Biology subjects

Xia, T.

Publications and source records attributed to Xia, T..

At least 19 recordsLinked to original sources

bioRxiv2026

Sleep reactivation shapes memories of event and its temporal structure

Episodic memory provides a temporal scaffold that organizes past experiences according to their sequential occurrence. While memories are reactivated and consolidated during sleep, it remains unknown how exogenous reactivation reshapes memories of complex events, particularly the temporal structure that binds individual elements across time. In this preregistered study, 35 participants encoded neutral and aversive narratives composed of image sequences, with auditory context shifts defining event boundaries. During post-encoding slow-wave sleep, closed-loop targeted memory reactivation (TMR) delivered boundary-associated auditory cues time-locked to the up-state of slow oscillations. Reactivation led to faster, more confident, and more liberal recognition responses for cued than uncued memories. However, reactivation exerted a time-dependent effect on temporal-order memory, impairing it immediately after sleep but attenuating forgetting one week later. This behavioral dissociation was paralleled at the neural level: stronger cue-elicited theta activity predicted superior recognition of neutral, but not aversive event content, whereas greater theta responses were associated with poorer temporal-order memory for the same events. Together, these findings suggest that sleep reactivation does not uniformly strengthen episodic memories. Rather, it selectively prioritizes access to event content while reorganizing temporal relations among event elements, revealing distinct, time-dependent consequences of sleep reactivation for different components of complex experiences.

neuroscience↗

bioRxiv2026

DCAF: Unifying Data Validity, Evaluation Validity and Predictive Modeling for Single-cell Perturbation Simulation

Single-cell perturbation profiling is limited by cost and throughput, while computational methods remain constrained by inconsistent evaluation and underperformance relative to simple baselines. Here we present DCAF, a unified framework for single-cell perturbation simulation that that jointly addresses data validity, evaluation validity and predictive modeling. Among its three components, DCAF-data addresses the single-measurement limitation by reconstructing pre- and post-perturbation status using entropy-regularized optimal transport, and is demonstrated on both cell line and multi-cell type datasets. DCAF-criterion evaluates log fold changes in non-silenced genes to improve sensitivity to model error and robustness to zero inflation noise, validated on synthetic datasets. DCAF-model combines JEPA with explicit modeling for regulatory-dynamics modeling for interpretable prediction. Across benchmarks, DCAF outperforms state-of-the-art deep-learning methods, as well as mean and linear baselines in cross-gene, cross-cell-type and cross-region generalization. In adult Mouse brain, DCAF recapitulates known cell-type-specific regulatory heterogeneity of Dgcr8 knockdown at both cell type and gene levels. DCAF thus provides a rigorous and biologically grounded foundation for reliable perturbation simulation across development, disease and therapeutic discovery.

bioinformatics↗

bioRxiv2026

Sum-h2, enabling genetic discovery for deep learning-derived phenotypes through a fast evaluation framework and arena of performance

In the recent growing interest of AI research toward biology, genetic association studies of AI- derived phenotypes from high-content modalities such as images emerges as a powerful means for biological discovery. However, such AI-phenotyping methods still lacks a good optimization target and an efficient evaluation framework. The number of discovered loci was the major criterion for evaluating the quality of AI-derived phenotypes. However, the time and computational resources required for running the GWAS and subsequent loci-clumping are substantial, limiting the rapid development and iteration of deep learning representation algorithms. Here we present a 1000x faster and lightweight framework, sum-h2, than traditional GWAS framework for evaluating genetic discovery through total heritability. We revisit tr(P^(-1) G), previously proposed in the context of evolutionary studies, as a measure of multi-phenotype heritability. We showed that sum-h2, the sum of heritability over phenotypic Principal Components (PCs), is equivalent to the linear transformation-invariant tr(P^(-1) G), through both theoretical proof and simulation studies. Moreover, sum-h2 can be estimated rapidly with minimal information loss over a relatedness-enriched sample, while preserving the relative ranking of endophenotypes by GWAS loci counts. Based on selected UKB data and sum-h2, we set up a Genetic Discovery Arena, enabling rapid and fair comparisons for the development and optimization of deep learning-derived phenotyping methods.

bioinformatics↗

bioRxiv2026

Mode of T cell priming durably shapes the TCR repertoire, effector function and α4β1 integrin expression of human virus-specific CD4+ T cells

The generation of an effective T cell response against an antigen depends on the recognition of the antigen by the T cell receptor (TCR), followed by T cell priming, initiating coordinated biophysical, biochemical and proliferative changes that drive differentiation into effector and memory clones. The immunological environments in which priming occurs, such as natural infection or vaccination, influences the quality and persistence of memory T cells, but the long-term impacts remain incompletely understood. Here, we investigate how the mode of priming shapes durable antigen-specific CD4+ T cell memory, utilising two cohorts 3-4 years after initial antigen encounter: individuals recovered from SARS-CoV-2 infection and infection-naive individuals who received a SARS-CoV-2 vaccination. Using ex vivo single-cell RNA sequencing, paired TCR sequencing and in vitro functional analyses, we characterise the transcriptional, clonal and functional profile of Spike-specific CD4+ T cells. Across both cohorts, CD4+ T cell responses against spike epitopes S166-180, S751-765 and S866-880, were immunodominant, with shared public TCR clonotypes indicating conserved antigen-recognition regardless of mode of priming. Despite this shared specificity, infection-primed individuals exhibited greater TCR repertoire diversity and lower CDR3{beta} sequence convergence. Transcriptionally, infection-primed cells exhibited a more cytotoxic and effector phenotype, while vaccine-primed cells preferentially adopted T follicular helper (Tfh)- and Th1-associated phenotypes. Infection-primed individuals also displayed enrichment of cell adhesion and integrin signalling pathways, with a greater proportion of spike-specific CD4+ T cells expressing 4{beta}1 integrin subunits, consistent with enhanced migratory and effector potential. Collectively, our findings demonstrate that the mode of antigen priming may influence the long-term CD4+ T cell memory states, influencing TCR repertoire diversity, functional differentiation and tissue-homing potential, years after the initial immune response.

immunology↗

bioRxiv2026

Dissociable reactivation during NREM and REM sleep supports memory consolidation and emotional dissipation

Sleep is critical for memory consolidation and emotional regulation, yet the respective roles of non-rapid eye movement (NREM) and rapid eye movement (REM) sleep remain unclear. Here, using a within-subject crossover design, we recorded high-density electroencephalography (EEG) across two experimental nights while participants viewed neutral or aversive film clips in a counterbalanced order. Combining with daytime functional localizers establishing neural patterns of aversive vs. neutral emotional processing, multivariate pattern analysis revealed that the reactivation of aversive vs. neutral memory during nocturnal sleep was both stage-dependent and event-specific. In NREM sleep, valence-specific reactivation was time-locked to slow oscillation (SO)-spindle complexes but not to either event alone; in REM sleep, reactivation occurred selectively during phasic REM periods marked by rapid eye movements. Critically, NREM SO-spindle coupling percentage was associated with consolidation of temporal memories; whereas phasic REM reactivation strength was linked to overnight dissipation of negative affect. Our findings provide direct evidence that sleep reprocesses emotional experiences through dissociable stage- and event-specific mechanisms, laying out a framework for future targeted sleep-based interventions.

neuroscience↗

bioRxiv2026

Sleep reactivation of positive self-evaluation reveals neural-evaluative alignment that varies with depressive symptoms

Positive self-evaluation underpins resilience and mental health, yet how to enhance positive self-evaluation and the underlying neurocognitive mechanisms remain poorly understood. Existing data mostly examined self-evaluative processing of the wakeful minds, with the role of sleep remain unexplored. Here, we demonstrated that reactivation of positive self-evaluation during sleep revealed fine-grained neural-evaluative alignment that varies with depressive symptoms. We found that targeted reactivation of positive personality traits during non-rapid eye movement (NREM) sleep enhanced positive self-evaluation, but this benefit diminished with elevated depressive symptoms. Moreover, cueing positive traits during NREM sleep elicited trait-specific, 11-16 Hz EEG sigma power-based neural representations that aligned with post-sleep positive self-evaluative representational structures, as revealed by representational similarity analyses. Notably, individuals with elevated depressive symptoms exhibited weaker neural-evaluative representational alignments, identifying a maladaptive self-evaluative processing that bridges across sleep and wakefulness states. These findings provide mechanistic insights on how fine-grained neural-evaluative alignment can substantiate positive self-evaluation, and why it was disrupted as depressive symptoms increased. HighlightsO_LIPositive-trait reactivation during NREM sleep enhances positive self-evaluation C_LIO_LIThis behavioral benefit decreases as depressive symptoms increase C_LIO_LISleep cueing evokes trait-specific sigma-band neural representations C_LIO_LINeural-evaluative alignment weakens as depressive symptoms increase C_LI

neuroscience↗

bioRxiv2026

RepGene: Toward a Unified Gene Representation Space Robust to Missing Biological Views

This study evaluates how multi-view gene representations should be compared under a matched contract. Each of 18,460 genes is described by five aligned views: DNA sequence-derived features (D), RNA sequence-derived features (R), protein language-model embeddings (P), curated text knowledge (K), and single-cell expression (X). The contract fixes the gene universe, the 256-D output, label-free outer-training-only fitting, one random-forest probe, and task-equal Macro-F1 over nine registered binary tasks, and varies only the fusion operator across four protocol-defined input states: Core (DNA+RNA+protein), Core+K, Core+X, and Core+K+X. Under these states, the top-scoring operator changes with the available view combination and all task-bootstrap intervals overlap, establishing a condition-dependent rather than universal ranking pattern. An independently configured external RF track provides a separate check of the numerical range, while a gene-to-cell boundary diagnostic shows that gene-level differences attenuate toward a dimension-matched random control under expression-weighted aggregation, as predicted by a dilution identity. Historical log-derived multiplicative-interaction screening is retained only as magnitude-only supplementary evidence because fold-level records are unavailable. The contribution is therefore the matched contract and its audit trail: an evidence-graded protocol for turning fusion claims into comparable, condition-bounded measurements.

bioinformatics↗

bioRxiv2026

The emergent mental operation of the learning brain

How does the learning brain give rise to emergent mental operations that enable skill internalization and generalization? Using five-year longitudinal tracking of children acquiring abacus-based mental calculation, we reveal how sustained practice progressively reconfigures whole-brain states. Learning induces nonlinear transitions from stable baseline configurations through a transient exploratory phase toward restabilized associative brain networks. This hierarchical remodeling cascades from perceptual systems to action and default networks, mediated by adaptive neural variability and selective connectivity reweighting, with the salience-parietal memory network acting as a dynamic relay. These brain-state transitions accompany an ordered cognitive transfer cascade, emerging within arithmetic and extending to visuospatial and executive functions. Our findings are reproducible and provide a systems-level account of how ecological learning sculpts large-scale networks to enable complex skill acquisition.

neuroscience↗

bioRxiv2026

Closed-loop auditory stimulation during sleep shapes cortical responses and memory consolidation depending on spindle timing

Sleep spindles contribute to memory consolidation, and acoustic stimulation during sleep can modulate spindle activity. However, it remains unclear whether these effects depend on the timing of stimulation relative to the spindle cycle. Using closed-loop spindle detection, we delivered four variants of pink noise either during ongoing spindles or shortly after spindle offset. In Experiment 1, stimulation delivered during a spindle prolonged its duration, whereas stimulation applied shortly after spindle offset, i.e., within the refractory period, delayed the onset of the subsequent spindle. This pattern was consistent across all sound variants, indicating that the stimulation timing, rather than acoustic properties, drove the observed effects. Experiment 2 examined whether post-spindle stimulation influences memory consolidation. Relative to a no-stimulation nap, post-spindle stimulation reduced post-nap deterioration in motor sequence accuracy but did not significantly affect word-pair recall. EEG analyses revealed that stimulation elicited widespread slow oscillation activity without triggering new spindles, demonstrating that cortical processing of external input persists even during the refractory period. Notably, the motor benefit was greatest in participants with lower spindle density during the baseline adaptation nap. Together, these findings challenge the notion that the post-spindle refractory period is a functionally silent interval. Instead, this period remains receptive to external input, capable of eliciting cortical activity and supporting motor memory consolidation even in the absence of new spindle generation.

neuroscience↗

bioRxiv2026

Structural Connectivity Between the Zinc Linchpin Motif, the cluster, and the active site orchestrates DNA repair in MUTYH

The DNA glycosylase MUTYH protects genomic integrity by excising adenine mispaired with 8-oxoguanine (OG), which initiates base excision repair (BER). The [4Fe-4S] cluster DNA binding domain and Zn linchpin motif of MUTYH are hotspots for inherited cancer associated variants (CAVs) highlighting their critical functions in DNA repair. Here, we present three full-length human MUTYH crystal structures bound to DNA across three catalytic states, representing early and late transition states and product complexes. These structures reveal a previously unrecognized interplay between the Zn site and the Fe-S cluster mediated by His85 and a conserved Arg247/Arg307 bridge spanning [~]15 [A]. Disruption of this network impairs metal loading, reduces the active enzyme fraction, weakens lesion DNA binding, and diminishes OG:A repair in cells. Moreover, the Zn site adopts distinct coordination modes across the reaction coordinate, with a water molecule replacing a cysteine ligand in transition state analog complexes and cysteine coordination restored in a product-mimicked state. Functional analyses show that this ligand switching is dispensable for core glycosylase chemistry in vitro but disproportionately affects repair in cells, suggesting an additional role of the Zn linchpin in cellular OG:A repair beyond intrinsic glycosylase activity, likely involving interactions with BER partners. Together, these findings illustrate how the Zn linchpin enhances Fe-S cluster domain DNA substrate engagement to enable effective DNA repair and provides a rationale for how these functions are compromised by MUTYH CAVs.

biochemistry↗

bioRxiv2026

Replicability of unsupervised deep learning derived image phenotypes

Unsupervised deep-learning image phenotypes derived from brain MRI are propelling imaging genetics to link brain structure to genetic variation. However, their replicability across data sets has not been sufficiently evaluated, raising questions about whether they capture robust biological structure or reflect training-specific artifacts. Here, we assess the replicability of unsupervised deep-learning image phenotypes under variation in model initialization, data partitioning, and cohort, directly evaluating their stability across experimental conditions. We trained multiple models under (i) different training batch random seeds, (ii) cross-validation splits, and (iii) independent datasets (UKB and ADNI), across CNN and ViT architectures. We then derived representations from a separate UKB discovery cohort (N = 22,985) for both trained models and random initialized models without training. The representation stability was assessed using centered kernel alignment (CKA; mean ViT 0.74 vs random 0.27) and kernel canonical correlation analysis (KCCA; mean ViT 0.84 vs random 0.60), as well as genetic discovery stability using loci overlap ratio (mean ViT 0.45 vs random 0.08). We further applied weighted MAXVAR generalized CCA to 12 embeddings to extract a shared 30-dimensional subspace. Our result showed that UDIPs exhibit statistically significant stability (CKA, KCCA t test p < 0.001) across training perturbations and preserve biologically meaningful structure (loci overlap ratio t test p <0.001) across cohorts, supporting their use in imaging genetics.

bioinformatics↗

bioRxiv2026

Learning dynamics of unsupervised deep learning reveal epoch-specific genetic architectures of brain morphology

Representation learning is an emerging paradigm for deriving phenotypes from complex measurements (e.g., imaging) for genetic discovery. However, the learning dynamics of deep neural networks, especially the evolution of representations during training, while of interest in representation learning, were insufficiently investigated in the context of genetic discovery. In this study, using a 3D convolutional autoencoder trained on T1-weighted brain MRIs UK Biobank participants, we show that its learning trajectory forms an epoch-stratified landscape of brain morphology heritability. Different training epochs capture distinct genetic architectures at comparable heritability levels. Overall, ensembling across informative checkpoints identifies more genomic risk loci than the conventional single-checkpoint approach. Interpretability analysis reveals that epoch-specific loci, including MAPT and MCPH1, map onto biologically coherent and distinct neuroanatomical signatures, identified at different stages of the training process. Our results establish learning dynamics as a novel axis for genetic discovery using unsupervised deep learning and have practical implications for any architecture that saves multiple checkpoints during training.

bioinformatics↗

bioRxiv2026

Interactome mapping in human excitatory neurons reveals novel risk genes and pathways in Alzheimer's disease

Alzheimers disease (AD) is an irreversible neurodegenerative disease defined by its molecular hallmarks - amyloid beta peptide plaques and neurofibrillary Tau tangles. Despite significant progress that has been made in uncovering a large number of genetic risk factors through extensive genomic sequencing and genetic studies, the molecular mechanisms driving AD-associated pathology and cognitive decline remain poorly understood. Therefore, alongside the identification of more risk genes, it is also paramount to study how these genes function and influence each other within the cellular pathways and overall molecular networks in AD-relevant brain cell types. However, current human protein-protein interactome datasets were all generated in either yeast or generic human cell lines. Consequently, many important neuronal interactions, especially neuron-specific ones, have yet been discovered. To address this critical gap, we developed a highly scalable, high-quality interactome mapping pipeline in human excitatory neurons derived from induced pluripotent stem cells (iPSC), and generated a comprehensive, neuron-specific interactome map, named ADNeuronNet, for key AD risk genes. ADNeuronNet consists of 1,767 high-confidence interactions among 1,189 proteins and is the only dataset enriched with neuron-specific genes when compared to known protein interactions, including previous large-scale interactome maps, for the same baits in the literature. Within ADNeuronNet, we identified 1,375 novel interactions, many of which are likely neuron specific. For example, we identified a neuron-specific interactor, RIN2, for major AD risk factor BIN1 and confirmed RIN2s function in recruiting BIN1 to RAB5 positive early endosomes, a process that has been well-associated with AD etiology. Additionally, we performed quantitative interaction perturbation analyses on AD risk genes with AD-associated mutations or isoforms and identified significant changes in 99 protein interactions among 11 different protein variants. Finally, we found that subunits from the anaphase-promoting complex/cyclosome (APC/C), another novel BIN1 interactors identified by ADNeuronNet, mediated modulation of Tau-aggregation in neurons via regulation of APOE expression, uncovering a previously unrecognized BIN1-APC/C-APOE regulatory axis in AD pathobiology. In summary, these findings illustrate how our neuron-specific ADNeuronNet can be leveraged to uncover new risk gene candidates and cellular pathways that help advance our understanding of molecular mechanisms underlying AD etiology.

systems biology↗

bioRxiv2026

Reactivating conflicting evaluative memories during sleep reduces decision ambivalence

Memory guides everyday evaluations and decision-making. Yet people often encounter inconsistent information about the same target, giving rise to conflicting evaluative memories and decision ambivalence. Decision ambivalence is not only aversive but also reduces confidence, increases hesitation, and leads to maladaptive choices. While sleep consolidates memories, its role in resolving these evaluative conflicts and shaping decision dynamics remains unknown. Here, we investigated how memory reactivation during sleep, a critical period for memory consolidation and transformation, would reconstruct conflicting evaluative memories and thus influence next-day decision ambivalence. In a valence reversal learning procedure, participants first encoded positive or negative cue-outcome associations (A-B) on Day 1, followed by learning A-C associations yet with opposite valences on Day 2 (i.e., Day 1 negative-to-Day 2 positive and Day 1 positive-to-Day 2 negative). During subsequent non-rapid eye movement (NREM) sleep on Day 2 night, half of the cues were re-presented to sleeping participants to reactivate the cue-associated memories. Upon waking up, participants completed post-sleep evaluation and memory tests. Our results showed that cueing reduced decision ambivalence, especially in the negative-to-positive condition, as evidenced by less curved mouse-tracking trajectories. Meanwhile, cueing promoted the integration of conflicting evaluative memories, again in the negative-to-positive condition. Critically, cueing-induced ambivalence reduction was evident only for items that were integrated after sleep. Electrophysiologically, stronger cue-elicited delta power during NREM sleep predicted next-day ambivalence reduction, while higher cue-elicited spindle probabilities were associated with better memory integration. Together, our findings suggest that memory reactivation during post-learning NREM sleep actively reorganizes conflicting memories, providing a mechanistic pathway through which offline memory reprocessing resolves waking decision ambivalence.

neuroscience↗

bioRxiv2026

Learning heritable multimodal brain representation via contrastive learning

Magnetic resonance imaging (MRI)-derived phenotypes (IDP) has enabled the discovery of numerous genomic loci associated with brain structure and function. However, most existing IDPs and learned representations are derived from a single imaging modality, missing complementary information across modalities and potentially limiting the scope of genetic discovery. Here, we introduce a multimodal contrastive learning framework to derive heritable representations from paired T1- and T2-weighted MRIs. Unlike single-modality reconstruction-based models, we designed a momentum-based contrastive learning framework. As a result, our approach offers improved prediction of traditional IDPs, age, and brain disorders. Notably, genome-wide association studies (GWAS) of the learned representations reveal a substantially higher overlap of genetic loci across modalities, indicating improved alignment of their underlying genetic architecture. Analysis of the GWAS loci identified shared protein and drug targets, yielding meaningful biological insights. Overall, our framework learns shared representations across brain imaging modalities that exhibit anatomical and genetic coherence.

bioinformatics↗

bioRxiv2026

A biofidelic Goat Model of Traumatic Optic Neuropathy with Optic Canal Fracture via Transnasal Endoscopy

Traumatic optic neuropathy (TON) is a major cause of irreversible vision loss following blunt cranial trauma, yet the absence of clinically relevant large animal models that faithfully recapitulate human TON has significantly impeded translational research. Current rodent models do not reproduce optic canal fracture, a key injury mechanism in many patients with TON. Here, we combined high-fidelity finite element analysis (FEA) with iterative engineering to establish a reproducible goat model of TON. We first built a high-resolution human-head finite element model to characterize force transmission to the optic nerve. Across clinically relevant periorbital impacts, stress preferentially concentrated in the intracanalicular segment, reaching a peak force density of approximately 500 N/m2 at 50.6 ms, about fivefold higher than in the intraorbital segment. Simulations further showed that direct optic canal impact reproduced comparable intracanalicular stress with a markedly lower input force: 195 N, compared with 3900 N for periorbital impact. Guided by these insights, we developed transnasal endoscopic impact systems capable of inducing optic canal fractures in goats. TON was confirmed within 24 hours by a characteristic relative afferent pupillary defect (RAPD), and at 1 month post-injury goats (n = 14) exhibited a 10%-20% reduction in ganglion cell complex (GCC) thickness and 40%-65% reductions in flash visual evoked potential (FVEP) and pattern electroretinogram (PERG) amplitude ratios (all P [≤] 0.001), with structural and functional preservation of the fellow eye. This study presents a robust, standardized, and clinically relevant large-animal platform for investigating TON pathophysiology.

neuroscience↗

bioRxiv2026

O-GlcNAcylation and low glycolysis underpin Th2 polarization by dendritic cells

Activation of dendritic cells (DCs) is dependent on rewiring of their cellular metabolism. However, the metabolic requirements for DCs to prime T helper 2 (Th2) responses are still poorly understood. Using unbiased transcriptomics and non-targeted metabolomics we find that helminth antigen-conditioned human DCs suppress glycolysis while increasing hexosamine biosynthesis to fuel protein O-GlcNAcylation. Functionally, glycolytic inhibition of DCs selectively enhanced, while blocking O-GlcNAcylation impaired, Th2-priming capacity. In helminth infection and allergic challenge, Th2 responses were also attenuated in vivo in mice with specific deletion of O-GlcNAc Transferase (OGT) in CD11c-expressing cells. Mechanistically, through proteomic analysis and functional validation, we identified O-GlcNAcylation as a critical negative regulator of immune synapse formation by controlling cytoskeletal organization via Fascin-1 and Zyxin, thereby dampening TCR signalling to promote Th2 polarization. Altogether we reveal a novel metabolic program in DCs that governs Th2 polarization, that could potentially be harnessed to treat type 2 mediated inflammatory diseases.

immunology↗

bioRxiv2026

BiOmics: A Foundational Agent for Grounded and Autonomous Multi-omics Interpretation

While AI has automated bioinformatic workflows, biological interpretation remains fragmented and often disconnected from mechanistic insights. Existing AI is bifurcated between statistical "black-box" models that lack logical grounding and simple agents restricted to shallow knowledge retrieval. To bridge this divide, we introduce BiOmics, a foundational agent that synthesizes multi-omics data with adaptive knowledge for biological interpretation. BiOmics introduces a novel dual-track architecture comprising a harmonized explicit reasoning space for grounded logic and a unified latent embedding space for high-dimensional association mapping. This architecture enables a transformative "Retrieving-Reasoning-Predicting" paradigm for purposeful, cross-scale inference traversing the biological hierarchy, from molecular variants to disease phenotypes. Empirical evaluations demonstrate that BiOmics surpasses state-of-the-art AI agents and specialized algorithms, markedly augmenting the granularity and depth of biological insights. Specifically, BiOmics exhibits unique superiority in uncovering indirect pathogenic variants, achieving reference-free cell annotation, and prioritizing drug repurposing candidates tailored to specific datasets. BiOmics further enriches the interpretive landscape of biological entities, leveraging its reasoning-grounded knowledge graph to uncover deep functional contexts. Ultimately, BiOmics provides a versatile engineering foundation to transition AI for Science from descriptive "data fitting" to autonomous, knowledge-driven interpretation.

bioinformatics↗