Biology subjects

Lee, J.

Publications and source records attributed to Lee, J..

At least 199 records · Page 11Linked to original sources

bioRxiv2025

Molecular and cellular dynamics of measurable residual disease progression in myelodysplastic syndromes

Cancer relapse after treatment invariably proceeds from the persistence and progression of measurable residual disease (MRD), an ultrasmall malignant population. Despite the poor prognostic impact established by increasingly sensitive MRD detection methods, the molecular pathways defining MRD remain unknown. To identify the unique features of MRD and the molecular forces shaping its progression, we performed single-cell multi-omic profiling on DNA, RNA and protein layers for a longitudinal patient cohort with relapsed myelodysplastic syndromes (MDS) after stem cell transplantation (SCT), the only curative modality for MDS. We provide a comprehensive molecular portrait of MRD cells with novel markers, shared across genetically heterogeneous patients. MDS relapse after SCT manifested universally with marked phenotypic evolution. Genotype and phenotype analyses revealed MRD progression as a dynamic, evolutionary process rather than a static expansionary one, driven by both subclonal sweeping and cell state transitions. Malignant cells adapted to infiltrating T cells by rewiring IFN-{gamma} responses to activate a key immunoevasive pathway. Our study demonstrates the power of longitudinal, single-cell multi-omic analysis for identifying, tracking, and understanding MRD cells, opening new avenues to target MRD persistence and progression.

cancer biology↗

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

DNMT3A-R882 mutations intrinsically drive dysfunctional neutropoiesis from human haematopoietic stem cells

Clonal haematopoiesis (CH) arises from the expansion of hematopoietic stem cells (HSCs) carrying leukaemia-associated somatic mutations. CH is linked to pathological immune dysregulation and a greater risk of age-related inflammatory diseases. Yet, how CH mutations impact HSC differentiation into immune effector cells remains understudied. Here, we report a single-cell resolution functional and multi-omic investigation of HSC clonal and differentiation dynamics in individuals with DNMT3A-R882 CH. DNMT3A-R882 reshapes the clonal architecture of haematopoiesis towards an aged phylogenetic structure. Functionally, DNMT3A-R882 HSCs produce decreased monocytic output but more abundant and mature neutrophil progeny compared to WT HSCs in the same individual. Whereas DNMT3A-R882 myeloid progenitors display attenuated inflammatory transcriptional programmes, DNMT3A- R882 mature neutrophils acquire proinflammatory and immunomodulatory features typical of maladaptive immunity and CH co-morbidities. Our findings, validated in humanised mice, identify aberrant DNMT3A-R882 HSC-driven neutropoiesis as a key link between CH, immune dysregulation and risk of inflammatory disease.

cell biology↗

bioRxiv2025

Characterization of the Cell Division-Associated Peptidoglycan Amidase AmiA of Chlamydia trachomatis

Chlamydia is an obligate intracellular bacterium that differentiates between infectious, non-dividing EBs and non-infectious, dividing RBs. Pathogenic Chlamydia species are unusual in lacking a peptidoglycan sacculus, yet they do synthesize a transient and localized peptidoglycan structure at the divisome of the RB during their polarized division process. Although several studies have described the components of the chlamydial divisome necessary to generate peptidoglycan at a specific site on the membrane, less is understood about how the peptidoglycan structure is degraded to allow for the daughter cell to form and the division process to complete. Amidases are key components of the cell wall in model system bacteria as they catalyze the degradation and remodeling of peptidoglycan, including in the division septum. Here, we characterized the cell division-associated amidase, AmiA_Ct, of Chlamydia trachomatis both in vitro and in vivo. Our in vitro data show that AmiA_Ct is a bona fide, metal-dependent amidase capable of cleaving peptidoglycan. AmiA_Ct complemented an E. coli amidase deficient strain and supported the growth and separation of daughter cells. To assess the function of AmiA_Ct in C. trachomatis, we generated a transformant strain carrying an inducible CRISPR interference system targeting the amiA gene. Knocking down expression of amiA resulted in altered bacterial morphology, a reduction in infectious EBs, and the accumulation of peptidoglycan in the organisms. These data indicate a critical function for AmiA_Ct in the unique cell division process of Chlamydia. ImportancePeptidoglycan is an important structural cell wall polymer that serves to give bacteria their shape and resistance to changes in extracellular solute concentrations. For Chlamydia trachomatis, an obligate intracellular pathogen that divides within a host cell, peptidoglycan is only used for cell division and is not a component of its cell wall. In this study, we characterize the function of a chlamydial amidase that helps degrade peptidoglycan during cell division. We show a critical function for amidase activity in facilitating changes to the peptidoglycan structure during chlamydial cell division that support normal growth and development of this pathogenic bacterium.

microbiology↗

bioRxiv2025

Clonal dynamics of monozygotic twinning in early human embryogenesis

Monozygotic twins are derived from the split of a single zygote early in embryogenesis. Although it was hypothesized that the timing of twining is overall associated with fetal membrane configuration of twins, i.e., chorionicity and amnionicity, our understanding of early embryonic clonal dynamics underlying human twinning is limited. Here we explored the segregations of early embryonic lineages in 7 dichorionic diamniotic (DCDA), 7 monochorionic diamniotic (MCDA), 8 monochorionic monoamniotic (MCMA) monozygotic twins, and 1 dichorionic triamniotic (DCTA) monozygotic triplets, using post-zygotic early embryonic mutations (EEMs) as endogenous lineage barcodes. Patterns of the early lineage distributions among monozygotic twins revealed three apparent clonal categories, referred to as para-identical, sub-identical, and full-identical twins, which largely correlated with the amnionicity of the twins. Rather, despite conventional wisdom, chorionicity was not substantially associated with early clonal compositions, but with blood exchanges in utero. In sub-identical twins, where one co-twin was clonally a part of the other, our data suggested that the foundation of the latter co-twin was established after acquisition of a median of 6 additional post-zygotic mutations (range: 2-13), corresponding to [~]5 early cell divisions. Additional in-depth analysis on the matched placenta from an MCDA twin suggested that separation of two co-twins can precede the separation of the placenta and embryonic proper, and a single chorion can be formed even with multiclonal origin. Our findings provide insights into the clonal dynamics, twinning processes, and cell fate decisions in early human embryogenesis.

genomics↗

bioRxiv2025

ETV4 restrains the formation of CD74+ cytotoxic exhausted CD8+ T cells

Exhausted CD8+ T (Tex) cells comprise heterogeneous subsets with varying cytotoxic and dysfunctional properties, making inhibitory receptor profiles unreliable indicators of function. Therefore, it is essential to define transcriptional regulators or markers that accurately reflect these states. Here, we identified ETS variant 4 (ETV4) as a transcription factor that restricts cytotoxic Tex differentiation within tumors. ETV4 deficiency promotes the emergence of a CD74+ cytotoxic Tex subset while preserving the progenitor Tex (Tpex) pool. CD74 serves as both a marker and regulator, reinforcing Tpex features and enhancing Tex cytotoxicity. Mechanistically, ETV4 loss derepresses T-bet expression, thereby increasing IFN{gamma} production that induces CD74 expression in a paracrine manner. Single-cell analyses across human cancers revealed that CD74+ Tex cells are enriched for proliferative and metabolic programs and correlate with improved survival. Collectively, we establish the ETV4-CD74 axis as a regulator of Tex fate and promising therapeutic target to augment anti-tumor immunity.

immunology↗

bioRxiv2025

AGO1 in neural progenitor cells orchestrates brain development and sociability via LIN28A-REELIN axis

AGO1, an essential RNA binding protein (RBP) in RNA interference, is associated with autism spectrum disorder (ASD). However, the precise functions of AGO1 in brain development and related disorders remain largely unexplored. Here, we report the critical roles of AGO1 in neural progenitor cells (NPCs) in promoting sociability and shaping the structure of the developing brain. Ago1 knockout (KO) in the mouse brain leads to hyposociability, a characteristic symptom of ASD. In human forebrain organoids, AGO1 KO disrupts the formation of ventricle-like structures and delays cortical layer development. We discovered that AGO1 KO results in a loss of polarity in NPCs, which subsequently reduces neuronal development. These AGO1 KO phenotypes originate from the failure to suppress LIN28A in NPCs. AGO1 is primarily localized in the nucleus within NPCs and binds to the LIN28A promoter region, thereby inhibiting LIN28A transcription. We found that increased LIN28A in AGO1 KO NPCs reduces REELIN expression, a key regulator of brain development, by binding directly to the REELIN mRNA. The aberrant polarity phenotype in AGO1 KO NPCs was successfully ameliorated by either LIN28A knockdown or recombinant REELIN treatment. Collectively, our findings elucidate the intricate molecular mechanisms orchestrated by nuclear AGO1 in NPCs and underscores its pivotal roles in brain development, providing significant insights into ASD.

neuroscience↗

bioRxiv2025

Circadian phosphotimer generates time cues through PER dimerization-mediated trans-phosphorylation

The circadian clock generates autonomous molecular rhythms that regulate sleep cycles, chronotypes, daily physiology, and expression patterns of almost the entire genome. Temporal phosphorylation of PERIOD (PER) proteins creates a phosphotimer that is critical for clock function. Although previous studies show that PER stably binds the casein kinase CK1, cis-acting phosphorylation of PER by CK1 is not compatible with slow PER phosphorylation extended over 24 hrs. So how can PER phosphorylation be programmed in a slow and controlled manner? Here, we show that the timing cues are encoded by PER-PER dimerization-mediated trans-phosphorylation, which enables the necessary time delay in phosphorylation and slow phosphorylation. When PER dimerization is disrupted, PER phosphorylation and circadian rhythms are severely compromised. In mouse models with point mutations in the PER dimerization domain, circadian period is shortened to [~]20 hrs, and the phase of wake/sleep cycles is dramatically advanced, switching a nocturnal animal to a half diurnal animal.

neuroscience↗

bioRxiv2025

Life-history evolution under artificial selection in a clonal plant

The response of natural populations to selection and the role of genetic correlations in constraining or facilitating evolutionary change is fundamental to adaptation. We use artificial selection to investigate the evolutionary response of clonal reproduction in the common monkeyflower (Mimulus guttatus), a species with extensive life history variation. We first characterize the standing genetic variation in a single perennial population, and then conduct four generations of divergent artificial selection on stolon number--the mechanism of clonal reproduction in this species. To start, stolon number had moderate heritability (H{superscript 2}=0.25) and was negatively genetically correlated with reproductive traits. Artificial selection produced a clear but asymmetrical response. High selection lines made significantly more stolons, while low lines diverged less from controls. Analyses of G matrices revealed that selection not only changed trait means but also genetic correlations, with high lines diverging more in multivariate genetic architecture. Our results demonstrate that single populations harbor sufficient genetic variation to respond rapidly to selection on clonality, and the response is shaped by existing patterns of genetic covariation. The capacity for rapid evolution of clonal traits is particularly relevant as climate change alters selection and shifts the relative advantages of sexual versus clonal life-history strategies.

evolutionary biology↗

bioRxiv2025

Why early birds sing early: rebound singing from nighttime suppression can explain dawn chorus

Many birds sing vigorously at dawn, a phenomenon widely known as the "dawn chorus." While its adaptive functions have been extensively studied, the proximate mechanisms underlying this widespread phenomenon have remained largely unexplored. Here, we demonstrate that intense morning singing resembling the dawn chorus is evoked as a rebound from singing suppression prior to dawn in captive zebra finches. Our findings indicate that when birds become hormonally aroused in darkness well before dawn, the suppression of spontaneous singing by darkness enhances their intrinsic motivation to sing. This elevated motivation subsequently triggers a rebound of intense singing immediately upon an increase in ambient light. These results reveal mechanisms by which intense morning singing arises from the interaction between the birds arousal state and ambient light, providing a plausible mechanistic explanation for the dawn chorus. Additionally, we found that rebound morning singing accelerates morning changes in song structure, supporting the existing hypothesis that the dawn chorus functions as a vocal warm-up to rapidly optimize song performance following nighttime silence. Together, our findings provide both mechanistic and functional insights into the dawn chorus and other diel vocal patterns in songbirds.

animal behavior and cognition↗

bioRxiv2025

Effects of Cholesterol Modulation on Cisplatin-Induced Hearing Loss

Cisplatin is a widely used and effective anticancer drug. However, it causes permanent sensorineural hearing loss in over 50% of treated patients. There are no FDA-approved therapies to prevent cisplatin-induced hearing loss (CIHL) in adults, highlighting a critical unmet clinical need. Previous studies suggest that statins, commonly prescribed cholesterol-lowering drugs, are associated with reduced incidence and severity of cisplatin-induced hearing loss. Statins are primarily used to lower cholesterol, but they also exert several pleiotropic effects, making the mechanism(s) underlying this protection unclear. Here we examine whether reduced plasma cholesterol confers protection against CIHL independent of statin treatment. We utilized mice lacking serine protease proprotein convertase subtilisin/kexin type 9 (Pcsk9 knockout (KO) mice) as a genetic model of reduced plasma cholesterol. We find that Pcsk9 KO mice are protected against cisplatin-induced hearing loss, as reflected by significantly lower ABR and DPOAE threshold shifts relative to wild type (WT) mice following treatment. Histological analyses confirmed preservation of cochlear outer hair cells (OHCs) in Pcsk9 KO mice treated with cisplatin, whereas WT mice showed significant OHC loss in the high-frequency cochlear regions. Finally, hearing loss positively correlated with baseline plasma cholesterol levels. Together our data demonstrate that systemic cholesterol reduction provides significant protection against CIHL, and they suggest that the protective effect of statins against CIHL is mediated by cholesterol reduction.

neuroscience↗

bioRxiv2025

Neuroanatomical Correlates of Negative Symptoms in Schizophrenia

BackgroundSchizophrenia is characterized by widespread structural brain abnormalities, but associations between structural abnormalities and negative symptom severity are not well understood. Negative symptoms have been conceptualized in a hierarchical structure of two second-order dimensions--motivation and pleasure (MAP) and expression (EXP)--and five first-order domains: anhedonia, avolition, and asociality (MAP), and blunted affect and alogia (EXP). A better understanding of the neural circuitry underlying negative symptom dimensions and domains is important given their reported association with poor functional outcome and lack of available treatments. Study DesignThe meta-analysis included 1,591 individuals with schizophrenia across 16 samples with structural imaging and Scale for Assessment of Negative Symptoms data. The study generated correlations of cortical thickness and subcortical volumes with the negative symptom dimensions and domains. Study resultsNegative symptoms showed mainly negative associations with cortical thickness and subcortical volumes. The effect sizes were small but there was a pattern of associations in predominantly frontal lobe cortical thickness and limbic subcortical volumes. The regional correlation patterns of cortical thickness and subcortical volumes with symptom domains support the conceptualized hierarchical structure of negative symptoms: correlations of MAP domains were stronger with the MAP than EXP dimension, and vice versa. Exploratory analyses with receptor densities further supported the hierarchy. ConclusionOur findings reveal small but consistent associations between negative symptom dimensions and predominantly prefrontal region cortical thickness, and limbic region volumes. These findings advance our understanding of the network of anatomical regions that may contribute to the severity of negative symptoms in schizophrenia.

neuroscience↗

bioRxiv2025

MAS cryoprobe enhances solid-state NMR signals of α-synuclein fibrils

Solid-state NMR spectroscopy is increasingly applied to structural and dynamics studies across a broad range of chemical, material, and biological systems. Although sensitivity has traditionally been a major limitation, the recently developed MAS cryoprobe has been shown to substantially overcome this challenge. Its ability to enhance the signal-to-noise (S/N) ratio without requiring sample freezing makes it particularly attractive for investigating non-isotropic systems, including soft materials (e.g., hydrogels), semi-solids (e.g., membrane mimetics) and rigid solids (e.g., amyloid fibrils). In this study, we report on the enhanced sensitivity of solid-state NMR experiments on -synuclein fibrils using a MAS cryoprobe. Nearly an order-of-magnitude improvement in S/N was observed in CPMAS, refocused-INEPT and 2D 13C-13C chemical shift correlation spectra of -synuclein fibrils compared with data collected on a conventional MAS probe. The improved S/N enables the acquisition of slowly decaying signals in the indirect dimension, facilitating faster, high-resolution multidimensional solid-state NMR spectroscopy. We therefore anticipate that MAS cryoprobe will become increasingly valuable for structural studies a wide range of samples that are less abundant, less stable, or transient, such as amyloid intermediates.

biophysics↗

bioRxiv2025

Mapping transcriptional responses to cellular perturbation dictionaries with RNA fingerprinting

Single-cell perturbation dictionaries provide systematic measurements of how cells respond to genetic and chemical perturbations, and create the opportunity to assign causal interpretations to observational data. Here, we introduce RNA fingerprinting, a statistical framework that maps transcriptional responses from new experiments onto reference perturbation dictionaries. RNA fingerprinting learns denoised perturbation "fingerprints" from single-cell data, then probabilisti-cally assigns query cells to one or more candidate perturbations while accounting for uncertainty. We benchmark our method across ground-truth datasets, demonstrating accurate assignments at single-cell resolution, scalability to genome-wide screens, and the ability to resolve combinatorial perturbations. We demonstrate its broad utility across diverse biological settings: identifying context-specific regulators of p53 under ribosomal stress, characterizing drug mechanisms of action and dose-dependent off-target effects, and uncovering cytokine-driven B cell heterogeneity during secondary influenza infection in vivo. Together, these results establish RNA fingerprinting as a versatile framework for interpreting single-cell datasets by linking cellular states to the underlying perturbations which generated them.

genomics↗

bioRxiv2025

RobPicker: A Meta Learning Framework for Robust Identification of Macromolecules in Cryo-Electron Tomograms

Accurate particle picking of macromolecules of all kinds, shapes, and sizes in cryogenic electron tomograms (cryo-ET) is critical for understanding the molecular architecture of biological systems in their native state. Current deep learning methods have shown potential in identifying macromolecules from tomograms, but they are vulnerable to issues of noise in the training dataset obtained through human or automatic labeling and imbalanced distribution of macromolecule species. To address these limitations, we developed RobPicker, a meta-learning framework that effectively mitigates these issues by automatically learning deep neural networks to correct label errors and give greater emphasis to underrepresented macromolecule species. In evaluations across diverse cryo-ET datasets with noisy labels and imbalanced species distributions, RobPicker substantially outperforms state-of-the-art methods, particularly in identifying small and rare macromolecules. The efficiency and robustness of RobPicker can also be used for rapid fine-tuning of the tilt-series alignment, leading to improved tomogram reconstruction and enabling high-resolution cellular structural biology analysis.

bioinformatics↗

bioRxiv2025

Identification of a novel CLPX variant in a mixed breed dog with anemia and spinocerebellar ataxia

Spinocerebellar ataxia (SCA) or hereditary ataxia is a progressive neurodegenerative disorder primarily manifesting as cerebellar or spinocerebellar dysfunction, resulting in the loss of motor control and voluntary muscle coordination. SCAs are typically inherited conditions, with causative genetic variants identified in multiple genes in people and across various dog breeds. Recently, an atypical case of SCA was documented in a mixed breed dog. In addition to the classic clinical signs and spinocerebellar lesions of SCA, the dog had retinal and optic nerve degeneration and severe, non-regenerative anemia. Whole-genome sequence (WGS) of the affected dog did not reveal any previously identified canine SCA-associated variants. Subsequent variant filtering against a control cohort of over 700 unaffected dog genomes identified a homozygous 4-base-pair frameshift deletion in caseinolytic mitochondrial matrix peptidase chaperone subunit X (CLPX) [XM_038580726.1:c.1723_1726del]. CLPX encodes a subunit of the ATP-dependent ClpXP protease, a molecular chaperone involved in mitochondrial protein degradation. The variant is predicted to cause a frameshift and a premature stop codon within 17 amino acids, truncating approximately 6.64% of the protein. Our study is the first to explore the association of CLPX variants with SCA in any species. Given the high evolutionary conservation of CLPX, this report of a CLPX variant associated with SCA in a dog may have relevance for understanding CLPX-related neurodegeneration and/or anemia in other species. Author SummaryA young mixed-breed dog developed a gait abnormality that progressively worsened, together with vision loss, and severe anemia. Despite treatment, the dogs condition deteriorated, and he was humanely euthanized. An autopsy revealed extensive abnormalities in the brain, spinal cord, eyes, and bone marrow. These histologic findings supported a diagnosis of spinocerebellar ataxia (SCA), also known as hereditary ataxia, which is a genetic neurological disorder that results in impaired movement and diminished coordination. Genetic analysis identified a previously unreported mutation in the CLPX gene. CLPX plays a key role in mitochondrial protein quality control by helping break down damaged or misfolded proteins within mitochondria--cell structures critical for energy production that are particularly crucial in high-demand tissues like the brain. This mutation likely disrupted normal CLPX protein function, leading to both nerve damage and impaired blood cell production. While related genes are known to cause similar conditions in humans, this is the first time a naturally occurring CLPX variant has been identified in an SCA case in any species. Because CLPX is highly conserved between dogs and humans, this finding may offer valuable insights into rare inherited neurological diseases in people.

genomics↗

bioRxiv2025

Efineptakin alfa (NT-I7) improves overall survival and induces tertiary lymphoid structures in murine lung tumors

Tertiary lymphoid structures (TLSs) are emerging as good predictive biomarkers of response to cancer immunotherapy. However, therapeutic strategies to induce these structures are currently limited. We evaluated the therapeutic benefit of efineptakin alfa (NT-I7), a long-acting form of IL-7, and its ability to induce TLSs in a murine lung tumor model. NT-I7 improved overall survival in tumor-bearing mice. It also increased the abundance of T, B, dendritic cells, and stem-like CD8 T cells and promoted the formation of immune aggregates in the tumor microenvironment (TME). Stem-like CD8 T cells were preferentially located in the immune aggregates. Spatial transcriptomic analyses of the TME further demonstrated that the immune aggregates induced by NT-I7 included TLSs with enrichment of Cd274 (PD-L1) transcripts and genes involved in antigen processing and presentation. Upregulation of Cd274 in the TLSs may provide opportunities for synergy between NT-I7 and PD-1-targeted immunotherapy. STATEMENT OF SIGNIFICANCEThis study demonstrates the ability of efineptakin alfa (NT-I7) to potentially augment the clinical efficacy of cancer immunotherapy by inducing tertiary lymphoid structures in the tumor microenvironment.

immunology↗

bioRxiv2025

An AI Agent for cell-type specific brain computer interfaces

Decoding how specific neuronal subtypes contribute to brain function requires linking extracellular electrophysiological features to underlying molecular identities, yet reliable in vivo electrophysiological signal classification remains a major challenge for neuroscience and clinical brain-computer interfaces (BCI). Here, we show that pretrained, general-purpose vision-language models (VLMs) can be repurposed as few-shot learners to classify neuronal cell types directly from electrophysiological features, without task-specific fine-tuning. Validated against optogenetically tagged datasets, this approach enables robust and generalizable subtype inference with minimal supervision. Building on this capability, we developed the BCI AI Agent (BCI-Agent), an autonomous AI framework that integrates vision-based cell-type inference, stable neuron tracking, and automated molecular atlas validation with real-time literature synthesis. BCI-Agent addresses three critical challenges for in vivo electrophysiology: (1) accurate, training-free cell-type classification; (2) automated cross-validation of predictions using molecular atlas references and peer-reviewed literature; and (3) embedding molecular identities within stable, low-dimensional neural manifolds for dynamic decoding. In rodent motor-learning tasks, BCI-Agent revealed stable, cell-type-specific neural trajectories across time that uncover previously inaccessible dimensions of neural computation. Additionally, when applied to human Neuropixels recordings-where direct ground-truth labeling is inherently unavailable-BCI-Agent inferred neuronal subtypes and validated them through integration with human single-cell atlases and literature. By enabling scalable, cell-type-specific inference of in vivo electrophysiology, BCI-Agent provides a new approach for dissecting the contributions of distinct neuronal populations to brain function and dysfunction.

neuroscience↗