Biology subjects

Sarkar, S.

Publications and source records attributed to Sarkar, S..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Siglec-G on B cells restrains the germinal center response by controlling T cell help during positive selection

The germinal center (GC) reaction requires tight regulation of B cell and T follicular helper (Tfh) cell interactions to ensure B cell expansion and antibody affinity maturation, while preventing oncogenesis. However, regulatory mechanisms fine-tuning B-T cell interactions within the GC to prevent aberrant activation and proliferation remain incompletely understood. Here, we identify Siglec-G, the mouse ortholog of human Siglec-10, as an immune checkpoint that restrains the GC by dampening B-T cell interactions. Selective and temporal ablation of Siglec-G on B cells after immunization triggers GC hyperplasia and enhanced plasma cell and antibody output. While Siglec-G is dispensable in B cell receptor (BCR)-mediated processes, it acts as an intrinsic inhibitory receptor of B-T cell interactions in the GC, ultimately limiting Myc and mTORC activation within positively selected GC B cells. Trans interactions of Siglec-G and its glycan ligands on Tfh likely contribute in fine-tuning the strength of bidirectional signaling following contact between GC B cells and Tfh cells. This interaction is further reinforced by glycan remodeling that occurs in the GC, resulting in concurrent decreased in glycan ligands on GC B cells and increased in glycan ligands on Tfh. This augmented binding of Siglec-G/10 on Tfh is mainly due to the upregulation of 2-6 linked sialic acid ligands. Moreover, APEX2-based proximity labeling revealed several candidate Siglec-G/10 binding partners on T cells, including BTLA, CD6, and Slamf6, which are known negative regulators of Tfh cell activation. Taken together, our findings identified that Siglec-G acts as a GC checkpoint receptor, restricting B cell proliferation by tuning T cell help following B-T cell interactions.

immunology↗

bioRxiv2026

ApoE Lipidation State Directs Immunometabolic Reprogramming of Human Microglia

IntroductionApoE4 is the strongest genetic risk factor for Alzheimers disease (AD). Emerging evidence suggests that ApoE4 increases AD risk by disrupting microglial metabolism and function. However, whether ApoE lipidation state contributes to microglial dysfunction remains poorly understood. MethodsHuman microglia were treated with lipid-free or lipid-bound ApoE3 or ApoE4. Label-free live-cell holotomography and global proteomics were used to assess isoform- and lipidation-specific effects on lipid droplet dynamics, mitochondrial morphology, and microglial phenotype. ResultsApoE4 treatment resulted in fewer but enlarged lipid droplets and increased mitochondrial fragmentation compared to ApoE3, effects that were enhanced by lipid-bound ApoE4. Proteomic analyses revealed a strong type I interferon response in cells exposed to lipid-free ApoE, which was exacerbated by lipid-free ApoE4. DiscussionThese findings indicate that lipid-bound ApoE4 drives metabolic reprogramming, whereas lipid-free ApoE4 promotes inflammatory signaling, identifying ApoE lipidation as a critical modifier of ApoE4-associated AD risk.

neuroscience↗

bioRxiv2026

Vascular tree structure-based perfusion phantom fabrication using modified Hele-Shaw Cell technique

Tissue phantoms that mimic microvasculature and perfusion are essential for modelling vascular function, guiding interventions, and calibrating imaging systems, which require faithful replication of vascular geometry and flow. Conventional fabrication strategies, including wire-based molding, lithographic micromachining, and additive manufacturing, offer useful capabilities but remain constrained by predefined designs, rectangular channel cross-sections, limited scalability, and high production costs. Reliance on predefined digital vascular models restricts design flexibility and limits the ability to capture the natural variability and complexity of real vascular systems. Here, we present a lithography-free, fractal-generating approach based on a modified Lifted Hele-Shaw Cell (LHSC) technique, in which vascular networks emerge spontaneously via interfacial fluid instabilities. Unlike pre-designed methods, these structures are governed by fluid properties and flow conditions, enabling adaptive, physiologically relevant geometries with smooth Gaussian cross-sections and natural diameter tapering. We demonstrate four phantom designs: a planar vascular tree, an anatomically guided cerebral network, a retinal vascular model, and a conformable curved substrate phantom. Validation using Laser Speckle Contrast Imaging confirms structural fidelity and physiologically relevant flow consistent with Murrays law. This platform uniquely integrates realistic vascular architecture with emergent, fractal driven formation, highlighting its potential as a reproducible and biologically relevant alternative to conventional vascular phantom fabrication. Furthermore, the availability of such realistic in vitro vascular models can reduce reliance on animal experiments and contribute towards more ethical and sustainable preclinical research.

bioengineering↗

bioRxiv2026

Endocytic accessory proteins assemble clathrin while simultaneously destabilizing protein condensates

During endocytosis a dynamic protein network is responsible for assembly of the clathrin coat. Early in the process the Eps15 protein is thought to undergo liquid-liquid phase-separation to form biomolecular condensates. These condensates facilitate clathrin assembly, producing clathrin-coated vesicles that are ultimately excluded from the condensate as the vesicles depart from the plasma membrane. An array of distinct clathrin accessory proteins drive this process, yet their interactions with protein condensates have not been explored. Here we show that clathrin accessory proteins promote the assembly of the clathrin lattice by simultaneously destabilizing protein condensates. By adding diverse accessory proteins to Eps15 condensates, we observed that accessory proteins which cross-linked Eps15 proteins stabilized condensates, while accessory proteins that compete with Eps15-Eps15 interactions destabilized condensates. Interestingly, accessory proteins that destabilized condensates also promoted the assembly of clathrin and its exclusion from condensates. In contrast, accessory proteins that stabilized condensates opposed exclusion of clathrin. Finally, we demonstrated that decreasing condensate stability enhanced clathrin assembly and exclusion, even when no additional clathrin assembly motifs were introduced. Taken together, these results help to explain how condensates can catalyze the assembly of coated vesicles by providing both a substrate for their initiation and a driving force for their exclusion and ultimate departure. Statement of SignificanceThis work is significant because it reveals how biomolecular condensates could actively drive, rather than merely host, clathrin-mediated endocytosis. By showing that clathrin accessory proteins can tune condensate stability to simultaneously promote clathrin lattice assembly and exclusion, this study uncovers a physical mechanism that could link phase separation to membrane trafficking. The finding that condensate destabilization alone enhances clathrin exclusion reframes condensates as dynamic regulators that could provide both an initiation platform and a force for vesicle maturation and release. More broadly, this work advances our understanding of how cells coordinate assembly and disassembly of complex macromolecular structures through regulated changes in condensate material properties.

biophysics↗

bioRxiv2026

Retrograde signalling mediates cellular adaptation to mitochondrial DNA copy number alterations

Eukaryotic cells maintain multiple copies of the mitochondrial genome, which is essential for cellular metabolism. Accordingly, alterations in the mitochondrial DNA (mtDNA) copy number are associated with severe human diseases and ageing. However, the mechanisms through which cells regulate mtDNA copy number and the cellular consequences of altered copy number remain poorly understood. Here, using budding yeast as a model, we show that mtDNA copy number is determined by the amount of three limiting factors, Mip1, Abf2, and Rim1. By synthetically tuning the concentrations of only these three proteins, we can modulate mtDNA dosage inside the cell. This revealed that cells are surprisingly robust to mtDNA copy number alterations, with increased copy numbers even accelerating cell growth. Our findings suggest that this robustness is due to protein dosage compensation and independent regulation of mitochondrial morphology. Mechanistically, we identified a critical role of the retrograde signalling pathway for this adaptation. We show that signalling from mitochondria to the nucleus is upregulated in cells with higher mtDNA copy number, and disruption of this regulation diminishes their faster growth. Taken together, our work reveals regulatory principles that enable cells to adapt to mtDNA copy number alterations.

cell biology↗

bioRxiv2026

Linguistic and Acoustic Biomarkers from Simulated Speech Reveal Early Cognitive Impairment Patterns in Alzheimers Disease

BackgroundAlzheimers disease (AD) causes progressive decline in language and cognition. Automated speech analysis has emerged as a promising screening tool, yet clinical data scarcity limits progress. To address this, we generated a large-scale simulated speech dataset to model linguistic and acoustic deterioration across cognitive stages, Control, Mild Cognitive Impairment (MCI), and AD. MethodsUsing Monte Carlo simulations, we emulated the Pitt DementiaBank "Cookie Theft" narratives. Acoustic features (speech rate, pause duration, jitter, shimmer) and linguistic features (type-token ratio, unique-word count, filler usage) were synthetically sampled from real-world DementiaBank distributions. We trained an XGBoost classifier to distinguish diagnostic groups, and applied SHAP (Shapley Additive exPlanations) to assess feature importance. ResultsThe model achieved high discriminative performance (AUC {approx} 0.94; accuracy {approx} 85%). Compared to controls, simulated MCI and AD groups showed progressive declines in fluency and lexical diversity, and increases in disfluencies and voice instability. SHAP analysis revealed that key predictors included reduced type-token ratio, higher pause and filler rates, and elevated jitter/shimmer. Classification was most accurate for Control vs. AD; MCI misclassifications highlighted intermediate profiles. InterpretationOur framework, FMN (Forget Me Not), captures clinically relevant speech changes using simulated data, offering an explainable and scalable approach for cognitive screening. While not a substitute for real datasets, FMN validates a pipeline that mirrors known AD markers and can guide future real-world deployments. External validation remains a key next step for translational impact.

neuroscience↗

bioRxiv2026

ARID2 loss destabilizes PBAF and drives colorectal cancer

The PBAF is one of three biochemically distinct BAF chromatin remodelers in humans. We previously proposed the role of ARID2, a PBAF component, as a bonafide tumor suppressor in colorectal cancer (CRC). Here, we validated loss of tumor suppression under conditions of ARID2 deficiency emanating from a marked reduction in PBAF complex assembly resulting from destabilization of PBAF-specific components BRD7, PHF10, and PBRM1. Transcriptome profiling of ARID2 deficient CRC cells revealed perturbation of disease processes, including CRC and neurodegenerative disorders, as well as CRC relevant pathways including Wnt/{beta}-catenin signalling, but transcript levels of PBAF-specific components remained unchanged, confirmed by RT-qPCR and TCGA data analysis. Our study establishes ARID2 as a critical stabilizer of the PBAF complex of relevance to CRC.

cancer biology↗

bioRxiv2026

Characterization of a novel soluble melanin secreted by Gliomastix polychroma

Melanin is a heterogeneous, amorphous biopolymer widely distributed in biological systems, yet most canonical forms are intrinsically insoluble. Soluble melanins are rare and remain poorly characterized, with pyomelanin and peptidomelanin representing the best-described fungal examples. Here, we report the isolation and characterization of a distinct, polyphyletic class of soluble melanin that we term `colloidal melanin'. This class of polymers was isolated from four different species of Ascomycetous fungi occupying varied environmental niches. Colloidal melanin was found to be phenolic and nitrogenous, with a low amino acid content. The polymers form particles with median diameters ranging from 79 to 127 nm, suggesting that small particle size may contribute to aqueous solubility via colloidal dispersion. Using chemical inhibition of melanin biosynthesis together with whole-genome sequencing, we show that colloidal melanin can be produced through multiple distinct biochemical pathways. Further, genes identified as involved in melanin biosynthesis across species do not share any detectable common ancestry, consistent with convergent evolution. Together, these findings identify a distinct class of soluble melanin that does not conform to existing soluble melanin classifications.

biochemistry↗

bioRxiv2026

Querying functional drivers in primary murine naive and memory T cells using RNP-mediated CRISPR-Cas9 Gene Deletion

Cytotoxic CD8 T lymphocytes (CTL) are critical for the clearance of pathogenic cells via antigen-dependent cell lysis, thus providing protection against infectious diseases and cancers. CTLs represent one of the key targets of immunotherapies and vaccine design. While microarray and single-cell RNA sequencing of effector and memory CD8 T cells have identified several promising gene targets that may modulate CD8 T cell responses, their development is slowed down by the financial and time constraints related to the generation of germline knockout mice for further validation studies. Here we present a protocol for conducting efficient deletion of genes from activated effector, as well as resting naive and memory primary murine CD8 T cells using RNP-based Crispr/Cas9 technology. This CRISPR modification of CD8 T cells was then adapted to study the effects of gene deletion in the context of acute memory differentiation as well as chronic exhaustion of CD8 T cells. To this end, we have titrated the necessary dose of antigen-specific CD8 T cells to study their differentiation in acute and chronic infections and confirmed the model by demonstrating rapid expansion of CRISPR mediated PD-1 ablated CD8 T cells in a chronic viral infection. Finally, by combining this methodology with a murine model of subcutaneous tumor challenge, this study provides a unique screening system for genes critical for mediating clearance of malignant tumor cells by CTLs. This study expands current technical capabilities for rapid evaluation of functional role of any candidate gene in CTL responses to infections and cancer through targeted gene deletion at any stage of CD8 T cell differentiation without the need for germline gene deletion mouse models.

immunology↗

bioRxiv2026

Comparative Proteomic Analysis of Environmental and Genetic Models of Parkinsons Disease Highlights the Role of Purine Metabolism.

Parkinsons Disease (PD) is the second most common neurodegenerative disease, with many cases being attributed to environmental contaminant exposures. Paraquat (PQ), is a pesticide and environmental neurotoxicant that has been strongly associated with increased risk of PD. PQ is known to be a weak inhibitor of complex I of the electron transport chain, and while its acute toxicity is well understood, the underlying mechanism by which PQ exposure contributes to PD pathophysiology remains unclear. Additionally, the mechanism of PQ neurotoxicity has yet to be effectively compared and related to genetic forms of PD. Given that PD is a heterogeneous disease with both genetic and environmental determinants, we sought to systematically compare the proteomic changes that occur in different genetic and environmental models of PD. In this study, we leveraged untargeted omics approaches to differentiate between systemic, peripheral, and CNS-specific changes in the proteome. We did this by performing a comparative proteomic analysis on the heads and bodies of Drosophila models of PQ ingestion and neuronal -synuclein expression in males. Additionally, we validated the findings with metabolomic analysis of male and female brain stems from a murine PQ inhalation model using C57BL/6J mice. Our findings indicate shared dysregulated pathways across all models, highlighting similar mechanisms of action. Specifically, we identified a glia-specific role in purine nucleotide metabolism upstream of inosine catabolism, which may protect against PQ neurotoxicity. This work identifies potential early points for biomarker detection and potential targets for drug intervention. Significance StatementNeurodegenerative diseases such as Parkinsons disease (PD) pose a growing public health burden, yet disease-modifying therapies remain limited due to lack of mechanistic understanding and disease heterogeneity. Both genetic and environmental factors contribute to PD, complicating the identification of shared therapeutic targets. Here, we identify a convergent pathway common to genetic and environmental models of Parkinsonism that not only affects the brain but also systemically. Using integrated metabolomics, proteomics, and genome-scale metabolic modeling, we demonstrate that purine metabolism is dysregulated across models. Reverse genetic screening of key enzymes in this pathway mitigates locomotor deficits induced by neurotoxic pesticide exposure in Drosophila. These findings reveal a shared metabolic vulnerability in PD and highlight purine metabolism as a potential therapeutic target.

neuroscience↗

bioRxiv2026

Early detection of sudden transitions in Notch signalling

Identifying sudden transitions during phenotypic decision-making of complex biological systems can be crucial for our ability to control a cellular state. Yet, prior determination of these sudden transitions or tipping points remains challenging, as biological systems often exhibit only subtle early changes, which are often masked by inherent noise or rapid transition dynamics. Using Notch signaling as a model, we systematically analyze dynamical transitions in Notch-Delta (ND), Notch-Delta-Jagged (NDJ), and Fringe-mediated NDJ systems for both one and two-cell contexts. In the one-cell ND system, critical slowing down (CSD)-based early warning signals (EWSs) reliably capture transitions between sender (S) and receiver (R) states and remain robust to variation in forcing rate. We further find that flickering is a precursor to transitions in one-cell NDJ system. In contrast, flickering does not occur in the two-cell Notch model due to the presence of a supercritical bifurcation. Our analysis also offers insight into how NICD (Notch Intracellular Domain)-driven and Fringe-mediated asymmetries, along with the strength of external signals, control the emergence of flickering. Overall, this study identifies sudden transitions in Notch signaling under demographic noise and can be extended to other noisy biological systems, with potential applications in drug development and targeted therapeutic interventions.

systems biology↗

bioRxiv2026

Suppression of CD4 T cells by Tregs dictates metabolic remodeling of CD8 T cells during effector to memory transition

Metabolic transitions between naive, effector and memory T cell states are largely orchestrated by TCR, costimulatory and cytokine signals along with nutrient availability in the immune microenvironment. Treg cells have been shown to play a critical role in the effector-to-memory (E-M) transition of virus-specific CD8 T cells through regulation of proliferation and cytotoxic functional programs. However, the precise Treg-dependent metabolic changes that occur in the microniches and, the underlying molecular and cellular mediators of E-M transition remain undefined. Here we show that Treg cells promote the metabolic remodeling of memory-precursor effector CD8 T cells (MPEC) from aerobic glycolysis to fatty acid oxidation as they enter quiescence after antigen clearance. Our data implicate the anatomic microniche of the splenic white pulp as a site for Treg-MPEC interactions. We further show that optimal E-M metabolic transition requires regulation of effector CD4 T cells and inflammatory myeloid cells through inhibitory CTLA4 signals from Treg cells. Moreover, antagonism of inflammatory cytokine interferon-{gamma} (IFN-{gamma}) signals partially rescues the memory defects associated with absence of Treg cells. Together, these findings support a metabolic triad model of memory CD8 T cell differentiation where Treg-dependent regulation of inflammation from effector CD4 T cells promotes the transition of CD8 T cells from cytotoxic effector to quiescent memory metabolic programs. These studies define novel molecular targets that may be exploited to manipulate metabolism, migration and memory function during vaccination.

immunology↗

bioRxiv2026

MicroRNA-17~92 drives metabolic programming of virus-specific effector CD4 and CD8 T cell responses

The miR-17[~]92 microRNA cluster drives oncogenesis by inducing proliferation and survival of cancer cells. A similar pro-proliferative role of miR-17[~]92 has also been identified in pathogen- and tumor-reactive T cells. However, the metabolic underpinnings of miR-17[~]92-drivenT cell expansion and effector differentiation remain undefined. Using a murine model of conditional miR-17[~]92 deletion or constitutive overexpression in viral antigen-specific CD4 or CD8 T cells, here we show that miR-17[~]92 drives terminal differentiation of CD4 and CD8 T cells through sustained activation of mTOR and glycolytic bioenergetics pathways. Constitutively increased expression of miR-17[~]92 led to a universal increase in the numbers of antigen-specific T follicular helper (TFH), T helper 1 (TH1) and cytotoxic T lymphocyte (CTL) effector subsets. During early stages of T cell activation, miR-17[~]92 overexpression was associated with increased glucose uptake and heightened glycolytic and oxidative metabolism to sustain increased proliferation in both CD4 and CD8 T cells. However, prolonged overexpression of miR-17[~]92 led to a loss of polyfunctionality and compromised metabolic fitness (glycolysis and mitochondrial respiration) by the peak of the effector responses. These studies establish miR-17[~]92 as a critical driver of T cell proliferation and effector differentiation through metabolic regulation of both effector CTL and CD4 T cell subsets (TFH and TH1) that are critical for combating viral infections. These studies form the basis for future manipulation of miR-17[~]92 gene/metabolic regulatory network to commandeer T cell immunity during infection, vaccination or cancer immunotherapy.

immunology↗

bioRxiv2026

FRAPPe - a novel FRET-based biosensor to detect inorganic polyphosphates

Inorganic polyphosphate (polyP), a linear polymer of orthophosphate residues, is involved in a range of cellular processes, including phosphate storage, blood clotting, chromatin remodeling, RNA processing, and mitochondrial function. Despite its significance, tools for monitoring polyP with specificity and in real time are lacking, limiting insights into its dynamics. Here, we present FRAPPe (FRET-based Ratiometric Analysis of Polyphosphate), a FRET-based sensor for polyP. FRAPPe consists of an eCFP-eYFP FRET pair flanking CHAD, a polyP-binding domain. The sensor remains unresponsive to nucleic acids and other polyanions, while polyP binding induces a significant decrease in FRET, allowing quantitative estimation of polyP. We further show that FRAPPe can qualitatively estimate polyP levels from crude cell extracts, supporting its utility in high-throughput genetic and pharmacological screens to identify regulators of polyP metabolism and signaling.

biochemistry↗

bioRxiv2026

Treatment of Saccharomyces cerevisiae with cigarette smoke extract causes vacuolar fragmentation to combat cigarette smoke-induced cellular toxicity

Exposure to cigarette smoke is one of the major risk factors for developing various diseases such as chronic obstructive pulmonary disease (COPD), cardiovascular disorders, and cancer mediated via cellular oxidative stress and organelle dysfunction. To this end, the current study investigated how cigarette smoke extract (CSE) affects vacuole structure and function in Saccharomyces cerevisiae, as vacuole plays a crucial role in handling oxidative stress-induced misfolded proteins. Our results showed that CSE exposure causes transient vacuolar fragmentation up to 1 h to increase its surface area to facilitate microautophagy in clearing CSE-mediated misfolded protein and promoting cell survival. However, excessive fragmentation or vacuolar fusion sensitizes cells towards CSE-mediated cellular toxicity. Towards understanding the underlying mechanism, the current study demonstrated the involvement of PI3P and PI (3,5) P2-mediated signaling and phospholipase-driven remodeling of lipid moieties. Moreover, the current study also showed the importance of mitochondrial activity in CSE-mediated vacuolar fragmentation. Prolonged exposure to CSE impairs mitochondrial function and thus disrupts fragmentation, the adaptive survival strategy against CS. It results in proteostasis collapse, which is a characteristic shared by many inflammatory and degenerative disorders. Taken together, the current study reveals a previously unrecognized cellular protection mechanism induced by cigarette smoke and highlights potential therapeutic targets for mitigating CS-mediated diseases

cell biology↗

bioRxiv2026

Conformalized Multiview Learning

SO_SCPLOWUMMARYC_SCPLOWThere is an overwhelmingly large body of literature and numerous algorithms already available on "multimodal AI", based on different modeling and fusion paradigms. As multimodal AI becomes fully integrated into healthcare applications, the ability to say "Im not sure" or "I dont know" when uncertain is a necessary capability for safe clinical deployment, requiring the augmentation of current multimodal AI techniques. Our goal in this work is not to add yet another method to the growing toolbox of multimodal AI, but rather to (i) clarify how existing supervised multimodal AI methods can be understood through the lens of uncertainty quantification, and (ii) develop a unified framework that provides valid and interpretable uncertainty estimates for multimodal AI practitioners. We introduce Coracle, a conformalized framework for multimodal AI that seamlessly adapts to early, late, and intermediate fusion. Coracle provides theoretical marginal confidence guarantees and achieves valid finite-sample coverage without relying on distributional assumptions. Through extensive simulations and analyses of diverse biomedical multiview datasets, we demonstrate that Coracle consistently outperforms existing non-conformalized methods in both uncertainty estimation and calibration. By equipping published multiview models with valid, interpretable confidence measures, Coracle advances the development of trustworthy, uncertainty-aware statistical tools for biomedical decision support. Coracle is publicly available as an open-source R package at https://github.com/himelmallick/Coracle.

systems biology↗

bioRxiv2026

Cell-Permeable Morpholino Inhibits Programmed Death-Ligand 1 at mRNA Level and Potentiates Antitumor Immunity in Breast Cancer

Morpholino-based ASOs with nuclease-resistant backbone and high sequence specificity could modulate the PD-L1/PD-1 axis to overcome T regulatory cell mediated inhibition but poor cell-permeability limits their clinical applications. Here, we develop a guanidinium-modified cell-permeable morpholino specifically targeting PD-L1 in breast cancer. It suppresses PD-L1 expression through effectively mitigating cytokine induced PD-L1 upregulation and enhancing autophagic flux. Notably, the ASO synergistically with durvalumab (anti-PD-L1 antibody) disrupt, hyperglycosylated PD-L1 by mRNA-level interference and also reduced PD-1 T-cell populations. Importantly, systemic and intra-tumoral administration of ASO demonstrated no abnormalities in cardiac tissues which show no off-target toxicity but it reprogrammed the tumor microenvironment by upregulating inducible nitric-oxide synthase and suppressed regulatory cells, collectively enhancing antitumor immune responses. Together, these findings establish morpholino ASO as a promising targeting therapeutic strategy to overcome tumor-induced immunosuppression, opening new avenues for precision treatment in breast-cancer.

cancer biology↗

bioRxiv2025

Reconstructing developmental and disease progression with sample-level embeddings

Single-cell genomics is transformative for characterizing cellular heterogeneity, but many translational questions require comparing samples rather than individual cells. Typical analyses compare "case" and "control" groups, ignoring sample-level variation within them. Here we present scSLIDE, a framework that transforms each samples single-cell data into a compact profile describing where its cells fall in high-dimensional space. By comparing these density profiles, scSLIDE calculates embeddings to characterization variation across samples. Applied to COVID-19 infection, Alzheimers disease, and zebrafish embryogenesis, we show that scSLIDE can be used to cluster patients, reconstruct sample-level disease trajectories, and identify coordinated cellular programs across samples. We discover independent axes of infection and severity, a molecular disease progression that aligns with pathology-based estimates of neurodegeneration, and embryonic "pseudostages" varying across and within timepoints. In each case, we demonstrate how case-control analyses collapse rich biological variation into binary labels, and how sample-level embedding represents a powerful analytical alternative.

genomics↗