1758-Single-Cell Reference Atlas Construction Toolbox

The paper introduce SQUINT, a graph vector-quantized variational autoencoder designed to process spatially-resolved transcriptomics data by simultaneously learning distinct cell and niche codebooks. This framework utilizes residual vector quantization to capture both cell-intrinsic identities and broader microenvironmental contexts in a hierarchical discrete token format. By incorporating domain-specific training objectives such as adjacency, contrastive, and batch-correction losses, the model successfully merges spatial structural information across multiple tissue sections and assays. Extensive benchmarking across diverse datasets demonstrates that SQUINT either outperforms or remains competitive with leading continuous representation and single-cell foundation models on tasks ranging from cellular identification to cross-section integration. Furthermore, the resulting discrete vocabulary facilitates advanced biological applications, including gene-expression imputation in hidden regions and streamlined query-to-reference mapping for complex tissue atlases.References:Birk S, Merchant A, Vahidi A, et al. Learning Discrete Cell and Niche Codes from Spatial Transcriptomics Using Dual Residual Vector Quantization[J]. bioRxiv, 2026: 2026.08. 07.743490.前往小宇宙评论区与主播互动

The single-cell Atlas Toolbox (scAtlasTb) is a modular, Python-based framework designed to streamline and standardize the iterative, complex process of constructing large-scale single-cell reference atlases. Built using Snakemake and integrated with the scverse ecosystem, the toolbox provides a robust infrastructure for data preparation, quality control, workflow management, and downstream analysis. To address the historical lack of uniform standards in atlas building, scAtlasTb incorporates quantitative benchmarking metrics—such as Moran's I for label-free biological conservation and graph dissimilarity scores—to systematically evaluate methodological choices. Through case studies on the Human Lung Cell Atlas, Human Retina Cell Atlas, and Human Cell Atlas Reference for PBMCs, the authors demonstrate how this framework optimizes lineage-specific quality control, batch variable definition, and feature selection strategies. Ultimately, scAtlasTb transforms reference atlas creation into a transparent, reproducible, and evidence-driven procedure capable of scaling to millions of cells while preserving subtle biological signals.References:Mueller M F, Cujba A M, Romanovskaia D, et al. Building optimized single-cell reference atlases with scAtlasTb[J]. bioRxiv, 2026: 2026.07. 30.741695.前往小宇宙评论区与主播互动

The paper introduces TERRA, a spatial foundation model pretrained on 112 million human cells using a joint-embedding predictive architecture to analyze imaging-based spatial transcriptomics. By tokenizing each cell alongside its local environment, the model simultaneously generates gene-, cell-, and neighborhood-level embeddings from a single pretrained backbone without requiring task-specific fine-tuning. This multi-scale framework supports zero-shot tissue analysis, cross-platform transfer, and in silico gene perturbations that predict cellular and microenvironmental responses to injury or therapeutic intervention. Furthermore, the sources demonstrate TERRA's practical utility by uncovering a developmental precursor of islet microvasculature during pancreas organogenesis, predicting a checkpoint-blockade injury program in the human kidney, and cataloging conserved macrophage niche archetypes across multiple organs and disease states.References:Birk S, Sanian M V, Vahidi A, et al. Multi-scale modeling of human tissues from spatial transcriptomics with TERRA[J]. BioRxiv, 2026: 2026.07. 29.741565.前往小宇宙评论区与主播互动

The paper introduces the omnigenic neural network (OGM), a biologically structured architecture designed for human genome-scale phenotype prediction using the Gene Ontology (GO) as an inductive bias. By integrating individual-level genotypes, sex, and genetic principal components into a hierarchical gene and ontology layer, the model effectively captures non-additive interactions, such as epistasis, and accommodates multimodal data like metabolomics. Evaluated across large cohorts including the UK Biobank and the All of Us cohort, the OGM matches or outperforms traditional polygenic scores (PGS) and baseline models, especially in multilabel predictions and cross-ancestry transferability. Furthermore, the architecture provides systems-level interpretability and captures eQTL structures through methods like Integrated Gradients and Integrated Hessians, establishing a flexible framework for advanced genomic research.References:Upmeier zu Belzen J, Arnoldt L, Hollmann N, et al. An interpretable omnigenic neural network architecture for the human genome[J]. medRxiv, 2026: 2026.07. 28.26359187.前往小宇宙评论区与主播互动

Researchers created a multimodal cellular and spatial atlas comparing Broca’s area and the motor cortex across human development from infancy to adulthood to understand the biological foundations of language. By combining single-nucleus RNA sequencing, chromatin accessibility profiling, and spatial transcriptomics, the investigation mapped distinct cell types, gene expression programs, and structural architectures within these critical brain regions. The findings revealed area-specific cellular signatures, including unique glial populations, interneuron subtypes, and an early molecular code driven by cadherin genes that facilitates precise neural communication. Additionally, the study linked specific cell types, such as somatostatin-positive interneurons, to increased vulnerability for neurodevelopmental disorders like autism spectrum disorder and attention-deficit/hyperactivity disorder. Finally, evolutionary analyses of differentially accessible chromatin regions suggested that genetic adaptations supporting human linguistic abilities accumulated gradually over millions of years following the divergence of human and chimpanzee lineages.References:Moriano J, Mukhtar T, Lee J T H, et al. A multimodal interrogation of Broca’s area in the pediatric and adult human brain[J]. bioRxiv, 2026: 2026.07. 22.739934.前往小宇宙评论区与主播互动

The paper introduces RisQ, an innovative transformer-based framework designed to map human health by integrating multimodal data, including genetics, lifestyle factors, and medical histories. By treating health as a unified and interconnected system rather than isolated conditions, this model effectively predicts disease risks across a wide array of endpoints and continuous time horizons. Extensive evaluations using datasets from the UK Biobank and the All of Us cohort demonstrate that RisQ significantly outperforms traditional disease-specific and multi-disease forecasting models. Furthermore, the framework successfully generalizes to unseen medical categories and maintains high predictive performance even in low-prevalence disease regimes. Ultimately, the research illustrates that scaling input modalities, diseases, and time horizons enhances information transfer, leading to superior risk stratification and valuable biological discoveries.References:Hager P, Roth B, Buehler N, et al. Learning the shared structure of human health across diseases, modalities, and time[J]. medRxiv, 2026: 2026.07. 07.26357373.前往小宇宙评论区与主播互动

PertOmni is a novel multimodal representation-learning framework designed to analyze cellular perturbation effects across both genetic and small-molecule experiments. By employing a CLIP-style architecture with masked contrastive learning, the system aligns transcriptomic profiles and microscopic cell images with large language model-derived text embeddings of biological descriptions. This approach effectively isolates within-cell-type perturbation dynamics while minimizing confounding heterogeneity across different cell lines. The framework demonstrates robust predictive capabilities in bi-directional cross-modality retrieval, drug-gene interaction inference, and retrieval-based perturbation prediction. Ultimately, this method provides a unified computational strategy to link interventions directly to their functional targets and improve downstream drug discovery applications.References:Long W, Liu T, Szałata A, et al. Learning Perturbation Effects Through Contrastive Alignment of Multimodal Biological Embeddings[J]. bioRxiv, 2026: 2026.06. 23.734145.前往小宇宙评论区与主播互动

The paper introduce Phoenix, an advanced generative artificial intelligence framework designed to predict single-cell spatial gene expression directly from routine histology images. By scaling training across a massive, highly curated dataset named The Nest, this system successfully overcomes the high costs and slow speeds traditionally associated with spatial transcriptomics technologies. The research demonstrates that Phoenix can generalize exceptionally well zero-shot across diverse, previously unseen biological cohorts, organs, and species without requiring fine-tuning. Furthermore, the framework enables powerful in silico analyses, facilitating population-scale cancer atlases, precise biological hypothesis testing, and the discovery of actionable clinical biomarkers for predicting therapeutic responses.References:Tran M, Gindra R H, Putze P, et al. Pan-cancer virtual spatial transcriptomics from routine histology with Phoenix[J]. bioRxiv, 2026: 2026.04. 25.720812.前往小宇宙评论区与主播互动

The paper reviews the microevolutionary processes that drive the independent evolution of multicellularity across various life forms. It breaks down this transformative transition into key challenges, including group formation, the attainment of Darwinian individuality, and the development of functional integration and cellular specialization. Through comparative genomics and laboratory evolution experiments, the authors demonstrate that multicellular traits often emerge readily by repurposing ancestral unicellular machinery or through biophysical and ecological scaffolding. Finally, the sources explore how environmental shifts and genomic architecture have influenced why complex multicellularity remains a rare, highly contingent phenomenon.References:Ozan Bozdag G, Tong K, Yunker P J, et al. Overcoming key challenges in the evolutionary transition to multicellularity[J]. Nature, 2026, 658(8135): 327-336.前往小宇宙评论区与主播互动

Recent scientific findings demonstrate that human antiviral immunity and bacterial anti-phage defense mechanisms share remarkably similar evolutionary origins, structures, and functional strategies across all kingdoms of life. Both human cells and bacteria employ conserved protein pathways, such as cGAS–STING and inflammasomes, alongside nucleotide-based immune signaling molecules to detect and disrupt viral replication. Furthermore, these cellular defense systems utilize comparable mechanisms for pathogen recognition, signal amplification, and targeted effector functions like nucleic acid degradation and membrane disruption. Although human cells prioritize early innate and adaptive responses while bacteria often deploy late-stage cell death as a protective barrier, the overall architectural rules of host-virus interactions remain fundamentally unified.References:Kranzusch P J. Shared principles of human and bacterial antiviral immunity[J]. Nature, 2026, 656(8127): 307-319.前往小宇宙评论区与主播互动