Biology subjects

Zeng, H.

Publications and source records attributed to Zeng, H..

At least 19 recordsLinked to original sources

bioRxiv2026

Spatial-filtering nanoscopy for 40-nm label-free Raman imaging

Super-resolution fluorescence microscopy overcomes the optical diffraction limit and has significantly advanced our understanding of biological complexity within the framework of fluorescence labelling (1). Fluorescence labelling underpins this capability, enabling high photon budgets, superior signal contrast, and tuneable photophysical properties essential for diverse super-resolution modalities (2,3). In contrast, label-free Raman imaging offers intrinsic chemical specificity (4), supporting applications ranging from biomolecular fingerprinting to cell metabolic mapping (5-8) and histopathological tissue characterization (9,10). Despite its label-free advantage and chemical specificity, Raman imaging remains fundamentally limited in both spatial resolution and imaging contrast due to inherently low signal throughput and weak intrinsic Raman contrast (11-13). Here, we introduce spatial-filtering nanoscopy (SFN), a physics-driven super-resolution strategy that achieves resolution enhancement through targeted signal purification instead of signal amplification, offering a conceptually distinct pathway beyond the diffraction limit. SFN synergistically integrates a sub-millimetre microsphere lens (SMML) with a standard confocal Raman microscope, harnessing two complementary physical effects: (i) the photonic redistribution effect (PRE), which narrows the lateral excitation profile, and (ii) the three-dimensional spatial filtering (3D-SFE), which effectively suppresses both lateral and axial background. We demonstrate SFN-enabled super-resolution Raman imaging of silicon nanostructures, intact cells, and tissue sections, achieving an effective lateral resolution of approximately 40 nm, and chemically resolving subcellular features including organelles and pseudopodia without exogenous labels. These findings establish SFN as a broadly generalizable hardware-based super-resolution strategy, readily deployable on standard confocal platforms and extensible across diverse optical imaging modalities.

bioengineering↗

bioRxiv2026

Mural-VISTA: A tool for mural cell-vessel interaction assessment and multiscale single-cell topo-morphological analysis

Three-dimensional (3D) mural cell morphology is heterogeneous and coupled to vessel geometry, however, measurements from two-dimensional (2D) maximum intensity projections (MIP) obscure overlapping processes and cell-vessel contacts. Accordingly, we developed Mural-VISTA, a semi-automated Python workflow for mural cell-vessel interaction and single-cell topo-morphology analysis of reconstructed surface meshes. This workflow integrates mesh pretreatment, interactive centerline extraction, hierarchical segmentation of cell soma, main axis and secondary processes (branches), and extraction of 36 multiscale (cell process segment level, process level, and whole cell level) topo-morphological and vessel-referenced metrics. Mural-VISTA identified morphological changes in pericytes and vascular smooth muscle cells (vSMCs) with altered RhoA activity. Constitutive active RhoA (RhoA CA) over-expression reduced branch complexity and increased process alignment in both cell types, while increased whole-cell and branch solidity only in vSMCs. Dominant negative RhoA (RhoA DN) over-expression increased branch abundance and reduced branch solidity in pericytes but not vSMCs, suggesting cell-type specific effect of reduced RhoA activity. In conclusion, Mural-VISTA enables quantitative 3D profiling of mural cell architecture and its spatial relationship with the vessel.

bioinformatics↗

bioRxiv2026

Hypothalamic Cell Type Evolution across Vertebrates

The hypothalamus regulates vital functions and is highly conserved among vertebrates occupying diverse niches, but little is known about the evolution of its cell types. Using comparative single-cell transcriptomics across major vertebrate classes, we identified deeply conserved and clade-specific cell types. While homologous cell types express conserved transcription factors and neuropeptides, their regulatory linkages are often rewired across evolution. The paraventricular nucleus largely retains its neuropeptide expression and responds to dehydration with fine cellular resolution from amphibians to mammals, whereas lamina terminalis neurons are specific to terrestrial vertebrates and exhibit gene expression divergence in cell types activated by dehydration and heat. We propose that a conserved set of keystone cell types sustain essential functions across vertebrates, while flexible gene regulatory programs alongside numerous clade-specific cell types offer evolutionary fluidity.

evolutionary biology↗

bioRxiv2026

Blood-Mediated Opto-Acoustic Stimulation of Brain Cortex at Sub-millimeter Precision

Direct modulation of neural activity with high spatiotemporal precision is a cornerstone in experimental neuroscience. Here, we present a blood-mediated optoacoustic stimulation (BOAS) approach that utilizes blood as an endogenous transducer for brain stimulation. By delivering 532-nm nanosecond pulsed laser to the cortex, we demonstrate that the absorption of hemoglobin generates sufficient acoustic pressure to trigger neuronal activity. By integrating BOAS with calcium imaging in GCaMP6f-expressing mice, localized neuronal responses were observed. Quantitative analysis reveals that BOAS produces responses comparable to natural visual stimulation and is significantly more efficient than the photothermal stimulation. Furthermore, we show that the response is dose-dependent. At high energy doses, BOAS induces cortical spreading depression. Histological evaluation confirmed that the brain maintains tissue integrity even under these stimulation parameters. Together, this work establishes a versatile method for precise brain stimulation as an alternative method for stimulating neuron at cortex.

bioengineering↗

bioRxiv2026

Iterative co-creation of harmonized human and non-human primate cellular and structural ontologies and 3D common coordinate frameworks for the basal ganglia

A major goal of the BRAIN Initiative Cell Atlas Network (BICAN) is to create a suite of foundational reference cell atlases and associated standards for human and non-human primate brains. Central to this goal is the creation of cross-species harmonized cellular taxonomies and structural parcellations with formal ontologies that can be mapped into 3D reference frameworks bridging neuroimaging and cellular and histological resolutions. We describe here an iterative approach, focused initially on the basal ganglia, to co-create structural and cellular ontologies in human, macaque and marmoset brains, including a Harmonized Ontology of Mammalian Brain Anatomy (HOMBA), and to map and refine structural parcellations into neuroimaging-based common coordinate frameworks. These references provide the framework for documenting and mapping all experimental sampling in BICAN, allowing analyses of cellular and molecular variation as a function of topographic position, and enabling comparisons of cellular, molecular and neuroimaging-based functional variation within and between primate species. HighlightsO_LIA hierarchical Harmonized Ontology of Mammalian Brain Anatomy (HOMBA) covering 2348 structures C_LIO_LIHOMBA-annotated 3D common coordinate frameworks (CCFs) of the basal ganglia across species C_LIO_LIHistologically informed 3D parcellation/atlas of 280 human subcortical structures indexed by HOMBA C_LIO_LIMapping and integration of structural, cellular and functional data with HOMBA and CCFs C_LI

neuroscience↗

bioRxiv2026

Genetic background diversifies mutational paths but converges on membrane depolarization for bleomycin resistance in yeast

Drug resistance evolution is shaped by both selective agent and genetic background in which resistant mutations arise, but most experimental evolution studies rely on a single genetic background. Here we performed parallel experimental evolution across six genetically diverged Saccharomyces cerevisiae strains, propagating 48 replicate populations for [~]225 generations under the selection of DNA-damaging chemotherapeutic bleomycin. Over 70% of the evolved populations acquired resistance with the magnitude and repeatability of adaptation varying markedly across genetic backgrounds. Whole-genome sequencing showed that bleomycin elevated single-nucleotide and insertion-deletion mutation rates, and induced background-specific aneuploidies and structural variants, including Ty-element-mediated translocations and tandem duplications. Notably, mutation rates varied up to 10-fold among backgrounds, with the most commonly-used lab strain exhibiting the highest mutagenic susceptibility, underscoring that evolutionary insights derived from single-background models may not universally generalize across a species. Despite this background-dependent genomic heterogeneity, we identified recurrent loss-of-function mutations in PMA1 and PTK2 that arose independently across nearly all genetic backgrounds and conferred strong resistance. Functional and biophysical assays revealed that these mutations impair Pma1p H+-ATPase activity and depolarize the plasma membrane, thereby limiting the cellular uptake of the positively charged drug. Our findings demonstrate that while the mutational routes to drug resistance are heavily shaped by genetic background, a convergent, background-independent mechanism centered on membrane potential regulation can dominate phenotypic adaptation. This work underscores the necessity of integrating genetic diversity into experimental evolution paradigm and nominates plasma membrane potential as an important lever for modulating chemotherapeutic efficacy.

genetics↗

bioRxiv2026

Reconstructing whole-organism cell phylogenies with resolved ancestral transcriptional states

Combining cell lineage tracing with single-cell RNA sequencing can reconstruct a phylogenetic tree to integrate single-cell transcriptomic atlas. However, the internal nodes of this tree - representing ancestral cells - remain transcriptionally silent, preventing along-lineage longitudinal tracing of cell state dynamics. Here, we overcome this by reconstructing high-resolution zygote-to-larva developmental cell phylogenies for 15 zebrafish larvae, with directly measured transcriptomes of the terminal nodes (i.e., sampled cells). Leveraging a set of lineage-committed upregulated genes (LUGs), we developed LUG-encoded ancestral projection (LEAP), a novel phylogeny-based computational framework, and successfully imputed the transcriptional states of internal nodes of the phylogenies. This enabled, for the first time in a non-nematode organism, lineage-informed longitudinal analysis of cell state dynamics throughout development. Our analysis revealed a major, previously unappreciated wave of fate specializations associated with hatching, distinct from the well-characterized events of gastrulation. Furthermore, we uncovered abundant incipient cell states that are already fate-determined but exhibit minimal transcriptional differentiation, revealing a hidden layer of developmental fate specializations. In sum, by resolving ancestral transcriptional states of a reconstructed cell phylogeny, this work paves the way for constructing lineage-resolved cell atlases in complex organisms to characterize comprehensive cell state dynamics.

developmental biology↗

bioRxiv2026

Direct detection of alternative DNA conformations with long-read sequencing and machine learning approaches

Progress has been made in identifying G-quadruplexes (G4s) and other non-canonical (non-B) DNA structures in live cells. However, these experiments have been limited by methodological constraints, including low resolution and specificity, and GC-sequencing bias inherent to the short-read sequencing technologies. Direct, single-molecule, and long-read technologies have the potential to address these shortcomings. Here, we investigated the use of long-read DNA sequencing with Oxford Nanopore Technologies (ONT) to detect G4 and other non-B DNA structures. We applied ONT sequencing to oligos known to form G4s in vitro, as well as to DNA from live cells. Both experiments were probed with potassium permanganate, a chemical that preferentially oxidizes single-stranded DNA (ssDNA). We determined that G4 structures can be resolved by ONT sequencing, and a G4 motifs sequencing error profile depends on the original structural state of the G4 (as determined in vitro). We next applied machine learning algorithms (logistic regression, random forest, XGBoost, and 1D convolutional neural networks) to the raw ONT sequencing current data. Using these approaches, we could determine whether a sequencing read originated from a G4 motif in the G4 vs. B-form conformation with [~]90% accuracy. Further, we demonstrated that it is possible to sequence permanganate-modified DNA directly using ONT, with minimal effects on the read length, yield, and alignment accuracy. We conclude that at high ONT sequencing depth, this approach can identify B-folded vs. single-stranded DNA regions in cells, with ssDNA frequently present across several non-B DNA structures.

bioinformatics↗

bioRxiv2026

Conserved Cell Type Signatures Across the Brainstem and Spinal Cord in the Mouse Central Nervous System

Understanding how cell types are organized across the central nervous system (CNS) is key to uncovering neural function. Here, we integrate single-nucleus Multiome (RNA+ATAC) sequencing, spatial transcriptomics, and computational analyses to map conserved cell type signatures in the adult mouse brainstem and spinal cord. We identify a shared core of neuronal and non-neuronal cell types, alongside region-specific specializations reflecting distinct functions. Spatial data reveal conserved cellular niches across the brainstem-spinal cord boundary, indicating a continuous organizational logic. Cross-region comparisons uncover recurrent gene expression modules and signaling programs that may support shared circuit features. Chromatin accessibility profiling highlights cell-type-specific regulatory programs and implicates Hox transcription factors in positional identity. Notably, cell-type and positional identities are largely orthogonal, with varying regional influence across neuronal classes: motor neurons show strong positional coupling, whereas glutamatergic and GABAergic interneurons show minimal entrainment. This work provides a reference for the shared molecular architecture of these CNS regions.

neuroscience↗

bioRxiv2026

Mural cell contractility regulates vessel diameter by controlling cell morphology and vessel coverage

Mural cells are key regulators of vascular architecture, yet how their contractility and morphology jointly influence vessel diameter in vivo remains poorly understood. Here, using high-resolution live imaging and single-cell morphometric analysis in zebrafish, we show that vascular smooth muscle cells (vSMCs) and pericytes undergo a developmental reduction in cell size through the progressive retraction of actin-rich primary and secondary processes during vascular remodelling. By specifically manipulating RhoA activity to alter mural cell contractility and shape, we uncover vessel-specific roles for mural cells in diameter regulation. We find that vSMC contractility is dispensable for the initial constriction of the dorsal aorta but is required to stabilise its diameter following vasoconstriction and to maintain vascular tone. In contrast, pericyte contractility is dispensable for both the constriction and stabilisation of intersegmental vessels. In the brain vasculature, vessel diameter is governed by the balance between contractile force and the extent of vessel coverage by vSMCs. Together, our findings redefine the role of mural cell contractility in the control of vessel constraint, demonstrating that morphology and vessel coverage, rather than contractility alone, are key determinants of vascular diameter in vivo.

cell biology↗

bioRxiv2026

Transient suppression of the ECM1 gatekeeper is essential for HGF/c-MET-driven liver regeneration

Hepatocyte growth factor (HGF) is a multifunctional cytokine stored in the extracellular matrix as an inactive precursor and is essential for tissue repair. How HGF activity is dynamically regulated during regeneration remains unclear. Here, we identify extracellular matrix protein 1 (ECM1) as a physiological inhibitor of active HGF during liver regeneration. Following 70% partial hepatectomy, active HGF rapidly increases in parallel with a sharp decline in ECM1, and preventing this downregulation delays liver mass recovery. Mechanistically, ECM1 directly binds the active HGF -subunit through a mechanism dependent on residue R392, thereby suppressing c-MET-ERK-MYC signaling and hepatocyte proliferation. Loss of ECM1 permits activation of this pathway, whereas MYC overexpression rescues ECM1-mediated growth inhibition. In patients, proliferative hepatocytes localize to ECM1-negative regions. Supported by transcriptomic analyses and computational modeling, these findings identify ECM1 as an extracellular gatekeeper whose transient downregulation enables HGF-driven tissue repair. HighlightsO_LIECM1 is temporally downregulated during liver regeneration, while forced ECM1 expression delays liver mass recovery after 70% PHx by suppressing hepatocyte proliferation. C_LIO_LIRapid and transient ECM1 downregulation permits HGF-c-Met-ERK-MYC signaling to drive hepatocyte cell-cycle entry and expansion. C_LIO_LIECM1 directly binds the active HGF -subunit with residue R392 playing a critical role, revealing a novel extracellular mechanism for growth factor inhibition. C_LIO_LILoss of ECM1 marks proliferative hepatocytes in human liver repair, identifying ECM1 as a tunable regulator of regenerative capacity. C_LI

cell biology↗

bioRxiv2026

ACLY integrates metabolism and chromatin accessibility to enable B Cell activation and humoral immunity

B cell activation and differentiation into antibody-secreting cells require extensive metabolic and epigenetic remodeling, yet the molecular mechanisms that integrate these programs remain incompletely understood. ATP-citrate lyase (ACLY) links glucose metabolism to acetyl-CoA production, supporting lipid biosynthesis and protein acetylation. However, its role in humoral immunity has not been fully defined. Here, using genetic and integrated multi-omics approaches, we show that B cell activation is accompanied by coordinated metabolic, transcriptional, and epigenetic reprograming. Although ACLY is dispensable for B cell development and homeostasis, it is required to establish chromatin accessibility programs in activated B cells, with a more pronounced impact on the epigenetic landscape than on transcriptional output. ACLY-deficient B cells exhibit profound defects in TLR and BCR elicited activation, survival and metabolic fitness ex vivo. In vivo, B cell-intrinsic loss of ACLY results in impaired antigen-specific antibody production, associated with reduced germinal center and plasmablast formation, but normal homeostatic proliferation. Deletion of ACLY after B cell activation reduces plasmablast generation in vivo, indicating a continued requirement for ACLY beyond the initial activation phase. Together, these findings identify ACLY as a central regulator that links metabolism to epigenetic programing that supports B cell activation and humoral immunity.

immunology↗

bioRxiv2026

Targeting cancer-associated cell surface RNAs with oligonucleotide-drug conjugates enables broad antitumor activity

Recent studies have revealed the presence of RNAs on mammalian cell surfaces, yet the linkage of cell surface RNAs (csRNAs) to cellular states and their potential as extracellular accessible drug targets remains unexplored. Here we develop Cell-surface and Intracellular RNAs Co-mapping (CIRCmap), a highly multiplexed in situ profiling approach that simultaneously detects thousands of cell-surface and intracellular RNAs at single-cell resolution. Using CIRCmap, we uncover that csRNA distributions are correlated with cellular states and identified cancer-associated csRNAs. Integrative analysis of cell-surface and intracellular RNAs within the same cells implies that the csRNAs undergo endocytosis and endolysosomal trafficking, which is further supported by perturbation experiments and co-localization visualization. We design oligonucleotide-drug conjugates (ODCs) that target cancer-associated csRNAs for endocytic delivery of cytotoxic payloads selectively to cancer cells. ODCs exhibit broad antitumor activity in cell lines and an in vivo mouse tumor model, opening a promising new avenue for targeted cancer therapy.

cell biology↗

bioRxiv2026

Raster photostimulation of large-scale neural populations

Neural computations are implemented by distributed neural populations that typically span multiple brain areas. Simultaneous recording and photo-activation experiments can reveal the structure and logic of these neural computations, but such methods are typically limited to small subsets of neurons in restricted fields of view. Here we describe a new system called raster photostimulation for activating and recording thousands of neurons, over a short 300 ms time window and over a large 5 mm field-of-view on a two-photon mesoscope. The photo-activation is precisely matched to the neural recording configuration, as it uses the same optical path, although with a different laser that is independently gated. We demonstrate pixel-level precision, frame-by-frame mask updating, and single-frame photostimulation of thousands of neurons. While this method lacks the precise temporal control of alternative methods, it compensates with ease-of-use, spatial precision, cost of implementation and by pushing the limits on the number of near-simultaneously stimulated neurons.

neuroscience↗

bioRxiv2026

Lesion-Level-Dependent Neuroendocrine Surge Precedes Neuroinflammation and Endoplasmic Reticulum Stress in the Hypothalamus After Spinal Cord Injury: Dual-Cohort Transcriptomic Evidence for a Temporally Ordered AVP Cascade

BackgroundSpinal cord injury (SCI) triggers remote pathological changes in supraspinal regions, including neuroendocrine dysfunction that manifests clinically as hyponatremia and central diabetes insipidus. Clinical observations of lesion-level dependency and sequential transformation between these disorders suggest a temporally ordered hypothalamic cascade in which a compensatory arginine vasopressin (AVP)-driven neuroendocrine surge may precede a later neuroinflammation and endoplasmic reticulum (ER) stress-mediated neuronal exhaustion. Direct transcriptomic evidence for the temporal ordering of these events, however, has been lacking. MethodsWe performed a dual-cohort transcriptomic analysis. A discovery cohort (NCBI Sequence Read Archive PRJNA953752) comprised hypothalamic tissue from adult male Sprague-Dawley rats subjected to high-thoracic (T3) SCI, low-thoracic (T10) SCI, or sham surgery, sampled at post-injury day 7 and analyzed with edgeR/DESeq2 (|log2FC| > 1, Padj < 0.05). An independent chronic-phase validation cohort (Gene Expression Omnibus GSE297887) of hippocampal tissue from SCI and sham mice was interrogated as a sensitive supraspinal proxy for remote neuroinflammatory and ER-stress signatures. Pre-defined gene panels covered neuroendocrine, neuroinflammation, and ER-stress/unfolded-protein-response categories. ResultsIn the discovery cohort, high-thoracic SCI produced a lesion-level-dependent neuroendocrine surge in the hypothalamus: Avp (fold change 7.23; Padj = 0.002), Oxt (fold change 14.25; Padj = 2.3 x 10-7), and Ucn3 (fold change 9.22; Padj = 0.002) were among the most significantly upregulated genes genome-wide, whereas low-thoracic SCI failed to reach significance for any of these targets. Classical neuroinflammation markers and canonical ER-stress effectors remained transcriptionally silent (all Padj > 0.69). The PERK-pathway sentinel genes Trib3 and Ppp1r15a/GADD34 exhibited coordinated sub-threshold trends indicative of early activation, and Avp expression was tightly correlated with Mmp9 (r = 0.833; P = 0.0004). In the chronic-phase validation cohort, microglial P2ry12 and ferroptosis signatures were significantly upregulated (P2ry12 fold change 1.33; P = 0.008) suggesting a primed microglial state, while ER-stress effectors remained silent. ConclusionsThese data support a temporally ordered hypothalamic cascade after SCI in which an early compensatory neuroendocrine surge precedes -- and may precipitate, through biosynthetic overload and blood-brain-barrier disruption -- a subsequent neuroinflammation and ER-stress crisis. The defined molecular window between neuroendocrine activation and inflammatory/ER-stress engagement identifies a candidate therapeutic window for early neuroprotective intervention in acute SCI.

neuroscience↗

bioRxiv2026

A Community Standard Multispecies Cell Atlas of the Basal Ganglia

The NIH BRAIN Initiative Cell Atlas Network (BICAN) aims to generate a standardized, integrated cell atlas of the human, macaque, marmoset, and mouse brain that serves as a foundational community reference for the classification and study of brain cell types. Here we present the first major component of this effort: a cross-species, multimodal atlas of the basal ganglia, a group of subcortical nuclei central to motor control and implicated in a broad range of neurological disorders. Grounded in large-scale single-cell transcriptomic classification and integrated with epigenomic and spatial genomic modalities, this resource is enabled by coordinated cross-species sampling and harmonized analytical frameworks. It provides extensive phenotypic characterization of cell types, incorporates community-informed annotation, and establishes a highly curated, data-driven taxonomy with standardized nomenclature. The atlas is anchored to species-specific anatomical reference frameworks and linked across species through unified structural ontologies, enabling consistent cross-species comparisons. Multiple complementary datasets are mapped to this reference, including multiomic profiles and developmental trajectories aligned to adult cell states. Realization of this resource has required coordinated standards for tissue processing across human and model organisms, harmonization of donor metadata across brain banks, and the development of unified anatomical reference systems. To support these advances, BICAN has established an integrated ecosystem comprising standardized sequencing pipelines, neuroanatomically grounded data infrastructure, scalable visualization and mapping tools, and interoperable metadata standards. Analogous to the standardization achieved in genome science, this ecosystem provides a FAIR (findable, accessible, interoperable, and reusable) framework that enables researchers to map, compare, and interpret diverse datasets against a shared reference and associated knowledge base. The BICAN reference system is now being extended to the whole brain, with principles that are readily generalizable to other organ systems.

neuroscience↗

bioRxiv2026

The External Globus Pallidus is a Basal Ganglia Output Hub with Action-Specific Circuits

The external globus pallidus (GPe) is traditionally viewed as a homogeneous relay in the indirect basal ganglia pathway that broadly suppresses movement. Using whole-brain anterograde axon mapping, rabies tracing, single-neuron reconstruction, and single-nucleus RNA sequencing, we reveal that the GPe is a major basal ganglia output nucleus, composed of anatomically and molecularly distinct populations. Besides the neurons with canonical projections, and projections to striatum and cortex, the GPe contains specific neuronal populations that target the thalamus and brainstem directly, including the parafascicular thalamus (GPePf) and the pedunculopontine nucleus (GPePPN). These projection-defined subpopulations exhibit distinct behavioral functions. GPe-PPN neurons are selectively suppressed at locomotor onset, and bidirectional manipulation of this pathway is sufficient to promote or suppress locomotion. Notably, D2-MSN stimulation upstream of these neurons evokes locomotion, and co-activation of GPePPN pathway blocks this D2-MSN-evoked locomotion, demonstrating a disinhibitory circuit for action stemming from D2-MSNs. In contrast, GPePf neurons are not engaged during locomotion but are selectively recruited during skilled forelimb actions, and their activation disrupts forelimb movements without affecting locomotion. Together, these findings establish the GPe as a basal ganglia output hub composed of modules that mediate distinct behaviors. This organization revises canonical models of the indirect pathway by demonstrating that D2-MSNs can also facilitate, rather than only suppress, movement depending on the downstream GPe output channels they engage. In briefThe external globus pallidus (GPe), classically considered a relatively homogeneous relay within the indirect basal ganglia pathway, is revealed here as a major basal ganglia output hub. Distinct GPe subpopulations, defined by transcriptomic identity and projection target, form parallel channels to cortical, thalamic and brainstem structures with action-specific functions. These findings uncover a disinhibitory mechanism through which D2-MSNs can facilitate behavior, revising canonical models of basal ganglia function. HighlightsBrain-wide mapping reveals the GPe as a major basal ganglia output node GPe contains novel cell types defined by transcriptomic identity and projection specificity Projection-specific GPe circuits differentially control locomotion and skilled forelimb actions D2-MSNs can disinhibit downstream targets via specific GPe projections and promote action.

neuroscience↗

bioRxiv2026

Deep-Plant: a supervised foundation model for plant regulatory genomics

Large-scale sequence-to-function deep learning models have demonstrated unparalleled ability to model biological sequences and have revolutionized the field of regulatory genomics. However, the majority of such efforts have centered on human and mammalian systems, leaving plant regulatory genomics comparatively underexplored. To address this gap, we introduce DO_SCPLOWEEPC_SCPLOW-PO_SCPLOWLANTC_SCPLOW, a supervised foundation model trained to predict chromatin state directly from genomic sequence. In contrast to large language models, which are trained in a selfsupervised manner using sequence alone, our model is trained to predict chromatin state across tissues and conditions. Training the model on a large collection of genome-wide experiments including DNA accessibility, transcription factor binding, and histone modifications, provides it with added biological context beyond the sequence itself. We demonstrate that the resulting model is an effective platform for developing accurate models of regulatory activity relevant to gene expression and active enhancers, exhibiting large improvements in speed, accuracy, and interpretability over the complementary approach of fine-tuning DNA language models. DO_SCPLOWEEPC_SCPLOW-PO_SCPLOWLANTC_SCPLOW models are available in Arabidopsis and rice, and work well as a building block for sequence modeling in related species such as corn. Together, these results establish supervised, chromatin-informed foundation models as a practical and effective paradigm for regulatory sequence modeling in plants.

genomics↗