A human cell atlas of fetal gene expression

lncRNAlyzr: Enrichment Analysis for lncRNA Sets

lncRNAs make up a large portion of the human genome affecting many biological processes in normal physiology and diseases. However, human lncRNAs are understudied compared to protein-coding genes. While there are many tools for performing gene set enrichment analysis for coding genes, few tools exist for lncRNA enrichment analysis. lncRNAlyzr is a webserver application designed for lncRNAs enrichment analysis. lncRNAlyzr has a database containing 33 lncRNA set libraries created by computing correlations between lncRNAs and annotated coding gene sets. After users submit a set of lncRNAs to lncRNAlyzr, the enrichment analysis results are visualized as ball-and-stick subnetworks where nodes are lncRNAs connected to enrichment terms from across selected lncRNA set libraries. To demonstrate lncRNAlyzr, it was used to analyze the effects of knocking down the lncRNA CYTOR in K562 cells. Overall, lncRNAlyzr is an enrichment analysis tool for lncRNAs aiming to further our understanding of lncRNAs functional modules. lncRNAlyzr is available from: https://lncrnalyzr.maayanlab.cloud.

scCorrect: Cross-modality label transfer from scRNA-seq to scATAC-seq using domain adaptation

Cell type annotation in single-cell chromatin accessibility sequencing (scATAC-seq) is crucial for enabling researchers to identify subpopulations of cells associated with specific diseases, elucidate gene regulatory networks, and discover markers indicative of disease states. The prevailing approach for cell type annotation in single-cell research involves transferring well-delineated cell types from single-cell RNA sequencing (scRNA-seq) data to scATAC-seq data using a label propagation algorithm. However, the inherent modal discrepancies (i.e.biological interpretation) between scRNA-seq and scATAC-seq data, coupled with the intrinsic sparsity and high dimensionality of scATAC-seq data, pose significant challenges to the efficacy of this strategy. To address these challenges, we introduce a novel neural network framework, scCorrect, which operates in two distinct phases. In the first phase, scCorrect aligns the scRNA-seq and scATAC-seq datasets, generating initial annotation results. The second phase involves training a corrective network specifically designed to amend any erroneous annotations produced during the first phase. Empirical tests across multiple datasets have demonstrated that scCorrect consistently achieves superior recognition accuracy, underscoring its significant potential to enhance disease-related research in humans.

SMOPCA: spatially aware dimension reduction integrating multi-omics improves the efficiency of spatial domain detection

Technological advances have enabled us to profile multiple omics layers with spatial information, significantly enhancing spatial domain detection and advancing a variety of biomedical research fields. Despite these advancements, there is a notable lack of effective methods for modeling spatial multi-omics data. We introduce SMOPCA, a Spatial Multi-Omics Principal Component Analysis method designed to perform joint dimension reduction on multimodal data while preserving spatial dependencies. Extensive experiments reveal that SMOPCA outperforms existing single-modal and multimodal dimension reduction and clustering methods, across both single-cell and spatial multi-omics datasets derived from diverse technologies and tissue structures.

Sequence diversity lost in early pregnancy

Every generation, the human genome is shuffled during meiosis and a single fertilized egg gives rise to all of the cells of the body1. Meiotic errors leading to chromosomal abnormalities are known causes of pregnancy loss2,3, but genetic aetiologies of euploid pregnancy loss remain largely unexplained4. Here we characterize sequence diversity in early pregnancy loss through whole-genome sequencing of 1,007 fetal samples and 934 parental samples from 467 trios affected by pregnancy loss (fetus, mother and father). Sequenced parental genomes enabled us to determine both the parental and meiotic origins of chromosomal abnormalities, detected in half of our set. It further enabled us to assess de novo mutations on both homologous chromosomes from parents transmitting extra chromosomes, and date them, revealing that 6.6% of maternal mutations occurred before sister chromatid formation in fetal oocytes. We find a similar number of de novo mutations in the trios affected by pregnancy loss as in 9,651 adult trios, but three times the number of pathogenic small (<50 bp) sequence variant genotypes in the loss cases compared with adults. Overall, our findings indicate that around 1 in 136 pregnancies is lost due to a pathogenic small sequence variant genotype in the fetus. Our results highlight the vast sequence diversity that is lost in early pregnancy.

A single-cell and tissue-scale analysis suite resolves Mixl1's role in heart development

Perturbation studies using gene knockouts have become a key tool for understanding the roles of regulatory genes in development. However, large-scale studies dissecting the molecular role of development master regulators in every cell type throughout the embryo are technically challenging and scarce. Here, we systematically characterize the knockout effects of the key developmental regulators T/Brachyury and Mixl1 in gastrulation and early organogenesis using single-cell profiling of chimeric mouse embryos. For the analysis of these experimental data, we present COSICC, an effective suite of statistical tools to characterize perturbation effects in complex developing cell populations. We gain insights into T's role in lateral plate mesoderm, limb development, and posterior intermediate mesoderm specification. Furthermore, we generate Mixl1 -/- embryonic chimeras and reveal the role of this key transcription factor in discrete mesoderm lineages, in particular concerning developmental dysregulation of the recently identified juxta-cardiac field.

The path winds along isolation and analyses of fetal nucleated red blood cells in maternal peripheral blood: Past, present, and future toward non-invasive prenatal diagnosis

Traditional prenatal diagnosis detects fetal disorders through invading uterus to access fetal cells, which may cause maternal complications, fetal injury, or even miscarriage. Safe and convenient non-invasive prenatal testing (NIPT) by analyzing fetal materials (cell-free DNA/RNA, cells, and extracellular vesicles) that circulate in maternal peripheral blood attracts great attention and has been applied in risk evaluation of several fetal disorders. Among those fetal analytes, fetal nucleated red blood cells (fNRBCs) comprise entire fetal genome, possess distinct membrane antigens, and have a lifespan limited in every single gestation. They were once expected to be an ideal biomarker for NIPT and even definitive prenatal diagnosis. However, recent advances of fNRBC-based NIPT are limited and their applications toward clinical practices are still challenging. Herein, we comprehensively overview research on fNRBCs in maternal peripheral blood, trying to dissect current predicament and inspire potential solutions. The source and lineage of fNRBCs, their entrance into maternal peripheral blood, and their physiochemical characteristics are discussed, and various strategies of label-free or immuno-affinitive isolation and subsequential identification of fNRBCs from maternal blood cells are summarized. Although proof-of-concept analyses toward detecting a few fetal disorders are demonstrated, current fNRBC-based NIPT still suffers many challenges when applied to clinical practices. Nevertheless, via thorough investigation and new analytical technologies, it is believed fNRBC-based NIPT will provide a promising platform to supplement the insufficiency of current strategies.

Machine learning approaches enable the discovery of therapeutics across domains

Multi-modal datasets have grown exponentially in the last decade. This has created an enormous demand for machine learning models that can predict complex outcomes by leveraging cellular, molecular, and humoral profiles. Corresponding inference of mechanisms can help to uncover new therapeutic targets. Here, we discuss how biological principles guide the design of predictive models and how interpretable machine learning can lead to novel mechanistic insights. We provide descriptions of multiple learning techniques and how suited they are to domain adaptations. Finally, we talk about broad learning capabilities of foundation models on large datasets and whether they can be used to provide meaningful inference about biological datasets.

Single-cell MultiOmics and spatial transcriptomics demonstrate neuroblastoma developmental plasticity

Neuroblastoma, the most prevalent extracranial pediatric solid tumor, arises from neural crest progeny cells. It exhibits substantial developmental plasticity and intratumoral heterogeneity, leading to survival rates below 50% in high-risk cases. The regulatory mechanisms underlying this plasticity remain largely elusive. In this integrative study, we used single-cell MultiOmics from a mouse spontaneous tumor model and spatial transcriptomics from human patient samples to dissect the transcriptional and epigenetic landscapes that govern developmental states in neuroblastoma. We identified developmental intermediate states in high-risk neuroblastomas critical for malignant transitions and uncovered extensive epigenetic priming with latent capacity for diverse state transitions. Furthermore, we mapped enhancer gene regulatory networks (eGRNs) and tumor microenvironments sustaining these aggressive states. State transitions and malignancy could be interfered with by targeting transcription factors controlling the eGRNs.

A unified analysis of atlas single-cell data

Recent efforts to generate atlas-scale single-cell data provide opportunities for joint analysis across tissues and modalities. Existing methods use cells as the reference unit, hindering downstream gene-based analysis and removing genuine biological variation. Here we present GIANT, an integration method designed for atlas-scale gene analysis across cell types and tissues. GIANT converts data sets into gene graphs and recursively embeds genes without additional alignment. Applying GIANT to two recent atlas data sets yields unified gene-embedding spaces across human tissues and data modalities. Further evaluations demonstrate GIANT's usefulness in discovering diverse gene functions and underlying gene regulation in cells from different tissues.

scaLR: a low-resource deep neural network-based platform for single cell analysis and biomarker discovery

Single-cell ribonucleic acid (RNA) sequencing (scRNA-seq) produces vast amounts of individual cell profiling data. Its analysis presents a significant challenge in accurately annotating cell types and their associated biomarkers. Different pipelines based on deep neural network (DNN) methods have been employed to tackle these issues. These pipelines have arisen as a promising resource and can extract meaningful and concise features from noisy, diverse, and high-dimensional data to enhance annotations and subsequent analysis. Existing tools require high computational resources to execute large sample datasets. We have developed a cutting-edge platform known as scaLR (Single-cell analysis using low resource) that efficiently processes data into feature subsets, samples in batches to reduce the required memory for processing large datasets, and running DNN models in multiple central processing units. scaLR is equipped with data processing, feature extraction, training, evaluation, and downstream analysis. Its novel feature extraction algorithm first trains the model on a feature subset and stores the importance of the features for all the features in that subset. At the end of the training of all subsets, the top-K features are selected based on their importance. The final model is trained on top-K features; its performance evaluation and associated downstream analysis provide significant biomarkers for different cell types and diseases/traits. Our findings indicate that scaLR offers comparable prediction accuracy and requires less model training time and computational resources than existing Python-based pipelines. We present scaLR, a Python-based platform, engineered to utilize minimal computational resources while maintaining comparable execution times and analysis costs to existing frameworks.

Heterogeneity of mature oligodendrocytes in the central nervous system

Mature oligodendrocytes form myelin sheaths that are crucial for the insulation of axons and efficient signal transmission in the central nervous system. Recent evidence has challenged the classical view of the functionally static mature oligodendrocyte and revealed a gamut of dynamic functions such as the ability to modulate neuronal circuitry and provide metabolic support to axons. Despite the recognition of potential heterogeneity in mature oligodendrocyte function, a comprehensive summary of mature oligodendrocyte diversity is lacking. We delve into early 20 th -century studies by Robertson and Río-Hortega that laid the foundation for the modern identification of regional and morphological heterogeneity in mature oligodendrocytes. Indeed, recent morphologic and functional studies call into question the long-assumed homogeneity of mature oligodendrocyte function through the identification of distinct subtypes with varying myelination preferences. Furthermore, modern molecular investigations, employing techniques such as single cell/nucleus RNA sequencing, consistently unveil at least six mature oligodendrocyte subpopulations in the human central nervous system that are highly transcriptomically diverse and vary with central nervous system region. Age and disease related mature oligodendrocyte variation denotes the impact of pathological conditions such as multiple sclerosis, Alzheimer's disease, and psychiatric disorders. Nevertheless, caution is warranted when subclassifying mature oligodendrocytes because of the simplification needed to make conclusions about cell identity from temporally confined investigations. Future studies leveraging advanced techniques like spatial transcriptomics and single-cell proteomics promise a more nuanced understanding of mature oligodendrocyte heterogeneity. Such research avenues that precisely evaluate mature oligodendrocyte heterogeneity with care to understand the mitigating influence of species, sex, central nervous system region, age, and disease, hold promise for the development of therapeutic interventions targeting varied central nervous system pathology.

An integrated transcriptomic cell atlas of human endoderm-derived organoids

Human pluripotent stem cells and tissue-resident fetal and adult stem cells can generate epithelial tissues of endodermal origin in vitro that recapitulate aspects of developing and adult human physiology. Here, we integrate single-cell transcriptomes from 218 samples covering organoids and other models of diverse endoderm-derived tissues to establish an initial version of a human endoderm-derived organoid cell atlas. The integration includes nearly one million cells across diverse conditions, data sources and protocols. We compare cell types and states between organoid models and harmonize cell annotations through mapping to primary tissue counterparts. Focusing on the intestine and lung, we provide examples of mapping data from new protocols and show how the atlas can be used as a diverse cohort to assess perturbations and disease models. The human endoderm-derived organoid cell atlas makes diverse datasets centrally available and will be valuable to assess fidelity, characterize perturbed and diseased states, and streamline protocol development.

Advances and applications in single-cell and spatial genomics

The applications of single-cell and spatial technologies in recent times have revolutionized the present understanding of cellular states and the cellular heterogeneity inherent in complex biological systems. These advancements offer unprecedented resolution in the examination of the functional genomics of individual cells and their spatial context within tissues. In this review, we have comprehensively discussed the historical development and recent progress in the field of single-cell and spatial genomics. We have reviewed the breakthroughs in single-cell multi-omics technologies, spatial genomics methods, and the computational strategies employed toward the analyses of single-cell atlas data. Furthermore, we have highlighted the advances made in constructing cellular atlases and their clinical applications, particularly in the context of disease. Finally, we have discussed the emerging trends, challenges, and opportunities in this rapidly evolving field.

Spatially resolved mapping of cells associated with human complex traits

Depicting spatial distributions of disease-relevant cells is crucial for understanding disease pathology1,2. Here we present genetically informed spatial mapping of cells for complex traits (gsMap), a method that integrates spatial transcriptomics data with summary statistics from genome-wide association studies to map cells to human complex traits, including diseases, in a spatially resolved manner. Using embryonic spatial transcriptomics datasets covering 25 organs, we benchmarked gsMap through simulation and by corroborating known trait-associated cells or regions in various organs. Applying gsMap to brain spatial transcriptomics data, we reveal that the spatial distribution of glutamatergic neurons associated with schizophrenia more closely resembles that for cognitive traits than that for mood traits such as depression. The schizophrenia-associated glutamatergic neurons were distributed near the dorsal hippocampus, with upregulated expression of calcium signalling and regulation genes, whereas depression-associated glutamatergic neurons were distributed near the deep medial prefrontal cortex, with upregulated expression of neuroplasticity and psychiatric drug target genes. Our study provides a method for spatially resolved mapping of trait-associated cells and demonstrates the gain of biological insights (such as the spatial distribution of trait-relevant cells and related signature genes) through these maps.

Mapping the developmental profile of ventricular zone-derived neurons in the human cerebellum

The cerebellar ventricular zone (VZ) is the primary source of progenitors that generate cerebellar GABAergic neurons, including Purkinje cells (PCs) and interneurons (INs). This study provides detailed characterization of human cerebellar GABAergic neurogenesis using transcriptomic and histopathological analyses and reveals conserved and unique features compared to rodents. We show that the sequential progression of neurogenesis is conserved and occurs before 8 postconception weeks. Notably, PC differentiation occurs in the outer subventricular zone (SVZ), a region absent in the mouse cerebellum. Human PCs are generated during a compact two-week period before the onset of cerebral cortex histogenesis. A subset of human PCs retain proliferative marker expression weeks after leaving the VZ, another feature not observed in rodents. Human PC maturation is protracted with an extensive migration and reorganization throughout development with dendritic arborization developing in late gestation. We define a continuous transcriptional cascade of PC development from neuroepithelial cells to mature PCs. In contrast, while human interneuronal progenitors are born beginning in early fetal development, they exhibit an even more protracted differentiation across late gestation and into postnatal ages. These findings show dynamic developmental process for human cerebellar GABAergic neurons and underscore the importance of the embryonic environment, with early disruptions having potentially significant impacts.

Fate and state transitions during human blood vessel organoid development

Human blood vessel organoids (hBVOs) have emerged as a system to model human vascular development and disease. Here, we use single-cell multi-omics together with genetic and signaling pathway perturbations to reconstruct hBVO development. Mesodermal progenitors bifurcate into endothelial and mural fates in vitro, and xenografted BVOs acquire definitive arteriovenous endothelial cell specification. We infer a gene regulatory network and use single-cell genetic perturbations to identify transcription factors (TFs) and receptors involved in cell fate specification, including a role for MECOM in endothelial and mural specification. We assess the potential of BVOs to generate organotypic states, identify TFs lacking expression in hBVOs, and find that induced LEF1 overexpression increases brain vasculature specificity. Finally, we map vascular disease-associated genes to hBVO cell states and analyze an hBVO model of diabetes. Altogether, we provide a comprehensive cell state atlas of hBVO development and illuminate the power and limitation of hBVOs for translational research.

Novel Directly Reprogrammed Smooth Muscle Cells Promote Vascular Regeneration as Microvascular Mural Cells

BACKGROUND

Although cell therapy has emerged as a promising approach to promote neovascularization, its effects are mostly limited to capillaries. To generate larger or more stable vessels, layering of mural cells such as smooth muscle cells (SMCs) or pericytes is required. Recently, direct reprogramming approaches have been developed for generating SMCs. However, such reprogrammed SMCs lack genuine features of contractile SMCs, a native SMC phenotype; thus, their therapeutic and vessel-forming potential in vivo was not explored. Therefore, we aimed to directly reprogram human dermal fibroblasts toward contractile SMCs (rSMCs) and investigated their role for generating vascular mural cells in vivo and their therapeutic effects on ischemic disease.

METHODS

We applied myocardin and all-trans retinoic acid with specific culture conditions to directly reprogram human dermal fibroblasts into rSMCs. We characterized their phenotype as contractile SMCs through quantitative reverse-transcriptase polymerase chain reaction, flow cytometry, and immunostaining. We then explored their contractility using a vasoconstrictor, carbachol, and through transmission electron microscope and bulk RNA sequencing. Next, we evaluated whether transplantation of rSMCs improves blood flow and induces vessel formation as mural cells in a mouse model of hindlimb ischemia with laser Doppler perfusion imaging and histological analysis. We also determined their paracrine effects.

RESULTS

Our novel culture conditions using myocardin and all-trans retinoic acid efficiently reprogrammed human dermal fibroblasts into SMCs. These rSMCs displayed characteristics of contractile SMCs at the mRNA, protein, and cellular levels. Transplantation of rSMCs into ischemic mouse hind limbs enhanced blood flow recovery and vascular repair and improved limb salvage. Histological examination showed that vascular density was increased and the engrafted rSMCs were incorporated into the vascular wall as pericytes and vascular SMCs, thereby contributing to formation of more stable and larger microvessels. Quantitative reverse-transcriptase polymerase chain reaction analysis revealed that these transplanted rSMCs exerted pleiotropic effects, including angiogenic, arteriogenic, vessel-stabilizing, and tissue regenerative effects, on ischemic limbs.

CONCLUSIONS

A combination of myocardin and all-trans retinoic acid in defined culture conditions efficiently reprogrammed human fibroblasts into contractile and functional SMCs. The rSMCs were shown to be effective for vascular repair and contributed to neovascularization through mural cells and various paracrine effects. These human rSMCs could represent a novel source for cell-based therapy and research.

TF Profiler: a transcription factor inference method that broadly measures transcription factor activity and identifies mechanistically distinct networks

TF Profiler is a method of inferring transcription factor (TF) regulatory activity, i.e., when a TF is present and actively participating in the regulation of transcription, directly from nascent sequencing assays such as PRO-seq and GRO-seq. While ChIP assays have measured DNA localization, they fall short of identifying when and where the effector domain of a transcription factor is active. Our method uses RNA polymerase activity to infer TF effector domain activity across hundreds of data sets and transcription factors. TF Profiler is broadly applicable, providing regulatory insights on any PRO-seq sample for any transcription factor with a known binding motif.

Cell-type-specific subtyping of epigenomes improves prognostic stratification of cancer

BACKGROUND

Most molecular classifications of cancer are based on bulk-tissue profiles that measure an average over many distinct cell types. As such, cancer subtypes inferred from transcriptomic or epigenetic data are strongly influenced by cell-type composition and do not necessarily reflect subtypes defined by cell-type-specific cancer-associated alterations, which could lead to suboptimal cancer classifications.

METHODS

To address this problem, we here propose the novel concept of cell-type-specific combinatorial clustering (CELTYC), which aims to group cancer samples by the molecular alterations they display in specific cell types. We illustrate this concept in the context of DNA methylation data of liver and kidney cancer, deriving in each case novel cancer subtypes and assessing their prognostic relevance against current state-of-the-art prognostic models.

RESULTS

In both liver and kidney cancer, we reveal improved cell-type-specific prognostic models, not discoverable using standard methods. In the case of kidney cancer, we show how combinatorial indexing of epithelial and immune-cell clusters define improved prognostic models driven by synergy of high mitotic age and altered cytokine signaling. We validate the improved prognostic models in independent datasets and identify underlying cytokine-immune-cell signatures driving poor outcome.

CONCLUSIONS

In summary, cell-type-specific combinatorial clustering is a valuable strategy to help dissect and improve current prognostic classifications of cancer in terms of the underlying cell-type-specific epigenetic and transcriptomic alterations.

Functional maturation of preterm intestinal epithelium through CFTR activation

Preterm birth disrupts intestinal epithelial maturation, impairing digestive and absorptive functions. This study integrates analysis of single-cell RNA sequencing datasets, spanning fetal to adult stages, with human preterm intestinal models derived from the ileal tissue of preterm infants. We investigate the potential of extracellular vesicles (EVs) derived from human Wharton's jelly mesenchymal stem cells to promote intestinal maturation. Distinct enterocyte differentiation trajectories are identified during the transition from immature to mature stages of human intestinal development. EV treatment, particularly with the EV39 line, significantly upregulates maturation-specific gene expression related to enterocyte function. Gene set enrichment analysis reveals an enrichment of TGFβ1 signaling pathways, and proteomic analysis identifies TGFβ1 and FGF2 as key mediators of EV39's effects. These treatments enhance cell proliferation, epithelial barrier integrity, and fatty acid uptake, primarily through CFTR-dependent mechanisms-unique to human preterm models, not observed in mouse intestinal organoids. This highlights the translational potential of EV39 and CFTR activation in promoting the functional maturation of the premature human intestine.

Tracing the Shared Foundations of Gene Expression and Chromatin Structure

The three-dimensional organization of chromatin into topologically associating domains (TADs) may impact gene regulation by bringing distant genes into contact. However, many questions about TADs' function and their influence on transcription remain unresolved due to technical limitations in defining TAD boundaries and measuring the direct effect that TADs have on gene expression. Here, we develop consensus TAD maps for human and mouse with a novel "bag-of-genes" approach for defining the gene composition within TADs. This approach enables new functional interpretations of TADs by providing a way to capture species-level differences in chromatin organization. We also leverage a generative AI foundation model computed from 33 million transcriptomes to define contextual similarity, an embedding-based metric that is more powerful than co-expression at representing functional gene relationships. Our analytical framework directly leads to testable hypotheses about chromatin organization across cellular states. We find that TADs play an active role in facilitating gene co-regulation, possibly through a mechanism involving transcriptional condensates. We also discover that the TAD-linked enhancement of transcriptional context is strongest in early developmental stages and systematically declines with aging. Investigation of cancer cells show distinct patterns of TAD usage that shift with chemotherapy treatment, suggesting specific roles for TAD-mediated regulation in cellular development and plasticity. Finally, we develop "TAD signatures" to improve statistical analysis of single-cell transcriptomic data sets in predicting cancer cell-line drug response. These findings reshape our understanding of cellular plasticity in development and disease, indicating that chromatin organization acts through probabilistic mechanisms rather than deterministic rules.

Software availability

https://singhlab.net/tadmap.

Decoding the human prenatal immune system with single-cell multi-omics

The human immune system is made up of a huge variety of cell types each with unique functions. Local networks of resident immune cells are poised to sense and protect against pathogen entry, whereas more widespread innate and adaptive immune networks provide first rapid, then long-lasting and targeted responses. However, how we develop such a diverse and complex system remains unknown. Studying human development directly has been challenging in the past, but recent advances in single-cell and spatial genomics, together with the co-ordinated efforts of the Human Cell Atlas and other initiatives, have led to new studies that map the development of the human immune system in unprecedented detail. In this Review, we consider the timings, transitions, cell types and tissue microenvironments that are crucial for building the human immune system. We also compare and contrast the human system with model species and in vitro systems, and discuss how an understanding of prenatal immune system development will improve our knowledge of human disease.

Scalable high-performance single cell data analysis with BPCells

The growth of single-cell datasets to multi-million cell atlases has uncovered major scalability problems for single-cell analysis software. Here, we present BPCells, a package for high-performance single-cell analysis of RNA-seq and ATAC-seq datasets. BPCells uses disk-backed streaming compute algorithms to reduce memory requirements by nearly 70-fold compared to in-memory workflows with little to no loss of execution speed. BPCells also introduces high-performance compressed formats based on bitpacking compression for ATAC-seq fragment files and single-cell sparse matrices. These novel compression algorithms help to accelerate disk-backed analysis by reducing data transfer from disk, while providing the lowest computational overhead of all compression algorithms tested. Using BPCells, we perform normalization and PCA of a 44 million cell dataset on a laptop, demonstrating that BPCells makes working with the largest contemporary single-cell datasets feasible on modest hardware, while leaving headroom on servers for future datasets an order of magnitude larger.

PPDPF is not a key regulator of human pancreas development

Given their capability to differentiate into each cell type of the human body, human pluripotent stem cells (hPSCs) provide a unique platform for developmental studies. In the current study, we employed this cell system to understand the role of pancreatic progenitor differentiation and proliferation factor (PPDPF), a protein that has been little explored so far. While the zebrafish orthologue exdpf is essential for exocrine pancreas specification, its importance for mammalian and human development has not been studied yet. We implemented a four times CRISPR/Cas9 nicking approach to knockout PPDPF in human embryonic stem cells (hESCs) and differentiated PPDPFKO/KO and PPDPFWT/WT cells towards the pancreatic lineage. In contrast to data obtained from zebrafish, a very modest effect of the knockout was observed in the development of pancreatic progenitors in vitro, not affecting lineage specification upon orthotopic transplantation in vivo. The modest effect is in line with the finding that genetic variants near PPDPF are associated with random glucose levels in humans, but not with type 2 diabetes risk, supporting that dysregulation of this gene may only result in minor alterations of glycaemic balance in humans. In addition, PPDPF is less organ- and cell type specifically expressed in higher vertebrates and its so far reported functions appear highly context-dependent.

Genomic analysis of 11,555 probands identifies 60 dominant congenital heart disease genes

Congenital heart disease (CHD) is a leading cause of infant mortality. We analyzed de novo mutations (DNMs) and very rare transmitted/unphased damaging variants in 248 prespecified genes in 11,555 CHD probands. The results identified 60 genes with a significant burden of heterozygous damaging variants. Variants in these genes accounted for CHD in 10.1% of probands with similar contributions from de novo and transmitted variants in parent-offspring trios that showed incomplete penetrance. DNMs in these genes accounted for 58% of the signal from DNMs. Thirty-three genes were linked to a single CHD subtype while 12 genes were associated with 2 to 4 subtypes. Seven genes were only associated with isolated CHD, while 37 were associated with 1 or more extracardiac abnormalities. Genes selectively expressed in the cardiomyocyte lineage were associated with isolated CHD, while those widely expressed in the brain were also associated with neurodevelopmental delay (NDD). Missense variants introducing or removing cysteines in epidermal growth factor (EGF)-like domains of NOTCH1 were enriched in tetralogy of Fallot and conotruncal defects, unlike the broader CHD spectrum seen with loss of function variants. Transmitted damaging missense variants in MYH6 were enriched in multiple CHD phenotypes and account for ~1% of all probands. Probands with characteristic mutations causing syndromic CHD were frequently not diagnosed clinically, often due to missing cardinal phenotypes. CHD genes that were positively or negatively associated with development of NDD suggest clinical value of genetic testing. These findings expand the understanding of CHD genetics and support the use of molecular diagnostics in CHD.

Recursive Clustering of Cellular Diversity in scRNA-Seq Data

In scRNA-seq analysis, cell clusters are typically defined by a single round of feature extraction and clustering. This approach may miss phenotypic differences in cell types that are characterized by genes not sufficiently represented in the feature set derived using all cells, such as rare cell types. This work explores an alternative approach, where cell clusters are identified by recursively performing feature extraction and clustering on previously identified clusters, such that each subclustering step uses features that are more specific to distinguishing the higher-resolution subclusters. We benchmark this recursive approach against the conventional, nonrecursive clustering approach and demonstrate that the recursive method results in robust improvement in cell type detection on four scRNA-seq datasets across a wide range of clustering resolution parameters. We apply the recursive approach to cluster scRNA-seq data obtained from patients with Crohn's disease belonging to three clinical phenotypes and observe that recursive clustering captures phenotypic differences only visible at specific levels of granularity within an interpretable hierarchical framework while defining cell clusters within a gene expression feature space more specific to each cluster.

How to view the female reproductive tract through single-cell looking glasses

Single-cell technologies have emerged as an unprecedented tool for biologists and clinicians, allowing them to assess organs and tissues at the level of individual cells. In the field of women's reproductive biology, single-cell studies have provided insights into the cellular and molecular processes that regulate reproductive and obstetrical functions in health and disease. The knowledge that these studies generate is helping clinicians to improve the understanding and diagnosis of infertility related issues or pregnancy complications and to find new avenues for their treatment. However, navigating the expansive landscape of this type of transcriptomic data analysis represents a pivotal challenge in current research. Single cell RNA sequencing involves isolating cells into droplets, reverse transcribing RNA to generate complementary DNA, with each droplet content uniquely labeled by a barcode. Upon sequencing the complementary DNAs, the barcodes enable the reassignment of sequencing reads to individual droplets, facilitating the reconstruction of the cellular landscape of the sample obtained from a tissue or organ and beyond. Researchers, equipped with the metaphorical 'single-cell glasses,' must adequately choose from a plethora of strategies to dissect and interpret cellular information. Sophisticated algorithms and the decision-making process are often underestimated, resulting in artefactual or cumbersome interpreted results. Computational biologists apply and innovate computational tools designed to process, model, and interpret expansive datasets. The ramifications of their work extend far beyond the realm of data processing; they give shape to the outcome of analyses, playing a pivotal role in drawing meaningful conclusions from the wealth of information garnered. In this review, we describe the wide variety of approaches and analytical steps available with enough detail to gain a concise picture of what a complete examination of a single-cell dataset would be. We commence with a discussion on key points in experimental design, highlighting crucial questions one should consider. Following this, we delve into the various preprocessing and quality control steps essential for any single-cell dataset. The subsequent section offers a detailed guide on constructing a single-cell atlas, exploring nuances such as differential characteristics in visualization and clustering techniques, as well as strategies for assigning identity to cell populations through gene marker annotations. Moving beyond the creation of an atlas, we explore methods for investigating pathological conditions. This involves conducting cell population comparison tests between conditions and analyzing specific cell-to-cell communications and cellular differentiation trajectories in both health and disease scenarios. This work aims to furnish a newcomer researcher and/or clinician with essential guidelines to embark on a single-cell adventure without succumbing to common pitfalls. By bridging the gap between theory and practice, it facilitates the translation of single-cell technologies into clinically relevant applications. Throughout the manuscript, practical examples of its usage in women's reproductive health studies are provided. Various sections delve into specific clinical scenarios, demonstrating how these guidelines can be instrumental in unraveling the molecular landscapes of diseases and physiological processes related to women's reproduction.

The sodium leak channel NALCN is regulated by neuronal SNARE complex proteins

NALCN (sodium leak channel, nonselective) is vital for regulating electrical activity in neurons and other excitable cells, and mutations in the channel or its auxiliary proteins lead to severe neurodevelopmental disorders. Here, we show that the neuronal SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptors) complex proteins syntaxin and SNAP25 (synaptosome-associated protein 25), which enable synaptic transmission in the nervous system, inhibit the activity of the NALCN channel complex in both heterologous systems and primary neurons. The existence of this interaction suggests that the neurotransmitter release machinery can regulate electrical signaling directly and therefore modulate the threshold for its own activity. We further find that reduction of NALCN currents is sufficient to promote cell survival in syntaxin-depleted cells. This suggests that disinhibited NALCN may cause the puzzling phenomenon of rapid neuronal cell death in the absence of syntaxin. This interaction could offer opportunities for future drug development against genetic diseases linked to both NALCN- and SNARE protein-containing complexes.

Comprehensive molecular portrait reveals genetic diversity and distinct molecular subtypes of small intestinal neuroendocrine tumors

Small intestinal neuroendocrine tumors (siNETs) are rare bowel tumors arising from malignant enteroendocrine cells, which normally regulate digestion throughout the intestine. Though infrequent, their incidence is rising through better diagnosis, fostering research into their origin and treatment. To date, siNETs are considered to be a single entity and are clinically treated as such. Here, by performing a multi-omics analysis of siNETs, we unveil four distinct molecular groups with strong clinical relevance and provide a resource to study their origin and clinical features. Transcriptomic, genetic and DNA methylation profiles identify two groups linked to distinct enteroendocrine differentiation patterns, another with a strong immune phenotype, and the last with mesenchymal properties. This latter subtype displays the worst prognosis and resistance to treatments in line with infiltration of cancer-associated fibroblasts. These data provide insights into the origin and diversity of these rare diseases, in the hope of improving clinical research into their management.

Bioenergetic-related gene expression in the hippocampus predicts internalizing vs. externalizing behavior in an animal model of temperament

Externalizing and internalizing behavioral tendencies underlie many psychiatric and substance use disorders. These tendencies are associated with differences in temperament that emerge early in development via the interplay of genetic and environmental factors. To better understand the neurobiology of temperament, we have selectively bred rats for generations to produce two lines with highly divergent behavior: bred Low Responders (bLRs) are highly inhibited and anxious in novel environments, whereas bred High Responders (bHRs) are highly exploratory, sensation-seeking, and prone to drug-seeking behavior. Recently, we delineated these heritable differences by intercrossing bHRs and bLRs (F0-F1-F2) to produce a heterogeneous F2 sample with well-characterized lineage and behavior (exploratory locomotion, anxiety-like behavior, Pavlovian conditioning). The identified genetic loci encompassed variants that could influence behavior via many mechanisms, including proximal effects on gene expression. Here we measured gene expression in male and female F0s (n = 12 bHRs, 12 bLRs) and in a large sample of heterogeneous F2s (n = 250) using hippocampal RNA-Seq. This enabled triangulation of behavior with both genetic and functional genomic data to implicate specific genes and biological pathways. Our results show that bHR/bLR differential gene expression is robust, surpassing sex differences in expression, and predicts expression associated with F2 behavior. In F0 and F2 samples, gene sets related to growth/proliferation are upregulated with bHR-like behavior, whereas gene sets related to mitochondrial function, oxidative stress, and microglial activation are upregulated with bLR-like behavior. Integrating our F2 RNA-Seq data with previously-collected whole genome sequencing data identified genes with hippocampal expression correlated with proximal genetic variation (cis-expression quantitative trait loci or cis-eQTLs). These cis-eQTLs successfully predict bHR/bLR differential gene expression based on F0 genotype. Sixteen of these genes are associated with cis-eQTLs colocalized within loci we previously linked to behavior and are strong candidates for mediating the influence of genetic variation on behavioral temperament. Eight of these genes are related to bioenergetics. Convergence between our study and others targeting similar behavioral traits revealed five more genes consistently related to temperament. Overall, our results implicate hippocampal bioenergetic regulation of oxidative stress, microglial activation, and growth-related processes in shaping behavioral temperament, thereby modulating vulnerability to psychiatric and addictive disorders.

Erythroid progenitor cell-mediated spleen-tumor interaction deteriorates cancer immunity

Understanding both local and systemic immunity is essential to optimizing the effectiveness of immunotherapy. However, the dynamic alterations in systemic immunity during tumor development are yet to be clearly defined. Here, we identified a previously unrecognized connection that bridges the interaction between the spleen and tumor through erythroid progenitor cells (EPCs), which suppress tumor immunity and promote tumor progression. We performed the single-cell RNA-seq and RNA-seq to demonstrate the presence of EPCs and identify the characteristic and an immunomodulatory role of EPCs during tumor progression. These tumor-hijacked EPCs proliferate in situ in spleens and impaired systemic and local antitumor response through the interaction between tumor and spleen. Specifically, the splenic CD45- EPCs secreted heparin-binding growth factor to regulate PD-L1-mediated immunosuppression of splenic CD45+ EPCs. Educated CD45+ EPCs from the spleen then migrated to the tumors via the CCL5/CCR5 axis, thereby weakening local antitumor immunity. Consequently, targeting EPCs not only revitalized antitumor immunity but also improved the anti-PD-L1 effect by promoting intratumoral T cell infiltration. Importantly, CD45+ EPCs are associated with immunosuppression and reduced survival in patients with head and neck squamous cell carcinoma. Collectively, these findings reveal the role of EPCs in orchestrating the interaction between the spleen and tumor, which could have significant implications for the development of more effective cancer immunotherapy.

Smoothie: Efficient Inference of Spatial Co-expression Networks from Denoised Spatial Transcriptomics Data

Finding correlations in spatial gene expression is fundamental in spatial transcriptomics, as co-expressed genes within a tissue are linked by regulation, function, pathway, or cell type. Yet, sparsity and noise in spatial transcriptomics data pose significant analytical challenges. Here, we introduce Smoothie, a method that denoises spatial transcriptomics data with Gaussian smoothing and constructs and integrates genome-wide co-expression networks. Utilizing implicit and explicit parallelization, Smoothie scales to datasets exceeding 100 million spatially resolved spots with fast run times and low memory usage. We demonstrate how co-expression networks measured by Smoothie enable precise gene module detection, functional annotation of uncharacterized genes, linkage of gene expression to genome architecture, and multi-sample comparisons to assess stable or dynamic gene expression patterns across tissues, conditions, and time points. Overall, Smoothie provides a scalable and versatile framework for extracting deep biological insights from high-resolution spatial transcriptomics data.

Decoding the blueprints of embryo development with single-cell and spatial omics

Embryonic development is a complex and intricately regulated process that encompasses precise control over cell differentiation, morphogenesis, and the underlying gene expression changes. Recent years have witnessed a remarkable acceleration in the development of single-cell and spatial omic technologies, enabling high-throughput profiling of transcriptomic and other multi-omic information at the individual cell level. These innovations offer fresh and multifaceted perspectives for investigating the intricate cellular and molecular mechanisms that govern embryonic development. In this review, we provide an in-depth exploration of the latest technical advancements in single-cell and spatial multi-omic methodologies and compile a systematic catalog of their applications in the field of embryonic development. We deconstruct the research strategies employed by recent studies that leverage single-cell sequencing techniques and underscore the unique advantages of spatial transcriptomics. Furthermore, we delve into both the current applications, data analysis algorithms and the untapped potential of these technologies in advancing our understanding of embryonic development. With the continuous evolution of multi-omic technologies, we anticipate their widespread adoption and profound contributions to unraveling the intricate molecular foundations underpinning embryo development in the foreseeable future.

Multi-modal refinement of the human heart atlas during the first gestational trimester

Forty first-trimester human hearts were studied to lay groundwork for further studies of the mechanisms underlying congenital heart defects. We first sampled 49,227 cardiac nuclei from three fetuses at 8.6, 9.0, and 10.7 post-conceptional weeks (pcw) for single-nucleus RNA sequencing, enabling the distinction of six classes comprising 21 cell types. Improved resolution led to the identification of previously unappreciated cardiomyocyte populations and minority autonomic and lymphatic endothelial transcriptomes, among others. After integration with 5-7 pcw heart single-cell RNA-sequencing data, we identified a human cardiomyofibroblast progenitor preceding the diversification of cardiomyocyte and stromal lineages. Spatial transcriptomic analysis (six Visium sections from two additional hearts) was aided by deconvolution, and key spatial markers validated on sectioned and whole hearts in two- and three-dimensional space and over time. Altogether, anatomical-positional features, including innervation, conduction and subdomains of the atrioventricular septum, translate latent molecular identity into specialized cardiac functions. This atlas adds unprecedented spatial and temporal resolution to the characterization of human-specific aspects of early heart formation.

Batch correcting single-cell spatial transcriptomics count data with Crescendo improves visualization and detection of spatial gene patterns

Spatial transcriptomics facilitates gene expression analysis of cells in their spatial anatomical context. Batch effects hinder visualization of gene spatial patterns across samples. We present the Crescendo algorithm to correct for batch effects at the gene expression level and enable accurate visualization of gene expression patterns across multiple samples. We show Crescendo's utility and scalability across three datasets ranging from 170,000 to 7 million single cells across spatial and single-cell RNA sequencing technologies. By correcting for batch effects, Crescendo enhances spatial transcriptomics analyses to detect gene colocalization and ligand-receptor interactions and enables cross-technology information transfer.

The tumour microenvironment of pilocytic astrocytoma evolves over time via enrichment for microglia

Pilocytic astrocytoma (PA) is the commonest low-grade tumour affecting children and is frequently experienced as a chronic disease associated with extended treatment, periods of regrowth, and long-term disability. This contrasts with the view of PA as a benign tumour with positive clinical outcomes and raises the fundamental question of biologically driven change over time within these tumours, which will impact diagnosis, stratification, and management. To investigate the molecular, cellular, and pathological stability of PA we performed RNA sequencing, methylation array profiling, immunohistochemistry, and targeted panel DNA sequencing on a cohort of 15 PA patients with matched primary/longitudinal samples at a mean sampling interval of 2.7 years. Through pairwise analysis of primary versus longitudinal tumour samples we identified changes to immune-related pathways within the expression and methylation profiles of longitudinal PA. Further interrogation of these changes revealed an enrichment over time for microglial cell populations, which was validated by immunohistochemistry against common monocyte/microglial markers. Moreover, immunohistochemical characterisation revealed concurrent increases in the expression of M2-like and anti-inflammatory markers. Microglial enrichments were consistent across the cohort and were not adequately explained by a range of potential confounders, including receipt of adjuvant therapy. Taken together, these data challenge the idea of pilocytic astrocytoma as a static entity and indicate that they consistently accumulate microglia over time, potentially co-opting the immune microenvironment towards an anti-inflammatory phenotype that may affect the natural course and treatment response of the tumours.

Gene expression patterns of the developing human face at single cell resolution reveal cell type contributions to normal facial variation and disease risk

Craniofacial development gives rise to the complex structures of the face and involves the interplay of diverse cell types. Despite its importance, our understanding of human-specific craniofacial developmental mechanisms and their genetic underpinnings remains limited. Here, we present a comprehensive single-nucleus RNA sequencing (snRNA-seq) atlas of human craniofacial development from craniofacial tissues of 24 embryos that span six key time points during the embryonic period (4-8 post-conception weeks). This resource resolves the transcriptional dynamics of seven major cell types and uncovers distinct major cell types, including muscle progenitors and cranial neural crest cells (CNCCs), as well as dozens of subtypes of ectoderm and mesenchyme. Comparative analyses reveal substantial conservation of major cell types, alongside human biased differences in gene expression programs. CNCCs, which play a crucial role in craniofacial morphogenesis, exhibit the lowest marker gene conservation, underscoring their evolutionary plasticity. Spatial transcriptomics further localizes cell populations, providing a detailed view of their developmental roles and anatomical context. We also link these developmental processes to genetic variation, identifying cell type-specific enrichments for common variants associated with facial morphology and rare variants linked to orofacial clefts. Intriguingly, Neanderthal-introgressed sequences are enriched near genes with biased expression in cartilage and specialized ectodermal subtypes, suggesting their contribution to modern human craniofacial features. This atlas offers unprecedented insights into the cellular and genetic mechanisms shaping the human face, highlighting conserved and distinctly human aspects of craniofacial biology. Our findings illuminate the developmental origins of craniofacial disorders, the genetic basis of facial variation, and the evolutionary legacy of ancient hominins. This work provides a foundational resource for exploring craniofacial biology, with implications for developmental genetics, evolutionary biology, and clinical research into congenital anomalies.

SCEMENT: scalable and memory efficient integration of large-scale single-cell RNA-sequencing data

MOTIVATION

Integrative analysis of large-scale single-cell data collected from diverse cell populations promises an improved understanding of complex biological systems. While several algorithms have been developed for single-cell RNA-sequencing data integration, many lack the scalability to handle large numbers of datasets and/or millions of cells due to their memory and run time requirements. The few tools that can handle large data do so by reducing the computational burden through strategies such as subsampling of the data or selecting a reference dataset to improve computational efficiency and scalability. Such shortcuts, however, hamper the accuracy of downstream analyses, especially those requiring quantitative gene expression information.

RESULTS

We present SCEMENT, a SCalablE and Memory-Efficient iNTegration method, to overcome these limitations. Our new parallel algorithm builds upon and extends the linear regression model previously applied in ComBat to an unsupervised sparse matrix setting to enable accurate integration of diverse and large collections of single-cell RNA-sequencing data. Using tens to hundreds of real single-cell RNA-seq datasets, we show that SCEMENT outperforms ComBat as well as FastIntegration and Scanorama in runtime (upto 214× faster) and memory usage (upto 17.5× less). It not only performs batch correction and integration of millions of cells in under 25 min, but also facilitates the discovery of new rare cell types and more robust reconstruction of gene regulatory networks with full quantitative gene expression information.

AVAILABILITY AND IMPLEMENTATION

Source code freely available for download at https://github.com/AluruLab/scement, implemented in C++ and supported on Linux.

Understanding developing kidneys and Wilms tumors one cell at a time

Single-cell sequencing-based techniques are revolutionizing all fields of biomedical sciences, including normal kidney development and how this is disturbed in the development of Wilms tumor. The many different techniques and the differences between them can obscure which technique is best used to answer which question. In this review we summarize the techniques currently available, discuss which have been used in kidney development or Wilms tumor context, and which techniques can or should be combined to maximize the increase in biological understanding we can get from them.

Massive changes in gene expression and their cause(s) can be a unifying principle in the pathobiology of Alzheimer's disease

Understanding of the biology of Alzheimer's disease (AD) has long been fragmented, with various investigators concentrating on amyloid beta (Aβ) or tau, inflammation, cell death pathways, misfolded proteins, glia, and more. Yet data from multiple authors has repeatedly shown altered expression of myriad genes related to these seemingly disparate phenomena. In 2022, Morgan et al. organized the massive data on changes in AD in a meticulous survey of the literature and related these changes to Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Their data showed that 91% of the known KEGG pathways are involved in AD and that many of these pathways are represented by the known cellular/molecular phenomena of AD. Such data then raise the fundamental question: What mechanism(s) may be responsible for such widespread changes in gene expression? We review evidence for a unifying model based on sequestrations in stress granules and alteration of nucleocytoplasmic transport in AD. HIGHLIGHTS: In Alzheimer's disease (AD), critical changes take place in neurons before the appearance of plaques or tangles. Addressing these early changes provides a path to early detection and effective intervention in AD.

Comparing Alzheimer's genes in African, European, and Amerindian induced pluripotent stem cell-derived microglia

INTRODUCTION

Genome-wide association studies (GWAS) studies in Alzheimer's disease (AD) demonstrate ancestry-specific loci. Previous studies in the regulatory architecture have only been conducted in Europeans (EUs), thus studies in additional ancestries are needed. Given the prevalence of AD genes expressed in microglia, we initiated our studies in induced pluripotent stem cell (iPSC) -derived microglia.

METHODS

We created iPSC-derived microglia from 13 individuals of either high Amerindian (AI), African (AF), or EU global ancestry, including both AD and controls. RNA-seq, ATAC-seq, and pathway analyses were compared between ancestries in both AD and non-AD genes.

RESULTS

Twelve AD genes were differentially expressed genes (DEGs) and/or accessible between ancestries, including ABI3, CTSB, and MS4A6A. A total of 5% of all genes had differential ancestral expression, but differences in accessibility were less than 1%. The DEGs were enriched in known AD pathways.

DISCUSSION

This resource will be valuable in evaluating AD in admixed populations and other neurological disorders and understanding the AD risk differences between populations.

HIGHLIGHTS

First comparison of the genomics of AI, AF, and EU microglia. Report differences in expression and accessibility of AD genes between ancestries. Ancestral expression differences are greater than differences in accessibility. Good transcriptome correlation was seen between brain and iPSC-derived microglia. Differentially expressed AD genes were in known AD pathways.

The dynamics of hematopoiesis over the human lifespan

Over a lifetime, hematopoietic stem cells (HSCs) adjust their lineage output to support age-aligned physiology. In model organisms, stereotypic waves of hematopoiesis have been observed corresponding to defined age-biased HSC hallmarks. However, how the properties of hematopoietic stem and progenitor cells change over the human lifespan remains unclear. To address this gap, we profiled individual transcriptome states of human hematopoietic stem and progenitor cells spanning gestation, maturation and aging. Here we define the gene expression networks dictating age-specific differentiation of HSCs and the dynamics of fate decisions and lineage priming throughout life. We additionally identifiy and functionally validate a fetal-specific HSC state with robust engraftment and multilineage capacity. Furthermore, we observe that classification of acute myeloid leukemia against defined transcriptional age states demonstrates that utilization of early life transcriptional programs associates with poor prognosis. Overall, we provide a disease-relevant framework for heterochronic orientation of stem cell ontogeny along the real time axis of the human lifespan.

MetaQ: fast, scalable and accurate metacell inference via single-cell quantization

To overcome the computational barriers of analyzing large-scale single-cell sequencing data, we introduce MetaQ, a metacell algorithm that scales to arbitrarily large datasets with linear runtime and constant memory usage. Inspired by cellular development, MetaQ conceptualizes each metacell as a collective ancestor of biologically similar cells. By quantizing cells into a discrete codebook, where each entry represents a metacell capable of reconstructing the original cells it quantizes, MetaQ identifies homogeneous cell subsets for efficient and accurate metacell inference. This approach reduces computational complexity from exponential to linear while maintaining or surpassing the performance of existing metacell algorithms. Extensive experiments demonstrate that MetaQ excels in downstream tasks such as cell type annotation, developmental trajectory inference, batch integration, and differential expression analysis. Thanks to its superior efficiency and effectiveness, MetaQ makes analyzing datasets with millions of cells practical, offering a powerful solution for single-cell studies in the era of high-throughput profiling.

ERH Enables Early Embryonic Differentiation and Overlays H3K9me3 Heterochromatin on a Cryptic Pluripotency H3K9me3 Landscape in Somatic Cells

Enhancer of Rudimentary Homolog (ERH) is an evolutionarily conserved protein originally characterized in fission yeast 1 and recently shown to maintain H3K9me3 in human fibroblasts 2 . Here, we find that ERH depletion in fibroblasts reverts the H3K9me3 landscape to an embryonic stem cell (ESC) state and enables activation of naïve and pluripotency genes and transposable elements during induced pluripotent stem cell (iPSC) reprogramming. We find that ERH similarly represses totipotent and alternative lineage programs during mouse preimplantation development and is required for proper segregation of the inner cell mass and trophectoderm cell lineages. During human ESC differentiation into germ layer lineages, ERH silences naïve and pluripotency genes, transposable elements, and alternative lineage somatic genes. As in fission yeast, we find that mammalian ERH interacts with RNA-binding proteins to engage and repress its chromatin targets. Our findings reveal a fundamental role for ERH in cell fate specification via the initiation and maintenance of early developmental gene repression.

A spatially resolved multi-omic single-cell atlas of soybean development

Cis-regulatory elements (CREs) precisely control spatiotemporal gene expression in cells. Using a spatially resolved single-cell atlas of gene expression with chromatin accessibility across ten soybean tissues, we identified 103 distinct cell types and 303,199 accessible chromatin regions (ACRs). Nearly 40% of the ACRs showed cell-type-specific patterns and were enriched for transcription factor (TF) motifs defining diverse cell identities. We identified de novo enriched TF motifs and explored the conservation of gene regulatory networks underpinning legume symbiotic nitrogen fixation. With comprehensive developmental trajectories for endosperm and embryo, we uncovered the functional transition of the three sub-cell types of endosperm, identified 13 sucrose transporters sharing the DNA binding with one finger 11 (DOF11) motif that were co-upregulated in late peripheral endosperm, and identified key embryo cell-type specification regulators during embryogenesis, including a homeobox TF that promotes cotyledon parenchyma identity. This resource provides a valuable foundation for analyzing gene regulatory programs in soybean cell types across tissues and life stages.

Effects of the maternal and fetal proteome on birth weight: a Mendelian randomization analysis

Background

Fetal growth is an important indicator of survival, regulated by maternal and fetal genetic and environmental factors. However, little is known about the underlying molecular mechanisms. Proteins play a major role in a wide range of biological processes and could provide key insights into maternal and fetal molecular mechanisms regulating fetal growth.

Method

We used intergenerational two-sample Mendelian randomization to explore the effects of 1,139 maternal and fetal genetically-instrumented plasma proteins on birth weight. We used genome-wide association summary data from the Early Growth Genetics (EGG) consortium (n=406,063 with maternal and/or fetal genotype), with independent replication in the Norwegian Mother, Father and Child Cohort Study (MoBa; n=74,932 mothers and n=62,108 offspring). Maternal and fetal data were adjusted for the correlation between fetal and maternal genotype, to distinguish their independent genetic effects.

Results

We found that higher genetically-predicted maternal levels of NEC1 increased birth weight (mean-difference: 12g (95% CI [6g, 18g]) per 1 standard deviation protein level) as did PRS57 (20g [10g, 31g]) and ULK3 (140g [81g, 199g]). Higher maternal levels of Galectin_4 decreased birth weight (-206g [-299g, -113g]). In contrast, in the offspring, higher genetically-predicted offspring levels of NEC1 decreased birth weight (-10g [-16g, -5g]), alongside sLeptin_R (-8g [-12g, -4g]), and UBS3B (-78g [-116g, -41g]). Higher fetal levels of Galectin_4 increased birth weight (174g [89g, 258g]). We replicated these results in MoBa, and found supportive evidence for shared causal variants from genetic colocalization analyses and protein-protein network associations.

Conclusions

We find strong evidence for causal effects, sometimes in opposing directions, of maternal and fetal genetically-instrumented proteins on birth weight. These provide new insights into maternal and fetal molecular mechanisms regulating fetal growth, involving glucose metabolism, energy balance, and vascular function that could be used to identify new intervention targets to reduce the risk of fetal growth disorders, and their associated adverse maternal and fetal outcomes.

Modular organization of enhancer network provides transcriptional robustness in mammalian development

Enhancer clusters, pivotal in mammalian development and diseases, can organize as enhancer networks to control cell identity and disease genes; however, the underlying mechanism remains largely unexplored. Here, we introduce eNet 2.0, a comprehensive tool for enhancer networks analysis during development and diseases based on single-cell chromatin accessibility data. eNet 2.0 extends our previous work eNet 1.0 by adding network topology, comparison and dynamics analyses to its network construction function. We reveal modularly organized enhancer networks, where inter-module interactions synergistically affect gene expression. Moreover, network alterations correlate with abnormal and dynamic gene expression in disease and development. eNet 2.0 is robust across diverse datasets. To facilitate application, we introduce eNetDB (https://enetdb.huanglabxmu.com), an enhancer network database leveraging extensive scATAC-seq (single-cell assay for transposase-accessible chromatin sequencing) datasets from human and mouse tissues. Together, our work provides a powerful computational tool and reveals that modularly organized enhancer networks contribute to gene expression robustness in mammalian development and diseases.

Potential Adaptive Introgression From Dogs in Iberian Grey Wolves (Canis lupus)

Invading species along with increased anthropogenization may lead to hybridization events between wild species and closely related domesticates. As a consequence, wild species may carry introgressed alleles from domestic species, which is generally assumed to yield adverse effects in wild populations. The opposite evolutionary consequence, adaptive introgression, where introgressed genes are positively selected in the wild species, is possible but has rarely been documented. Grey wolves (Canis lupus) are widely distributed across the Holarctic and frequently coexist with their close relative, the domestic dog (C. familiaris). Despite ample opportunity, hybridization rarely occurs in most populations. Here we studied the geographically isolated grey wolves of the Iberian Peninsula, who have coexisted with a large population of loosely controlled dogs for thousands of years in a human-modified landscape. We assessed the extent and impact of dog introgression on the current Iberian grey wolf population by analysing 150 whole genomes of Iberian and other Eurasian grey wolves as well as dogs originating from across Europe and western Siberia. We identified almost no recent introgression and a small (< 5%) overall ancient dog ancestry. Using a combination of single scan statistics and ancestry enrichment estimates, we identified positive selection on six genes (DAPP1, NSMCE4A, MPPED2, PCDH9, MBTPS1, and CDH13) for which wild Iberian wolves carry alleles introgressed from dogs. The genes with introgressed and positively selected alleles include functions in immune response and brain functions, which may explain some of the unique behavioural phenotypes in Iberian wolves such as their reduced dispersal compared to other wolf populations.

CSI-GEP: A GPU-based unsupervised machine learning approach for recovering gene expression programs in atlas-scale single-cell RNA-seq data

Exploratory analysis of single-cell RNA sequencing (scRNA-seq) typically relies on hard clustering over two-dimensional projections like uniform manifold approximation and projection (UMAP). However, such methods can severely distort the data and have many arbitrary parameter choices. Methods that can model scRNA-seq data as non-discrete "gene expression programs" (GEPs) can better preserve the data's structure, but currently, they are often not scalable, not consistent across repeated runs, and lack an established method for choosing key parameters. Here, we developed a GPU-based unsupervised learning approach, "consensus and scalable inference of gene expression programs" (CSI-GEP). We show that CSI-GEP can recover ground truth GEPs in real and simulated atlas-scale scRNA-seq datasets, significantly outperforming cutting-edge methods, including GPT-based neural networks. We applied CSI-GEP to a whole mouse brain atlas of 2.2 million cells, disentangling endothelial cell types missed by other methods, and to an integrated scRNA-seq atlas of human tumors and cell lines, discovering mesenchymal-like GEPs unique to cancer cells growing in culture.

MAPbrain: a multi-omics atlas of the primate brain

The brain is the central hub of the entire nervous system. Its development is a lifelong process guided by a genetic blueprint. Understanding how genes influence brain development is critical for deciphering the formation of human cognitive functions and the underlying mechanisms of neurological disorders. Recent advances in multi-omics techniques have now made it possible to explore these aspects comprehensively. However, integrating and analyzing extensive multi-omics data presents significant challenges. Here, we introduced MAPbrain (http://bigdata.ibp.ac.cn/mapBRAIN/), a multi-omics atlas of the primate brain. This repository integrates and normalizes both our own lab's published data and publicly available multi-omics data, encompassing 21 million brain cells from 38 key brain regions and 436 sub-regions across embryonic and adult stages, with 164 time points in humans and non-human primates. MAPbrain offers a unique, robust, and interactive platform that includes transcriptomics, epigenomics, and spatial transcriptomics data, facilitating a comprehensive exploration of brain development. The platform enables the exploration of cell type- and time point-specific markers, gene expression comparison between brain regions and species, joint analyses across transcriptome and epigenome, and navigation of cell types across species, brain regions, and development stages. Additionally, MAPbrain provides an online integration module for users to navigate and analyze their own data within the platform.