Biology subjects

Liu, R.

Publications and source records attributed to Liu, R..

At least 73 records · Page 4Linked to original sources

bioRxiv2025

An EEG-based Multi-Scale Hybrid Attention and Squeeze Network for objective taste assessment

Taste perception is central to flavor experiences. Electroencephalography (EEG) signals carry rich information about taste perception. Because these neural data are independent of verbal reports and bypassing conscious filtering, EEG holds promise as a neural-sensing platform for objective taste representation. However, taste-evoked EEG signals exhibit complex spatiotemporal and spectral dynamics that require advanced computational approaches to decode effectively. This study developed an EEG-based Multi-Scale Hybrid Attention and Squeeze Network (MHASNet) for taste evaluation. EEG signals evoked by distinct taste stimuli were recorded, a dedicated taste-EEG dataset was compiled, and a novel deep learning architecture, MHASNet, was designed to classify these signals. MHASNet synergistically integrates multi-scale convolutions to capture temporal dynamics across different time scales, dual-attention mechanisms to localize discriminative brain regions and electrode positions, and a squeeze-and-excitation module to optimize frequency-band contributions-collectively enabling precise extraction of taste-specific neural signatures. Results showed that the proposed model achieved superior performance across five taste categories (sweet, sour, salty, bitter, and tasteless), with 94.33% accuracy, 91.37% F1-score, 91.89% precision, and 92.07% recall. These results surpass benchmark models while maintaining millisecond-level inference latency suitable for real-time applications. By complementing subjective evaluations and instrumental analyses, the model offers an objective, neurophysiology-based solution for taste evaluation.

neuroscience↗

bioRxiv2025

MDFIC2 is a sensory neuron-specific PIEZO channel auxiliary subunit

PIEZO channels are critical for sensory mechanotransduction. While MyoD-family inhibitor proteins were identified as PIEZO1 auxiliary subunits, their broader regulatory roles, particularly in sensory cells, remained unclear. Here we demonstrate native MDFIC and MDFI regulate endogenous PIEZO channel currents in various non-sensory cell types. However, neither MDFIC nor MDFI are expressed in primary sensory neurons. In these cell types we identified an uncharacterised member of this family, Mdfic2/Gm765, that shares the ability to physically bind to PIEZO1 and PIEZO2. MDFIC2 is selectively expressed in subsets of mechanosensitive neurons, including dorsal root ganglia, trigeminal ganglia, and vagal sensory neurons. Like its paralogues, MDFIC2 alters PIEZO1/2 mechanosensitivity and inactivation kinetics, converting them into high-threshold slowly inactivating mechanoreceptors. Extensive cryo-EM reveals a conserved binding pocket for these auxiliary subunits in the pore modules of both PIEZO1 and PIEZO2 mediated by the post-translationally modified distal C-termini of MyoD-family inhibitor proteins. This provides a comprehensive structural and functional characterisation of MyoD-family inhibitor proteins as PIEZO1/2 channel regulators and offers new insights into sensory physiology and mechanical pain mechanisms.

biophysics↗

bioRxiv2025

Lipid Metabolism Remodeling in Human Cardiomyocyte Differentiation and Maturation

During cardiac development, metabolic remodeling is characterized by a transition from glycolysis-dependent energy production to reliance on fatty acid oxidation. Lipids function as essential energy substrates and structural components throughout cardiac formation and maturation. However, the dynamic changes in endogenous lipid species during human cardiac development remain incompletely elucidated. In this study, we delineated distinct lipidomic profiles of human embryonic stem cells (ESCs), mesoderm (MES), cardiac progenitors (CPCs), immature cardiomyocytes (CMs), and mature CMs derived from ESCs using high-performance liquid chromatography (HPLC). Notably, ceramide concentrations peaked at the CPC stage, suggesting a pivotal role in mediating the MES-to-CPCs transition. Concurrently, lipid metabolites including GM3, phosphatidylcholine (PC), lysophosphatidylcholine (LPC), and lysophosphatidic acid (LPA) were significantly upregulated in mature CMs. Integrative transcriptomic and chromatin accessibility analyses further refined the landscape of lipid remodeling during CM differentiation. Functional assays demonstrated that LPA enhances the stability of the beating rhythm in CPCs, while both PC and LPA facilitate CM maturation. Collectively, our findings establish a comprehensive and dynamic lipidomic atlas of human cardiogenesis, providing a novel framework to advance the understanding of lipid-mediated regulatory mechanisms in human heart development.

developmental biology↗

bioRxiv2025

METAHIT enables comprehensive and flexible genome-resolved microbiome analysis with metagenomic Hi-C

Metagenomic Hi-C (metaHi-C) adds within-cell proximity information to shotgun sequencing, but existing analyses are fragmented and short-insert intra-contig read pairs from Hi-C libraries are often underused. Here, we present METAHICT, a modular and configurable workflow spanning read processing, metagenome-assembled genome (MAG) reconstruction, and downstream analysis. METAHICT consolidates outputs from bin3C, MetaCC, and ImputeCC and uses an expectation-maximization model of mapped insert sizes to select short-insert, shotgun-like Hi-C pairs for per-bin reassembly. Across seven datasets, binning and consolidation generally recovered more near-complete and medium-quality MAGs than individual Hi-C binners. On the sheep-gut long-read assembly, METAHICT's binning and consolidation stage recovered 929 MAGs meeting at least the medium-quality threshold, including 487 near-complete MAGs. Within the reassembly module, adding EM-selected Hi-C reads to shotgun-only inputs yielded further significant reductions in contamination across multiple environments. The workflow also integrates scaffolding, taxonomic annotation, contact-map visualization, and analysis of candidate mobile genetic element (MGE)-host pairs. Together, these capabilities provide standardized outputs for comparative and hypothesis-driven genome-resolved microbiome analysis.

bioinformatics↗

bioRxiv2025

Long-term dynamic profiles of cognitive kinases induced by different learning protocols

Learning is associated with activation of multiple protein kinases, but few details are known about the activation dynamics in response to different learning protocols. We addressed this issue by examining the long-term dynamics of kinases critical for long term synaptic facilitation (LTF) of the Aplysia sensorimotor synapse. Three serotonin (5-HT) protocols have been found to induce LTF with distinct effectiveness: the five-pulse regular-spaced Standard protocol; the five-pulse irregular-spaced Enhanced protocol; and the two-pulse protocol with an interval of 45 min. We previously compared long-term dynamics of the mitogen-activated protein kinase (MAPK) isoform ERK after these protocols. Here we examined the long-term dynamics of additional kinases critical for LTF: p38 MAPK, protein kinase A (PKA), and p90 ribosomal S6 kinase (RSK). All four kinases showed complex dynamics of activity during 24 h, with a first wave of increase occurring shortly after 5-HT treatment and ending within 5 h, and a second wave from [~] 5 h to 18 h. After the Standard and two-pulse protocols, all kinase activities returned towards basal at 24 h, but after the Enhanced protocol some remained elevated at 24 h. Interactions and multiple feedback loops among the kinase pathways, and with the growth factors Aplysia neurotrophin (NT) and transforming growth factor-{beta} (TGF-{beta}), contribute to development of molecular clocks underlying these complex dynamics. These results help to delineate the molecular mechanisms underlying the induction of LTF and provide insights that may help design improved training protocols for induction and maintenance of LTF and long-term memory.

neuroscience↗

bioRxiv2025

DeviceAgent: An autonomous multimodal AI agent for flexible bioelectronics

The development of flexible bioelectronics remains a complex, multidisciplinary process that demands specialized expertise and labor-intensive efforts, limiting scalability, adaptability and accessibility. Here, we introduce DeviceAgent, an autonomous multimodal AI agent that integrates large language models (LLMs), vision-language models (VLMs), and domain-specific computational tools into a unified framework for bioelectronics research. Leveraging the emergent reasoning abilities of LLMs and VLMs, DeviceAgent enables zero and few-shot generalization, contextual learning, and flexible task execution across modalities. A multimodal context memory system orchestrates these capabilities, providing end-to-end support across the experimental pipeline-from high-level design objectives to fabrication protocol generation, visual defect inspection, and electrophysiological signal analysis, while maintaining human oversight at critical decision points. We demonstrate its capabilities through the development of stretchable mesh electronics for interfacing with human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs), a representative application involving complex device architectures, heterogeneous material nanofabrication, and electrophysiology analysis. DeviceAgent autonomously (1) generates customized bioelectronic layouts; (2) creates comprehensive fabrication protocols tailored to specific materials and processes; (3) identifies microscopic defects using visual reasoning; and (4) analyzes cardiac electrophysiological recordings in an interpretable manner. By embedding LLMs and VLMs within a structured, tool-augmented architecture, DeviceAgent establishes a scalable and accessible paradigm for AI-scientist collaboration in nanofabrication and bioelectronics research.

bioengineering↗

bioRxiv2025

Quantitative genetics of trauma induced mortality in Drosophila melanogaster

Traumatic brain injury is a major cause of chronic neurological impairment worldwide, and there is evidence that both genetic and environmental variation contribute to the likelihood of recovery. Using an insect model of traumatic brain injury, we examined variation in the risk of mortality using quantitative genetic approaches applied previously for life history traits in Drosophila melanogaster. We quantified additive genetic variance for mortality risk using a controlled breeding design and found levels of variation consistent with existing data on major fitness components. We found no evidence for inbreeding effects on mortality risk, suggesting that dominance genetic variance makes little contribution to this trait. To explain the high level of standing genetic variation, we considered whether mortality risk depends on the metabolic resources available to an individual, also known as "condition". We manipulated condition by inducing random mutations and by restricting calories during larval development. We found that reduced condition due to both random mutations and resource limitation significantly increased the risk of mortality following trauma. Among inbred lines, greater mortality risk was associated with lower viability, fecundity and longevity, consistent with an effect of genome-wide genetic quality. Our results suggest that further consideration of individual condition would be valuable for understanding and predicting variation in the outcomes of traumatic brain injury.

genetics↗

bioRxiv2025

Acute kidney injury disrupts cardiac remodeling via SerpinA3N

Cardiorenal syndrome type 3 (CRS-3) also known as acute reno-cardiac syndrome, refers to a condition where acute kidney injury (AKI) leads to acute cardiac dysfunction or injury. The underlying mechanisms have not been elucidated. In the current study, we examined the potential mechanisms for the development of myocardial pro-fibrotic factors and cardiac remodeling in C57BL/6J mice after AKI. The mice were subjected to bilateral pedicle clamping for 30 min followed by 24 h of reperfusion. Heart tissue was collected from control (Ctrl) and AKI mice and subjected to proteomic analysis by liquid chromatography-mass spectrometry (LC-MS) and differential mRNA expression analysis by RNA-seq. Cardiac tissue was collected to assess serine protease activity, RNA, protein and cytokine expression. The results indicate that AKI significantly upregulated SerpinA3N in the heart which was negatively correlated with serine protease (Granzyme B) activity. Further, AKI induced the development of several pro-fibrotic factors, induced mitochondrial dysfunction and increased inflammation. Finally, using H9C2 cells we demonstrated that inhibiting SerpinA3N with XAV939 increased Granzyme B activity. Thus, this study suggests that AKI leads to the development of pro-fibrotic factors in the heart and disrupts cardiac remodeling by SerpinA3N mediated inhibition of serine protease Granzyme B activity. Decreased SerpinA3N expression in the heart followed by AKI could lead to a more balanced extracellular matrix (ECM) composition in the heart with an alleviated pro-fibrotic response.

molecular biology↗

bioRxiv2025

RhoV Promotes Proliferation, Migration, and MAPK Activation in Pancreatic Ductal Adenocarcinoma

Background and AimsPancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy characterized by diagnosis at advanced stages, limited therapeutic options, and frequent resistance to therapies. Although oncogenic KRAS mutations are central drivers of PDAC, alternative pathways indisputably contribute to its tumorigenesis and progression. RhoV, a member of the Rho family of small GTPases, has been implicated in tumor development in other cancer types, such as breast cancer and lung adenocarcinoma; however, its role in PDAC remains unclear. MethodsIn this study, we investigated the expression and functional impact of RhoV on PDAC. Analysis of publicly available datasets and immunohistochemical profiling of 114 PDAC patient specimens were used to evaluate the expression of RhoV in PDAC and its prognostic impact. Overexpression and CRISPR-Cas9-mediated knockout of RhoV were established in three pancreatic cancer cell lines. Functional analyses, such as cell proliferation, migration, invasion, colony formation, spheroid growth, and mouse xenograft, were used to evaluate the role of RhoV in PDAC cells. ResultsRhoV overexpression was associated with reduced overall and recurrence-free survival in public datasets and our own patient cohort. Functional assays demonstrated that RhoV overexpression promoted PDAC cell proliferation, colony formation, and spheroid growth, whereas knockout of RhoV suppressed these changes. Moreover, RhoV enhanced PDAC cell migration and invasion in vitro, accompanied by downregulation of E-cadherin and upregulation of N-cadherin and vimentin, indicating induction of epithelial-mesenchymal transition. Mechanistically, RhoV overexpression activated key MAPK pathway components, including phosphorylation of ERK, JNK, and p38. In vivo, xenograft models confirmed that RhoV drives tumor growth and increases tumor burden. ConclusionThese results establish RhoV as a novel oncogenic factor in PDAC progression and highlight its potential as a biomarker and therapeutic target, warranting further investigation into combinatorial targeting strategies to overcome KRAS inhibitor resistance.

cancer biology↗

bioRxiv2025

Illuminating the Druggable Proteome with an AI Protein Profiling Platform

Most human proteins lack chemical probes or pharmaceutical modulators, leaving much of the proteome unexplored. 1,2 Activity-based protein profiling has enabled proteome-scale discovery of protein-ligand interactions and covalent inhibitors, 3-5 but remains limited by probe chemistry, protein abundance, and discordant ligandability assignments across studies. 6-9 Machine-learning (ML) models can in principle generate proteomewide ligandability maps, but the predictive utility of current models is limited by the requirement for structures and the use of incomplete and weak training labels. 9-12 Here we developed an artificial intelligence protein profiling (AiPP), a sequence-based multitask platform built on the ESMC protein language model, 13,14 to accelerate proteomewide therapeutic discovery and target identification. While its primary task is identification of covalently ligandable cysteines, seven additional task heads and two external modules provide broader context by annotating reversible ligand-binding residues, disordered molecular recognition features, cysteine functional context, and reactivities. Central to the development is the LatentLift clustering approach, which leverages latent space similarities to reconcile conflicting experimental labels and facilitate model training. Applied to the human proteome, AiPP generated a cysteine-directed ligandability atlas that overcomes the limitations of current chemo-proteomic maps. As a proof of concept, we demonstrate that AiPP can guide the discovery of covalent allosteric inhibitors targeting the previously undruggable protein tyrosine phosphatase PTPN6. By linking sequence, ligandability and biological context, AiPP provides an open resource for therapeutic discovery, while LatentLift offers a general strategy for harmonizing proteomics data for ML applications.

biophysics↗

bioRxiv2025

Real-Time Viability Assessment of Ex Vivo Mouse Kidneys for Transplant Applications Using Dynamic Optical Coherence Tomography

Kidney transplantation remains the preferred treatment for patients with end-stage kidney disease. However, the ongoing shortage of donor organs continues to limit the availability of transplant treatments. Existing evaluation methods, such as the kidney donor profile index (KDPI) and pre-transplant donor biopsy (PTDB), have various limitations, including low discriminative power, invasiveness, and sampling errors, which reduce their effectiveness in organ quality assessment and contribute to the risk of unnecessary organ discard. In this study, we explored the dynamic optical coherence tomography (DOCT) as a real-time, non-invasive approach to monitor the viability of ex vivo mouse kidneys during static cold storage over 48 hours. The dynamic metrics logarithmic intensity variance (LIV), early OCT correlation decay speed (OCDSe), and late OCT correlation decay speed (OCDSl) were extracted from OCT signal fluctuations to quantify temporal and spatial tissue activity and deterioration. Our results demonstrate that DOCT provides complementary functional information to morphological assessment offered by conventional OCT imaging, showing potential to improve pre-transplant organ evaluation and clinic decision-making.

bioengineering↗

bioRxiv2025

Broad brain biodistribution conferred by an AAV to restore TDP-43 function mitigates Frontotemporal Demenia-like deficits

TDP-43 dysfunction is an early pathogenic determinant of frontotemporal lobar degeneration with TDP-43 pathology (FTLD-TDP), a devastating disorder currently without effective therapy. Here, we exploit a blood-brain-barrier (BBB)-permeable AAV (AAV-PHP.eB) that confers broad brain biodistribution to restore TDP-43 function in a TDP-43 deficient model (CamKIIa-CreER;Tardbp mice) that mimics the early stage of TDP-43 dysfunction occurring in FTLD-TDP. Intracerebroventricular delivery by AAV-PHP.eB of CTR, our previously characterized splicing repressor, revealed its accumulation in [~]40% of adult hippocampal neurons. Remarkably, treatment of adult CamKIIa-CreER;Tardbpf/fmice with AAV-PHP.eB-CTR restored TDP-43 function, attenuated neuronal aberrant activity and memory deficits, and rescued neuron loss. Importantly, we showed that TDP-43s autoregulatory element restricts CTR expression to a physiological range. No overt phenotype was observed after long-term exposure to AAV-PHP.eB-CTR in aged mice, highlighting a favorable safety profile for this gene therapy. These results validate that BBB-crossing AAVs can deliver CTR with a biodistribution in the adult brain that is broad enough to rescue FTD-like phenotypes, supporting clinical testing of this gene therapy for FTLD-TDP.

neuroscience↗

bioRxiv2025

Asymmetric Social Representations in the Prefrontal Cortex for Cooperative Behavior

Cooperation is a hallmark of social species, enabling individuals to achieve goals that are unattainable alone. Across species, cooperative behaviors are often organized by distinct social roles such as leaders and followers, yet the neural mechanisms supporting such role-based coordination remain elusive. Here we introduce a new paradigm for studying cooperation in mice, where pairs of animals engage in a joint spatial foraging task that naturally gives rise to stable leader-follower roles predictive of learning speed. Disruption of medial prefrontal cortex (mPFC) activity, particularly in followers, impairs cooperation and induces reciprocal shifts in how animals weigh self- and partner-related cues for decision-making. Calcium imaging reveals that mPFC encodes both leadership dynamics and an egocentric social value map of the partners position, each in an asymmetric, role-specific manner. Combining this behavior with a novel multi-agent inverse reinforcement learning framework, we identify latent value functions that guide cooperative decisions and are decodable from mPFC activity. These findings uncover fundamental neural computations that support cooperation, revealing how social roles shape decision-making in real time. Our work opens new avenues for investigating the cellular and circuit basis of social cognition and collective behavior.

neuroscience↗

bioRxiv2025

Microtubule Inhibition Triggers MYC-Mediated Immunogenic Cell Death in Breast Cancer

Oncogenic MYC promotes cancer cell proliferation, metabolism, and death, while also driving immunosuppression in the tumour microenvironment, complicating immune-based therapies. To counter MYC-driven immune evasion while leveraging MYC-dependent synthetic lethality (MYC-SL), we identified microtubule-targeting agents, including eribulin, as potent inducers of immunogenic cell death in MYChigh triple-negative breast cancer (TNBC). A screen of 528 oncology compounds using damage-associated molecular pattern (DAMP) reporters revealed that microtubule inhibitors induced key DAMPs, including HMGB1 secretion, calreticulin exposure, and double-stranded DNA release, leading to gasdermin-E associated cell death in MYChigh TNBCs. Immune cell co-culture assays showed immune activation, and patient-derived explant cultures confirmed pro-inflammatory cytokine responses. In vivo, cell-free media from eribulin-treated MYChigh murine TNBCs enhanced tumour protection in vaccination models compared to MYC-knockdown controls, linking MYC-dependent DAMP release to immunogenicity. These findings highlight a dual-function therapeutic strategy: agents that selectively induce MYC-dependent immunogenic cell death can provide both targeted cytotoxicity and local immune stimulation, thereby addressing a key limitation of conventional chemotherapeutics, offering a new approach for MYC-driven cancers.

cancer biology↗

bioRxiv2025

Identifying a Vaginal Microbiome-Derived Selective Antibiotic Metabolite via Microbiome Pharmacology Analysis

The vaginal microbiome plays a critical role in maintaining immune and epithelial homeostasis in the female reproductive tract. Bacterial Vaginosis (BV) is deleterious to female health, causing the loss of beneficial Lactobacillus species, overgrowth of anaerobic taxa, changes in vaginal pH, breakdown of protective mucins and epithelial barriers, and activation of the immune system. Treatment with gel-based antibiotics (Metronidazole or Clindamycin) resolves BV for 85% of patients, but 50% of those cases recur, indicating a need to identify strategies for overcoming antibiotic resistance and achieving a more durable response. Here, we developed a systems biology approach termed Microbiome Pharmacology Analysis to characterize the antibiotic potential of vaginal microbes, their metabolites and functions, via computational fusion of human cohort multi-omics and post-drug perturbation transcriptomic profiles. We focused on Clindamycin and Metronidazole as candidate drugs and screened 780 vaginal microbiome-drug mimicry candidates to identify candidate taxa and metabolites with antibiotic potential. We demonstrate experimentally that Lactobacillus crispatus-derived Hydroxyisocaproate (HICA) selectively kills Gardnerella vaginalis and that HICA enhances epithelial barrier integrity in a human vagina-on-a-chip system. Our work demonstrates the first use of Pharmacobiome Analysis, for discovering novel, selective antibiotic metabolites for BV with implications for charting the full pharmacologic potential of the vaginal microbiome.

microbiology↗

bioRxiv2025

HerbComb: an integrated database for the discovery of novel combinational therapies from herbal medicines

Drug combinations have gained increasing interest due to their potential for optimal efficacy and the ability to overcome drug resistance. Herbal medicine is a complex system with multiple ingredients and has illustrated apparent therapeutic effects through thousands of years of clinical experience, offering a rich resource for combinational therapies. However, the underlying mechanism of their synergistic therapeutic effects for disease treatment remains obscure for most herbal medicines. To fill this gap, we developed an integrated database named HerbComb (https://herbcomb.com/) to discover combinatorial therapies in herbal medicine, featuring three unique functions: drug prepositioning, customized combinatorial analysis, and biological and pharmacological features. Additionally, transcriptional gene signatures and ADMET properties were also interrogated as interactive functionalities to facilitate combinational analysis. Using Tongxinluo for Stroke as a case study, we identified and validated Oleanolic acid and Ferulic acid as synergistic ingredients. In summary, HerbComb is a versatile data exploration platform designed to characterize synergistic interactions among herbal medicines, enhancing the understanding of synergistic mechanisms and facilitating the discovery of effective drug combinations.

pharmacology and toxicology↗

bioRxiv2025

Population genomics of the endangered scaly-foot snail defines conservation units amid deep-sea mining threats in Indian Ocean vents

Hydrothermal vents in the Indian Ocean face imminent deep-sea mining, yet lack baseline high-resolution connectivity data for their endemic fauna. We present the first population genomic study of the endangered scaly-foot snail (Chrysomallon squamiferum), an iconic species distributed across three vent biogeographic provinces. Analysing 125 individuals from eight vent fields using 14 million single-nucleotide polymorphisms, we identify five genetic groups that warrant recognition as evolutionarily significant units. Demographic modelling reveals critical contributions from extinct or unsampled phantom populations to the contemporary genetic structure. Together with physical ocean modelling, we show that the deep currents drive asymmetric south-to-north gene flows, while transform faults act as dispersal barriers. We propose Longqi-Duanqiao fields on the Southwest Indian Ridge and Wocan field on the Carlsberg Ridge as two isolated populations prioritised for protection. Given its exceptional adaptations, wide distribution, and public recognition, the scaly-foot snail is well-positioned to serve as both a flagship and umbrella species to ensure the survival of a broader suite of vent biodiversity. As the International Seabed Authority finalises its Mining Code in 2025, our findings provide essential genomic evidence to mitigate the impacts of mining on genetic diversity and inform transboundary conservation strategies for this vulnerable ecosystem.

ecology↗

bioRxiv2025

DAXX governs the silencing of LINE1 during spermatogenesis in mice

The suppression of transposable elements in germline is critical for safeguarding fertility. Here we identify that the DAXX, a potent transcription repressor and an H3.3 chaperone, is indispensable for transposable elements silencing during spermatogenesis. Male mice with conditional knockout of Daxx in germ cells mediated by Ddx4-cre display delayed meiotic progression, deformed and reduced production of sperms, and an age-dependent decline in fertility. The ablation of DAXX results in activation of evolutionarily young LINE1 and ERVs subfamilies. Mechanistically, DAXX interacts with DMAP1 to maintain DNA methylation of LINE1 in meiotic cells. In summary, this study identifies DAXX as a previously unknown regulator of spermatogenesis and demonstrates that it functions as an epigenetic regulator to maintain the silencing the young LINE1 by DNA methylation.

developmental biology↗