

Unifying Genetic and Chemical Perturbagen Representation through a Hybrid Deep Learning Framework

Artificial Intelligence virtual cell (AIVC) holds transformative potential for biomedical research. Central to this vision is the systematic modeling of genetic and chemical perturbation phenotypes to accurately predict cellular dynamic states from diverse interventions. However, disparities in screening agents, library scales, experimental technologies, and data production efficiency hinder the integration, modeling, and analysis of the cross-data. Here we present UniPert-G2CP, a two-phase deep learning approach comprising i) UniPert, a multimodal molecular representation model that bridges genetic and chemical domains, and ii) G2CP (Genetic-to-Chemical Perturbation transfer learning), which systematically transforms CRISPR screen-based genetic insights into chemical perturbation modeling for cost-effective in silico drug screening. UniPert not only encodes multimodal perturbagens into a unified functionally interpretable sematic embedding space, but also improves phenotypic effect prediction for previously unseen gene perturbations and drug treatments. Building upon UniPert, G2CP successfully modeled large-scale cellular post-perturbation states spanning 4,994 gene and 7,821 compound perturbagens, while reducing modeling data costs by over 60%. We demonstrate that UniPert-G2CP enables efficient, generalizable simulations of multicellular, multi-domain perturbation cause-effect spaces, revealing differential cellular biological causality and informing mechanism-driven therapy. UniPert-G2CP opens new avenues for biological causal foundation model building, AIVC creation, and AI-powered precision medicine.

bioinformatics↗

bioRxiv2025

Mac/Lac-tosylceramide regulates intestinal homeostasis and secretory cell fate commitment by facilitating Notch signaling

Cell-to-cell communication via Delta-Notch signaling is widely used in various tissues and organs to regulate development and patterning; however, the mechanisms regulating Notch signaling for precise cell fate decisions remain poorly understood. Similar to mammals, the intestinal stem cells (ISCs) in the adult Drosophila midgut generate both absorptive and secretory cell progeny, guided by differential levels of Notch activation. Here we performed a forward genetic screen in Drosophila and identified glucosylceramide synthase (GlcT), a rate-limiting enzyme for glycosphingolipid (GSL) production, whose mutation causes the development of secretory cell tumors. Genetic analysis of the GSL synthesis pathway, combined with metabolite rescue experiments, revealed that the tumor formation is linked to a deficiency in Mactosylceramide/Lactosylceramide. This deficiency impaired the endocytic recycling of the Delta, subsequently reducing Notch signaling activation. Conditional knockout of UGCG, the mammalian ortholog of GlcT, in mouse small intestine caused an excessive differentiation of goblet cells, phenotypes similar to these caused by Notch inhibition. Our study suggests an evolutionarily conserved role for a specific GSL metabolite in modulating Notch signaling during stem cell fate decisions and provides a molecular connection between ceramide metabolism and Notch signaling in regulating tissue homeostasis and tumor formation.

developmental biology↗

bioRxiv2025

scGraphETM: Graph-Based Deep Learning Approach for Unraveling Cell Type-Specific Gene Regulatory Networks from Single-Cell Multi-Omics Data

Multi-omics sequencing technologies can jointly measure transcriptome and chromatin accessibility at the single-cell resolution. This enables inference of gene regulatory networks (GRNs) at the cellular level, thereby elucidating high-resolution differential GRNs associated with diseases. However, existing methods lack interpretability and scalability. We present single-cell Gene Regulatory Inference with Prior knowledge (scGRIP), a graph variational autoencoder with 3 technical contributions. First, we treat transcription factors (TF), target genes (TG), and regulatory elements (RE) as nodes and their potential TF-RE and RE-TG interactions as edges using a prior cis-regulatory knowledge graph. Second, we tokenize the single-cell chromatin accessibility and gene expression with a shared codebook to compute cell-specific feature embedding. Third, we incorporate a GraphSHAP technique to infer GRN edge attribution at the single-cell level. scGRIP, evaluated against existing regulatory inference frameworks such as LINGER and scGLUE, demonstrates that integrating prior regulatory knowledge yields regulatory interaction features that improve discrimination of cell types and disease conditions. In particular, scGRIP identified coordinated activation of immune signaling, amyloid-related responses, and downstream target gene programs. The inferred GRNs for Alzheimers disease were supported by functional enrichment analyses and independent reference dataset.

genomics↗

bioRxiv2025

Left ventricular transcriptome and extracellular vesicle-derived miRNAs in porcine donation after circulatory death (DCD) hearts undergoing prolonged working mode ex situ perfusion

Donation after circulatory death (DCD) hearts suffer from warm ischemia-reperfusion injury, which compromises graft quality. We have previously demonstrated that a postconditioning-based cardioprotective treatment (1% Intralipid, 2% (v/v) sevoflurane, 3 nM remifentanil) improved the function and viability of porcine DCD hearts undergoing ex situ heart perfusion (ESHP). However, transcriptional changes in DCD hearts subjected to prolonged working mode ESHP as compared with healthy hearts, and more so transcriptional changes in the presence or absence of cardioprotection, as well as the early reperfusion profile of extracellular vesicle (EV)-derived miRNAs, critical regulators of the transcriptional control, have not been investigated. A total of 5,483 differentially expressed transcripts were identified in left ventricular tissue samples of DCD hearts (N=10) collected after 6 hours of ESHP when compared with healthy hearts not subjected to ESHP (N=8), irrespective of cardioprotection. Downregulation of protein synthesis, metabolic, and DNA repair gene sets was most prominent, while inflammatory and remodeling gene sets were upregulated. Between DCD hearts that were treated with (pDCD_ESHP, N=5) or without (uDCD_ESHP, N=5) cardioprotection, only 43 transcripts were differentially regulated. Specifically, lipid breakdown, beta-oxidation, and DNA repair gene sets were upregulated in pDCD_ESHP, whereas lipid accumulation, inflammation, oxidative stress, and maladaptive remodeling gene sets were upregulated in uDCD_ESHP. Predominantly cardioprotective EV-derived miRNAs were released into the perfusate in pDCD_ESHP, while predominantly cardiac injury-associated miRNAs were released into the perfusate in uDCD_ESHP. Cardioprotection promoted adaptive rather than maladaptive transcriptional changes and enabled the release of potentially beneficial miRNAs.

molecular biology↗