Biology subjects

Regev, A.

Publications and source records attributed to Regev, A..

At least 19 recordsLinked to original sources

bioRxiv2026

Archetype analysis of multiplexed antibody imaging reveals interpretable morphological programs in optical pooled screens

High-content imaging-based screens with small molecule or genetic perturbations rely on rich optical phenotypes, yet how distinct subcellular markers contribute complementary biological information remains unclear. Here, we introduce CellMAPP, a multiplexed organelle imaging platform that uses optical pooled screens with iterative antibody staining to map cellular architecture under perturbation. Using 17 markers across 524 genetic perturbations in MCF-7 cells, we systematically examined how individual, and combinations of, markers capture diverse morphological responses. We find that markers provide complementary views of perturbation effects, with increased multiplexing improving signal and the breadth of hit recovery, although some phenotypes are better recovered from a subset of features rather than a full profile. Because non-molecular image-based profiles can be challenging to link with interpretable biology, we developed CHART (available at https://github.com/Genentech/CHART), an archetype analysis framework that identifies distinct reference morphological states for each marker and uses multimodal large language models (MLLMs) to automatically annotate them, connecting cellular images to recognizable organelle phenotypes. Integrating archetypes across multiple markers revealed modular morphological programs that describe coordinated organelle reorganization in response to perturbation. This framework identified regulators of defined organelle states such as micronuclei, nominated novel gene functions and revealed complex patterns of multi-organelle reposition. Together, optical profiling with MAPP and CHART provides an interpretable, scalable framework for image-based screens that sheds light on rules governing cellular structural reprogramming.

systems biology↗

bioRxiv2026

AutoScreen: AI Co-Scientist System for Target Discovery in Functional Genomics

Target discovery in functional genomics remains largely manual and time-consuming, lacking systematic tools for efficient and reproducible gene-level hypothesis generation. We introduce AutoScreen, an AI co-scientist system supporting target discovery through both Pre-screen Design, which constructs perturbation libraries de novo from free-text research descriptions, and Post-screen Analysis, which re-ranks experimental screen hits by integrating statistical scores with biological context. AutoScreen leverages the complementary strengths of multiple specialist agents that perform deep research into multi-modal evidence, information restructuring, parallel searches across >26 biomedical databases, evidence synthesis, and target review to provide transparent, explainable gene prioritization with pipeline provenance. Across 320 genome-scale CRISPR screens as expert-curated benchmarks, AutoScreen achieved a ~19% increase in validated-hit recovery among its top 100 predictions, relative to the strongest agent baseline, and required a 1.2-fold smaller library to recover the same number of hits at the top-500 reference point. AutoScreen reached mean average precision more than two orders-of-magnitude above random baseline. Further, we validated AutoScreen in cancer immune-evasion case studies focusing on natural killer (NK) and T-cell therapeutics. AutoScreen recovered NK-resistance genes that were initially lower-ranked in a leukemia screen, moving MUC1, PDPN, and LRRC15 from raw ranks of 118, 81, 1384 to 5, 44, 659, respectively. In follow-up tumor killing assay with primary human NK cells, individual perturbation validated all three hits successfully. Next, in prospective benchmarking across two cytotoxic T-cell-killing screens, AutoScreen recovered 77.1% of ground-truth hits called by the consensus of two gold-standard analysis pipelines (FDR<0.1), achieving 8% improvement over the strongest general-purpose LLM agent baseline. Finally, we re-analyzed >500 public datasets to construct the AutoScreen Resource Hub, a growing knowledge base of pre-computed, annotated reports for CRISPR screens, RNA-seq differential expression, and gene-level UK Biobank genome-wide association studies. This agentic AI approach enables real-time genomics benchmark construction that are continuously updated, expanded, for evaluating frontier AI co-scientists. Overall, AutoScreen enables AI-powered target discovery to be more auditable, scalable, extensible, and reproducible.

bioengineering↗

bioRxiv2026

Perturb-ME: Scalable mechanism discovery from phenotype-enriched genome-wide screens

Perturb-seq enables pooled genetic screens with rich single-cell profiling readouts, but genome-scale profiling remains costly and may not be associated with other established functional characteristics. Moreover, as screens grow in size and complexity, interpreting the resulting data comprehensively is challenging and slow. Here, we introduce Perturb-seq with Marker Enrichment (Perturb-ME), which combines genome-scale CRISPR screening, phenotype-based enrichment and multimodal single-cell profiling. Applied to MHC-I cell surface protein expression in melanoma, Perturb-ME profiled HLA-low and HLA-high cells with matched RNA, surface-protein and guide measurements. A regulatory model with 221 impactful regulators affecting 1,998 responsive genes recovered seven coherent co-functional regulatory modules governing nine gene programs, including the canonical IFN{gamma}-MHC-I axis regulating an antigen-presentation and interferon-response program. Agentic interpretation of the entire model with an AI co-scientist linked additional modules to trafficking, proteostasis and chromatin regulation. Perturb-ME, along with agentic interpretation, provide a scalable framework for comprehensive functional discovery from phenotype-enriched genetic screens.

cell biology↗

bioRxiv2026

An Integrated Atlas of the Human Kidney Spanning Health and Diseases

The human kidney contains highly specialized cell populations. Despite numerous single-cell and single-nucleus transcriptomics studies, differences in cohorts, technologies, analytical pipelines, and annotation frameworks have limited the ability to define consensus kidney cell states, identify disease-associated populations and interpret kidney disease genetic susceptibility. Here, we assembled 18 human kidney single-cell and single-nucleus RNA-sequencing datasets spanning 232 donors and five major disease contexts into a uniformly processed and computationally integrated Human Kidney Cell Atlas (HKCA), comprising over one million high-quality cells (816,895) and nuclei (215,308). The HKCA resolves 63 cell types and 120 harmonized cell states, including rare epithelial and stromal populations associated with kidney diseases . Integration with spatial transcriptomics, intercellular communication networks, and human genetic association data further defined the anatomical context and disease relevance of these populations. The HKCA also provides a framework for automated annotation of independent kidney human and mouse datasets. Together, the HKCA establishes a comprehensive reference for human kidney biology, enabling disease interpretation and genetic risk localization at cellular resolution.

genomics↗

bioRxiv2026

Orion: Towards Lab Automation with Computer-Using Agents

Laboratory discovery increasingly depends on computational workflows that connect experimental data to analysis, interpretation and follow-up hypotheses. Yet these workflows remain constrained by labor-intensive use of specialized software, visual inspection through graphical user interfaces, and integration of knowledge across multiple sources. Here, we present Orion, a computer-using AI agent for biomedical image analysis and interpretation that moves towards lab automation by automating this computational layer of laboratory work. Orion combines large language models with terminal execution, GUI control and adaptive multi-step reasoning in a shared computing environment. It can inspect visual data, operate standard scientific software, mine web resources and conduct end-to-end analysis and interpretation workflows without requiring bespoke software integrations. Across benchmarks, Orion achieved over 90% accuracy on biomedical database and literature retrieval tasks, learned to use the popular tools CellProfiler and QuPath for quantitative analysis of cellular and tissue images, respectively, and facilitated autonomous discovery in experimental imaging data. In 100 hours of autonomous exploration of a large-scale perturbation imaging dataset, Orion generated 52 research reports, of which human scientist review prioritized 22 plausible mechanistic hypotheses. These results show that computer-using AI agents can substantially expand the reach of laboratory automation, providing a scalable and auditable route from experimental imaging data to quantitative analysis, reports and biologically grounded hypotheses. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=45 SRC="FIGDIR/small/732095v1_ufig1.gif" ALT="Figure 1"> View larger version (20K): org.highwire.dtl.DTLVardef@476af5org.highwire.dtl.DTLVardef@bf1ac4org.highwire.dtl.DTLVardef@765c18org.highwire.dtl.DTLVardef@983419_HPS_FORMAT_FIGEXP M_FIG Overview of Orion Orion operates within a digital lab environment, using both graphical user interfaces and terminals just like a human scientist. This dual approach allows Orion to interact seamlessly with scientific software while also viewing figures and web databases to capture their nuanced visual information. C_FIG

bioinformatics↗

bioRxiv2026

Scalable multimodal mapping of macrophage regulatory architecture by integrating optical and transcriptomic pooled screens

Understanding how genetic perturbations reshape cellular states requires measuring diverse phenotypic modalities at scale. Here, we present PerturbPair, a platform that combines parallel Perturb-Seq and optical pooled screening (OPS) in primary mouse bone marrow-derived macrophages stimulated with lipopolysaccharide. Profiling over 334,000 single-cell transcriptomes across [~]1,000 gene perturbations and 7.8 million imaging-phenotyped cells across [~]3,000 gene perturbations, we reveal high concordance between transcriptomic and optical perturbation signatures. While RNA and imaging phenotypes were broadly concordant, OPS demonstrated superior sensitivity for weak-effect perturbations owing to greater cell throughput, and captured post-transcriptional regulatory events--such as protein trapping and mTOR-dependent phosphorylation--that left no detectable transcriptional footprint. To exploit cross-modal relationships, we developed EB-MoCAVI, an empirical Bayesian variational inference framework that both imputes RNA profiles for perturbations measured only by imaging--effectively tripling our Perturb-Seq dataset in silico--and denoises transcriptomic estimates for perturbations with sparse cell coverage by leveraging matched imaging data as a regularizer. A secondary screen validated the imputed transcriptional profiles and corroborated the cytosolic iron-sulfur assembly pathway as a candidate restraint on macrophage interferon tone. Integrating measured and imputed profiles with rare-variant burden statistics from the UK Biobank identified disease-specific macrophage gene programs and causal regulatory nodes for monocyte counts, type 2 diabetes, and inflammatory bowel disease. PerturbPair establishes a generalizable framework for multimodal perturbation atlases, pointing toward quantitative, causally-informed cross-modal models of cellular behavior.

genomics↗

bioRxiv2026

Massively multiplex multimodal chemical screens at single-cell resolution

Recent applications of scRNA-seq for massively multiplexed chemical screens have enabled comprehensive profiling of drug responses at unprecedented scale and resolution. However, current assays remain limited to RNA readouts, lacking information on other phenotypic and mechanistic layers such as chromatin accessibility, protein abundance and post-translational modifications. Here, we introduce a scalable framework for multimodal chemical screens, combining parallel small-molecule perturbations with multimodal readouts. We extend existing experimental platforms into icCITE-plex and DOGMA-plex, enabling joint profiling of RNA, protein, and epigenomic responses to thousands of chemical perturbations in parallel. To systematically decode the regulatory circuitry underlying these responses, we develop MoCAVI, a contrastive analysis framework that disentangles the effect of small molecules from control variation in multimodal measurements, and PERCISTRA, a pipeline that infers causal links between chromatin accessibility and gene expression. Applied across ~410,000 primary T cells under ~2,800 conditions, our approach resolves compound-specific mechanisms, highlights off-target effects, and links chromatin accessibility changes to transcription factor networks in primary T cells. Our results establish a generalizable platform for profiling and analyzing cellular responses to chemical perturbations across multiple modalities.

genomics↗

bioRxiv2026

SCALLOPS: a scalable, integrated computational framework for Optical Pooled Screens

Optical pooled screens (OPS) link pooled genetic perturbations to high-dimensional image-based phenotypes at scale, but their widespread adoption is hindered by computational bottlenecks in processing terabyte-scale, multimodal image data. We present SCALLOPS, a unified, modular, and cloud-native computational framework that overcomes these bottlenecks. SCALLOPS implements a "well-centric" processing strategy that integrates robust stitching with a non-linear two-stage registration strategy, enabling accurate alignment of multi-magnification images, reliable single-cell genotype-phenotype linkage, and efficient morphological feature extraction. Benchmarking with public and newly-generated datasets demonstrated SCALLOPS superior performance over existing solutions. Crucially, SCALLOPS uniquely enables robust processing of 4x magnification in situ sequencing data, accelerating image acquisition by around six-fold. We applied SCALLOPS to an optical pooled screen investigating the estrogen receptor (ER) degrader vepdegestrant in a breast cancer cell line, successfully recovering its known mechanism of action, highlighting the value of OPS in translational research. SCALLOPS provides a scalable end-to-end solution, making large-scale OPS routine.

bioinformatics↗

bioRxiv2026

Spatiotemporal profiling reveals the role of inflammatory niche in driving prostate cancer

Prostate cancer (PCa) is a lethal malignancy that displays profound resistance to immune checkpoint blockade (ICB), via mechanisms that are poorly understood. Here, we investigate the causes of CD8 T cell exhaustion and mechanisms of tumor progression in a PCa animal model, by single cell and spatial profiling, along a time course, following orthotopic transplantation of RB1/TP53/PTEN-deficient mouse organoids, competent to express neoantigens. The resulting tumors were castration resistant, consisting of largely basal and L2 malignant cells with upregulated inflammatory gene programs, and a specific spatial distribution of macrophages, cancer associated fibroblast (CAF) subtypes, and CD8 T-cells that was not previously reported. Using Zman-seq, we demonstrate that the effector function of tumor-infiltrating CD8 T cells was rapidly impaired as early as 24hrs after their infiltration, likely driven by signals from proinflammatory macrophages, Ccl2-Jak2+ inflammatory CAFs, and malignant basal cells, thus driving resistance to ICB. Interestingly, dual blockade of JAK1/2 and PD1 induced potent anti-tumor effects in tumor epithelial cells, decreased malignant epithelial cells and pro-inflammatory macrophages, and increased the proportion of normal (Pi16+) fibroblasts in the TME. Our results underscore the therapeutic potential of targeting JAK1/2 to enhance the efficacy of ICB, providing a rationale for clinical investigation of this combination in PCa.

cancer biology↗

bioRxiv2026

Unified Transcriptome and Mechanics Map of the Intact Mammalian Preimplantation Embryo In Situ

The development and maintenance of multicellular tissues requires that cell states be closely coupled to their local environment, including geometric and mechanical cues. However, studying this coupling in intact tissues has been challenging, because existing measurement technologies cannot simultaneously assess mechanical properties and molecularly defined cell states. To address this gap, we introduce the Unified Transcriptome and Mechanics Map (UTMM), a method for concurrent measurement of the transcriptome and cytoplasmic stiffness (high-frequency elastic modulus) within intact 3D multicellular structures. UTMM relies on two innovations: (1) a targeted in situ RNA sequencing approach for intact 3D embryos (3DISS), and (2) a strategy that leverages high-frequency intracellular organelle fluctuations to infer cell-level stiffness in vivo. We applied UTMM to mouse embryos from the zygote through the morula stage to characterize how RNA expression and cytoplasmic stiffness become coupled during early lineage specification. Our data reveal that, in the early morula, transcriptional and morphological distinctions emerge between trophectoderm and inner cell mass (ICM) lineages, coinciding with a gradual decrease in cytoplasmic stiffness (softening) across all cells from the 2-cell through morula stages. Furthermore, we observe that early lineage biases align with differential mechanical properties, reflecting distinct emerging developmental programs. When we delayed this softening process via mechanical perturbation, embryonic progression was impeded, highlighting the functional importance of coordinated mechanical and transcriptional changes. Together, these results demonstrate UTMMs ability to bridge molecular and mechanical dynamics in multicellular systems, providing a powerful framework for investigating how biomechanical cues shape cell fate decisions in intact tissues. One Sentence SummaryJoint 3D in situ quantification of spatial transcriptome and cytoplasm mechanics in preimplantation mammalian embryos

developmental biology↗

bioRxiv2025

DISCO-seq: 3D single-cell transcriptomics of intact biological systems

Single-cell transcriptomics has transformed tissue analysis, yet current methods struggle to integrate whole-tissue 3D architecture. Conventional techniques restrict molecular profiling to pre-selected 2D sections, losing systemic context and introducing anatomical bias by sampling less than 0.001% of a whole organism. To overcome these challenges, we developed DISCO-seq, a tissue-clearing chemistry that enables superior RNA accessibility compared to fresh or fixed tissues. DISCO-seq integrates whole-organ or organism 3D imaging with both untargeted and targeted transcriptomics, yielding high-quality RNA from cleared tissues comparable to standard samples. We demonstrate its versatility by investigating tumor heterogeneity in a syngeneic glioblastoma mouse model, using 3D imaging to identify spatially distinct microenvironments and characterize their unique transcriptomic signatures. Moreover, DISCO-seq enabled unbiased, whole-body mapping of SARS-CoV-2 S1 protein deposition in mice, followed by transcriptomic profiling of spatially defined niches. By bridgingmesoscale 3D imaging with single-cell transcriptomics, DISCO-seq establishes a paradigm for anatomically contextualized, hypothesis-free tissue interrogation. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=111 SRC="FIGDIR/small/693352v1_ufig1.gif" ALT="Figure 1"> View larger version (69K): org.highwire.dtl.DTLVardef@758ee5org.highwire.dtl.DTLVardef@1f862c6org.highwire.dtl.DTLVardef@1cf3b9org.highwire.dtl.DTLVardef@c5082e_HPS_FORMAT_FIGEXP M_FIG C_FIG HighlightsO_LIDISCO-seq integrates RNA-preserving tissue-clearing chemistry with whole-organ or organism 3D imaging, enabling anatomically unbiased single-cell transcriptomics. C_LIO_LIDISCO-seq yields RNA quality and transcriptome profiles equivalent to those obtained from matched fresh or fixed tissues. C_LIO_LIDISCO-seq identifies discrete glioblastoma microenvironments and defines their transcriptomic states within the intact brain. C_LIO_LIDISCO-seq enables whole-body mapping of SARS-CoV-2 S1 and uncovers region-specific immune and metabolic responses across anatomical niches. C_LI Supplementary movies can be seen at: http://discotechnologies.org/DISCO-seq/

systems biology↗

bioRxiv2025

Human brain organoids record the passage of time over multiple years in culture

The human brain develops and matures over an exceptionally prolonged period of time that spans nearly two decades of life. Processes that govern species-specific aspects of human postnatal brain development are difficult to study in animal models. While human brain organoids offer a promising in vitro model, they have thus far been shown to largely mimic early stages of brain development. Here, we developed human brain organoids for an unprecedented 5 years in culture, optimizing growth conditions able to extend excitatory neuron viability beyond previously-known limits. Using module scores of maturation-associated genes derived from a time course of endogenous human brain maturation, we show that brain organoids transcriptionally age with cell type-specificity through these many years in culture. Whole-genome methylation profiling reveals that the predicted epigenomic age of organoids sampled between 3 months and 5 years correlates precisely with time spent in vitro, and parallels epigenomic aging in vivo. Notably, we show that in chimeric organoids generated by mixing neural progenitors derived from "old" organoids with progenitors from "young" organoids, old progenitors rapidly produce late neuronal fates, skipping the production of earlier neuronal progeny that are instead produced by their young counterparts in the same co-cultures. The data indicate that human brain organoids can mature and record the passage of time over many years in culture. Progenitors that age in organoids retain a memory of the time spent in culture reflected in their ability to execute age-appropriate, late developmental programs.

neuroscience↗

bioRxiv2025

Gene-embedding-based prediction and functional evaluation of perturbation expression responses with PRESAGE

Understanding the impact of genetic perturbations on cellular behavior is crucial for biological research, but comprehensive experimental mapping remains infeasible. We introduce PRESAGE (Perturbation Response EStimation with Aggregated Gene Embeddings), a simple, modular, and interpretable framework that predicts perturbation-induced expression changes by integrating diverse knowledge sources via gene embeddings. PRESAGE transforms gene embeddings through an attention-based model to predict perturbation expression outcomes. To assess model performance, we introduce a comprehensive evaluation suite with novel functional metrics that move beyond traditional regression tasks, including measures of accuracy in effect size prediction, in identifying perturbations with similar expression profiles (phenocopy), and in prediction of perturbations with the strongest impact on specific gene set scores. PRESAGE outperforms existing methods in both classical regression metrics and our novel functional evaluations. Through ablation studies, we demonstrate that knowledge source selection is more critical for predictive performance than architectural complexity, with cross-system Perturb-seq data providing particularly strong predictive power. We also find that performance saturates quickly with training set size, suggesting that experimental design strategies might benefit from collecting sparse perturbation data across multiple biological systems rather than exhaustive profiling of individual systems. Overall, PRESAGE establishes a robust framework for advancing perturbation response prediction and facilitating the design of targeted biological experiments, significantly improving our ability to predict cellular responses across diverse biological systems.

bioinformatics↗

bioRxiv2025

Biomni: A General-Purpose Biomedical AI Agent

Biomedical research underpins progress in our understanding of human health and disease, drug discovery, and clinical care. However, with the growth of complex lab experiments, large datasets, many analytical tools, and expansive literature, biomedical research is increasingly constrained by repetitive and fragmented workflows that slow discovery and limit innovation, underscoring the need for a fundamentally new way to scale scientific expertise. Here, we introduce Biomni, a general-purpose biomedical AI agent designed to autonomously execute a wide spectrum of research tasks across diverse biomedical subfields. To systematically map the biomedical action space, Biomni first employs an action discovery agent to create the first unified agentic environment - mining essential tools, databases, and protocols from tens of thousands of publications across 25 biomedical domains. Built on this foundation, Biomni features a generalist agentic architecture that integrates large language model (LLM) reasoning with retrieval-augmented planning and code-based execution, enabling it to dynamically compose and carry out complex biomedical workflows - entirely without relying on predefined templates or rigid task flows. Systematic benchmarking demonstrates that Biomni achieves strong generalization across heterogeneous biomedical tasks - including causal gene prioritization, drug repurposing, rare disease diagnosis, micro-biome analysis, and molecular cloning - without any task-specific prompt tuning. Real-world case studies further showcase Biomnis ability to interpret complex, multi-modal biomedical datasets and autonomously generate experimentally testable protocols. Biomni envisions a future where virtual AI biologists operate alongside and augment human scientists to dramatically enhance research productivity, clinical insight, and healthcare. Biomni is ready to use at https://biomni.stanford.edu, and we invite scientists to explore its capabilities, stress-test its limits, and co-create the next era of biomedical discoveries.

bioinformatics↗

bioRxiv2025

PerTurboAgent: A Self-Planning Agent for Boosting Sequential Perturb-seq Experiments

Understanding how genetic interventions affect a cells phenotype is key to revealing causal gene regulatory mechanisms and finding novel drug targets. Pooled, high-content perturbation screening methods like Perturb-seq allow us to assess the impact of each of a large number of genetic interventions on a rich cellular profile of RNA or other features, facilitating such discoveries. However, the overall scale of perturbations, especially when considering combinations of genes for perturbation, cannot be tackled exhaustively in the lab. An alternative is to use an iterative design: By leveraging the modularity and sparsity of gene circuits along with prior biological knowledge, we can predict the impact of unseen genetic perturbations on these profiles and group genes with similar effects into co-functional modules, followed by a new round of Perturb-Seq to test these predictions and improve the overall performance of the model. These iterative cycles of experiment and prediction allow prioritizing genes for testing, maximizing the knowledge gleaned from fixed experimental resources, and opening the way to learn general predictive models. Designing these experiments requires a system that can analyze a cellular system, incorporate new and existing knowledge, use statistical tools, predict the effects of unseen perturbations, and prioritize the set of perturbations for the next iteration. These can be time-consuming tasks for scientists and require multiple different skills. Here, we developed PerTurboAgent, an LLM-based agent that excels in predicting candidate gene panels for iterative Perturb-Seq experiments through self-directed data analysis and knowledge retrieval. We evaluated PerTurboAgent based on its ability to identify genes with a phenotypic impact on gene expression upon perturbation in genome-scale perturbation data. PerTurboAgent outperforms existing agent-based and active learning strategies, offering an efficient and understandable approach to designing sequential perturbation experiments.

bioinformatics↗

bioRxiv2025

Prevalent cross-cell type QTL trans-regulatory genetic effects impacting innate lymphoid cells in the small intestine

Tissue function in homeostasis and disease arise from coordinated interactions between different cell types and can vary between individuals in a population, in part due to the impact of genetic variants. In barrier tissues such as the gut, innate lymphoid cells (ILCs) play critical roles in maintaining tissue homeostasis and immunity, but the genetic basis and associated regulatory circuitry affecting tissue ILCs remain largely unknown. Here, we systematically mapped the regulatory functions of genetic variants across 273,370 gut ILCs profiled by single cell RNA-seq from 274 diversity outbred (DO) mice. Computational analysis identified quantitative trait loci (QTL) impacting ILCs at three levels: cell subtype-specific mRNA expression (local and distal), cell subset proportions, and ILC gene expression programs (in trans), of which we experimentally validated a pivotal role for the transcription factor Rbpj in regulating ILC3s. All three classes of ILC QTL display polygenic inheritance signatures and broadly overlap with QTL affecting peripheral cytokine levels that we measured in another 261 DO mice. Strikingly, nearly half of trans-QTL affecting ILC traits did not overlap (within a 50-kb window) genes expressed in ILCs, and such loci were enriched for genes expressed by non-ILCs especially enteric neurons and intestinal glia. This suggests that such loci may act instead across cell type boundaries and allowed us to recover causal trans-cell type regulatory circuits in the tissue, including loci encoding multiple neuron-expressed peptides, epithelial-cell expressed Sox9, and phagocyte-expressed Ccl17. Finally, human orthologs of genes in QTL impacting gene expression and cell proportions are enriched for autoimmune disease risk heritability, suggesting their relevance to human disease. Our findings highlight how the impact of genetic variants may propagate to maintain tissue homeostasis and show a path to understand causality in tissue biology using quantitative genetics.

genetics↗

bioRxiv2025

CellFlow enables generative single-cell phenotype modeling with flow matching

High-content phenotypic screens provide a powerful strategy for studying biological systems, but the scale of possible perturbations and cell states makes exhaustive experiments unfeasible. Computational models that are trained on existing data and extrapolate to correctly predict outcomes in unseen contexts have the potential to accelerate biological discovery. Here, we present CellFlow, a flexible framework based on flow matching that can model single cell phenotypes induced by complex perturbations. We apply CellFlow to various phenotypic screens, accurately predicting expression responses to a wide range of perturbations, including cytokine stimulation, drug treatments and gene knockouts. CellFlow successfully modeled developmental perturbations at the whole-embryo scale and guided cell fate and organoid engineering by predicting heterogeneous cell populations arising from combinatorial morphogen treatments and by performing a virtual organoid protocol screen. Taken together, CellFlow has the potential to accelerate discovery from phenotypic screens by learning from existing data and generating phenotypes induced by unseen conditions.

bioinformatics↗

bioRxiv2025

SpatialAgent: An Autonomous AI Agent for Spatial Biology

Advances in AI are transforming scientific discovery, yet spatial biology, a field that deciphers the molecular organization within tissues, remains constrained by labor-intensive workflows. Here, we present SpatialAgent, an autonomous AI agent for spatial biology research. SpatialAgent couples large language models with a Plan-Act-Conclude architecture, dynamic tool and skill retrieval, multimodal interpretation, and verification modules that audit generated claims. It supports the full discovery loop, from gene-panel design and multimodal annotation to trajectory inference, cell-cell communication analysis, imputation, and hypothesis generation. Across human and mouse brain, heart, tonsil, colon, and prostate datasets, SpatialAgent outperformed established computational baselines and matched or surpassed expert scientists in key tasks. In open-ended case studies, it recovered known tissue organization and generated spatially grounded hypotheses. In a prospective mouse prostate cancer Xenium study, it designed a compact 100-gene add-on panel that profiled 4.2 million cells across 21 samples, improved cell-type and malignant-state prediction, and captured spatially structured tumor and microenvironment programs. SpatialAgent establishes a framework for autonomous and collaborative discovery in spatial biology.

bioinformatics↗