Metacell-2: a divide-and-conquer metacell algorithm for scalable scRNA-seq analysis

A comprehensive survey of dimensionality reduction and clustering methods for single-cell and spatial transcriptomics data

In recent years, the application of single-cell transcriptomics and spatial transcriptomics analysis techniques has become increasingly widespread. Whether dealing with single-cell transcriptomic or spatial transcriptomic data, dimensionality reduction and clustering are indispensable. Both single-cell and spatial transcriptomic data are often high-dimensional, making the analysis and visualization of such data challenging. Through dimensionality reduction, it becomes possible to visualize the data in a lower-dimensional space, allowing for the observation of relationships and differences between cell subpopulations. Clustering enables the grouping of similar cells into the same cluster, aiding in the identification of distinct cell subpopulations and revealing cellular diversity, providing guidance for downstream analyses. In this review, we systematically summarized the most widely recognized algorithms employed for the dimensionality reduction and clustering analysis of single-cell transcriptomic and spatial transcriptomic data. This endeavor provides valuable insights and ideas that can contribute to the development of novel tools in this rapidly evolving field.

Building and analyzing metacells in single-cell genomics data

The advent of high-throughput single-cell genomics technologies has fundamentally transformed biological sciences. Currently, millions of cells from complex biological tissues can be phenotypically profiled across multiple modalities. The scaling of computational methods to analyze and visualize such data is a constant challenge, and tools need to be regularly updated, if not redesigned, to cope with ever-growing numbers of cells. Over the last few years, metacells have been introduced to reduce the size and complexity of single-cell genomics data while preserving biologically relevant information and improving interpretability. Here, we review recent studies that capitalize on the concept of metacells-and the many variants in nomenclature that have been used. We further outline how and when metacells should (or should not) be used to analyze single-cell genomics data and what should be considered when analyzing such data at the metacell level. To facilitate the exploration of metacells, we provide a comprehensive tutorial on the construction and analysis of metacells from single-cell RNA-seq data ( https://github.com/GfellerLab/MetacellAnalysisTutorial ) as well as a fully integrated pipeline to rapidly build, visualize and evaluate metacells with different methods ( https://github.com/GfellerLab/MetacellAnalysisToolkit ).

Single-cell genomics and regulatory networks for 388 human brains

Single-cell genomics is a powerful tool for studying heterogeneous tissues such as the brain. Yet little is understood about how genetic variants influence cell-level gene expression. Addressing this, we uniformly processed single-nuclei, multiomics datasets into a resource comprising >2.8 million nuclei from the prefrontal cortex across 388 individuals. For 28 cell types, we assessed population-level variation in expression and chromatin across gene families and drug targets. We identified >550,000 cell type-specific regulatory elements and >1.4 million single-cell expression quantitative trait loci, which we used to build cell-type regulatory and cell-to-cell communication networks. These networks manifest cellular changes in aging and neuropsychiatric disorders. We further constructed an integrative model accurately imputing single-cell expression and simulating perturbations; the model prioritized ~250 disease-risk genes and drug targets with associated cell types.

SCIPAC: quantitative estimation of cell-phenotype associations

Numerous algorithms have been proposed to identify cell types in single-cell RNA sequencing data, yet a fundamental problem remains: determining associations between cells and phenotypes such as cancer. We develop SCIPAC, the first algorithm that quantitatively estimates the association between each cell in single-cell data and a phenotype. SCIPAC also provides a p-value for each association and applies to data with virtually any type of phenotype. We demonstrate SCIPAC's accuracy in simulated data. On four real cancerous or noncancerous datasets, insights from SCIPAC help interpret the data and generate new hypotheses. SCIPAC requires minimum tuning and is computationally very fast.

Modeling T cell temporal response to cancer immunotherapy rationalizes development of combinatorial treatment protocols

Successful immunotherapy relies on triggering complex responses involving T cell dynamics in tumors and the periphery. Characterizing these responses remains challenging using static human single-cell atlases or mouse models. To address this, we developed a framework for in vivo tracking of tumor-specific CD8+ T cells over time and at single-cell resolution. Our tools facilitate the modeling of gene program dynamics in the tumor microenvironment (TME) and the tumor-draining lymph node (tdLN). Using this approach, we characterize two modes of anti-programmed cell death protein 1 (PD-1) activity, decoupling induced differentiation of tumor-specific activated precursor cells from conventional type 1 dendritic cell (cDC1)-dependent proliferation and recruitment to the TME. We demonstrate that combining anti-PD-1 therapy with anti-4-1BB agonist enhances the recruitment and proliferation of activated precursors, resulting in tumor control. These data suggest that effective response to anti-PD-1 therapy is dependent on sufficient influx of activated precursor CD8+ cells to the TME and highlight the importance of understanding system-level dynamics in optimizing immunotherapies.

SuperCellCyto: enabling efficient analysis of large scale cytometry datasets

Advancements in cytometry technologies have enabled quantification of up to 50 proteins across millions of cells at single cell resolution. Analysis of cytometry data routinely involves tasks such as data integration, clustering, and dimensionality reduction. While numerous tools exist, many require extensive run times when processing large cytometry data containing millions of cells. Existing solutions, such as random subsampling, are inadequate as they risk excluding rare cell subsets. To address this, we propose SuperCellCyto, an R package that builds on the SuperCell tool which groups highly similar cells into supercells. SuperCellCyto is available on GitHub ( https://github.com/phipsonlab/SuperCellCyto ) and Zenodo ( https://doi.org/10.5281/zenodo.10521294 ).

Single-cell genomics and regulatory networks for 388 human brains

Single-cell genomics is a powerful tool for studying heterogeneous tissues such as the brain. Yet, little is understood about how genetic variants influence cell-level gene expression. Addressing this, we uniformly processed single-nuclei, multi-omics datasets into a resource comprising >2.8M nuclei from the prefrontal cortex across 388 individuals. For 28 cell types, we assessed population-level variation in expression and chromatin across gene families and drug targets. We identified >550K cell-type-specific regulatory elements and >1.4M single-cell expression-quantitative-trait loci, which we used to build cell-type regulatory and cell-to-cell communication networks. These networks manifest cellular changes in aging and neuropsychiatric disorders. We further constructed an integrative model accurately imputing single-cell expression and simulating perturbations; the model prioritized ~250 disease-risk genes and drug targets with associated cell types.

EpiCarousel: memory- and time-efficient identification of metacells for atlas-level single-cell chromatin accessibility data

SUMMARY

Recent technical advancements in single-cell chromatin accessibility sequencing (scCAS) have brought new insights to the characterization of epigenetic heterogeneity. As single-cell genomics experiments scale up to hundreds of thousands of cells, the demand for computational resources for downstream analysis grows intractably large and exceeds the capabilities of most researchers. Here, we propose EpiCarousel, a tailored Python package based on lazy loading, parallel processing, and community detection for memory- and time-efficient identification of metacells, i.e. the emergence of homogenous cells, in large-scale scCAS data. Through comprehensive experiments on five datasets of various protocols, sample sizes, dimensions, number of cell types, and degrees of cell-type imbalance, EpiCarousel outperformed baseline methods in systematic evaluation of memory usage, computational time, and multiple downstream analyses including cell type identification. Moreover, EpiCarousel executes preprocessing and downstream cell clustering on the atlas-level dataset with 707 043 cells and 1 154 611 peaks within 2 h consuming <75 GB of RAM and provides superior performance for characterizing cell heterogeneity than state-of-the-art methods.

AVAILABILITY AND IMPLEMENTATION

The EpiCarousel software is well-documented and freely available at https://github.com/biox-nku/epicarousel. It can be seamlessly interoperated with extensive scCAS analysis toolkits.

SEGCECO: Subgraph Embedding of Gene expression matrix for prediction of CEll-cell COmmunication

Recent advances in single-cell RNA sequencing technology have eased analyses of signaling networks of cells. Recently, cell-cell interaction has been studied based on various link prediction approaches on graph-structured data. These approaches have assumptions about the likelihood of node interaction, thus showing high performance for only some specific networks. Subgraph-based methods have solved this problem and outperformed other approaches by extracting local subgraphs from a given network. In this work, we present a novel method, called Subgraph Embedding of Gene expression matrix for prediction of CEll-cell COmmunication (SEGCECO), which uses an attributed graph convolutional neural network to predict cell-cell communication from single-cell RNA-seq data. SEGCECO captures the latent and explicit attributes of undirected, attributed graphs constructed from the gene expression profile of individual cells. High-dimensional and sparse single-cell RNA-seq data make converting the data into a graphical format a daunting task. We successfully overcome this limitation by applying SoptSC, a similarity-based optimization method in which the cell-cell communication network is built using a cell-cell similarity matrix which is learned from gene expression data. We performed experiments on six datasets extracted from the human and mouse pancreas tissue. Our comparative analysis shows that SEGCECO outperforms latent feature-based approaches, and the state-of-the-art method for link prediction, WLNM, with 0.99 ROC and 99% prediction accuracy. The datasets can be found at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE84133 and the code is publicly available at Github https://github.com/sheenahora/SEGCECO and Code Ocean https://codeocean.com/capsule/8244724/tree.

Clonal hematopoiesis of indeterminate potential is associated with acute kidney injury

Age is a predominant risk factor for acute kidney injury (AKI), yet the biological mechanisms underlying this risk are largely unknown. Clonal hematopoiesis of indeterminate potential (CHIP) confers increased risk for several chronic diseases associated with aging. Here we sought to test whether CHIP increases the risk of AKI. In three population-based epidemiology cohorts, we found that CHIP was associated with a greater risk of incident AKI, which was more pronounced in patients with AKI requiring dialysis and in individuals with somatic mutations in genes other than DNMT3A, including mutations in TET2 and JAK2. Mendelian randomization analyses supported a causal role for CHIP in promoting AKI. Non-DNMT3A-CHIP was also associated with a nonresolving pattern of injury in patients with AKI. To gain mechanistic insight, we evaluated the role of Tet2-CHIP and Jak2V617F-CHIP in two mouse models of AKI. In both models, CHIP was associated with more severe AKI, greater renal proinflammatory macrophage infiltration and greater post-AKI kidney fibrosis. In summary, this work establishes CHIP as a genetic mechanism conferring impaired kidney function recovery after AKI via an aberrant inflammatory response mediated by renal macrophages.

Bispecific dendritic-T cell engager potentiates anti-tumor immunity

Immune checkpoint inhibition treatment using aPD-1 monoclonal antibodies is a promising cancer immunotherapy approach. However, its effect on tumor immunity is narrow, as most patients do not respond to the treatment or suffer from recurrence. We show that the crosstalk between conventional type I dendritic cells (cDC1) and T cells is essential for an effective aPD-1-mediated anti-tumor response. Accordingly, we developed a bispecific DC-T cell engager (BiCE), a reagent that facilitates physical interactions between PD-1+ T cells and cDC1. BiCE treatment promotes the formation of active dendritic/T cell crosstalk in the tumor and tumor-draining lymph nodes. In vivo, single-cell and physical interacting cell analysis demonstrates the distinct and superior immune reprogramming of the tumors and tumor-draining lymph nodes treated with BiCE as compared to conventional aPD-1 treatment. By bridging immune cells, BiCE potentiates cell circuits and communication pathways needed for effective anti-tumor immunity.

SEACells infers transcriptional and epigenomic cellular states from single-cell genomics data

Metacells are cell groupings derived from single-cell sequencing data that represent highly granular, distinct cell states. Here we present single-cell aggregation of cell states (SEACells), an algorithm for identifying metacells that overcome the sparsity of single-cell data while retaining heterogeneity obscured by traditional cell clustering. SEACells outperforms existing algorithms in identifying comprehensive, compact and well-separated metacells in both RNA and assay for transposase-accessible chromatin (ATAC) modalities across datasets with discrete cell types and continuous trajectories. We demonstrate the use of SEACells to improve gene-peak associations, compute ATAC gene scores and infer the activities of critical regulators during differentiation. Metacell-level analysis scales to large datasets and is particularly well suited for patient cohorts, where per-patient aggregation provides more robust units for data integration. We use our metacells to reveal expression dynamics and gradual reconfiguration of the chromatin landscape during hematopoietic differentiation and to uniquely identify CD4 T cell differentiation and activation states associated with disease onset and severity in a Coronavirus Disease 2019 (COVID-19) patient cohort.

MCProj: metacell projection for interpretable and quantitative use of transcriptional atlases

We describe MCProj-an algorithm for analyzing query scRNA-seq data by projections over reference single-cell atlases. We represent the reference as a manifold of annotated metacell gene expression distributions. We then interpret query metacells as mixtures of atlas distributions while correcting for technology-specific gene biases. This approach distinguishes and tags query cells that are consistent with atlas states from unobserved (novel or artifactual) behaviors. It also identifies expression differences observed in successfully mapped query states. We showcase MCProj functionality by projecting scRNA-seq data on a blood cell atlas, deriving precise, quantitative, and interpretable results across technologies and datasets.

An immune cell atlas reveals the dynamics of human macrophage specification during prenatal development

Macrophages are heterogeneous and play critical roles in development and disease, but their diversity, function, and specification remain inadequately understood during human development. We generated a single-cell RNA sequencing map of the dynamics of human macrophage specification from PCW 4-26 across 19 tissues. We identified a microglia-like population and a proangiogenic population in 15 macrophage subtypes. Microglia-like cells, molecularly and morphologically similar to microglia in the CNS, are present in the fetal epidermis, testicle, and heart. They are the major immune population in the early epidermis, exhibit a polarized distribution along the dorsal-lateral-ventral axis, and interact with neural crest cells, modulating their differentiation along the melanocyte lineage. Through spatial and differentiation trajectory analysis, we also showed that proangiogenic macrophages are perivascular across fetal organs and likely yolk-sac-derived as microglia. Our study provides a comprehensive map of the heterogeneity and developmental dynamics of human macrophages and unravels their diverse functions during development.

Single cell Hi-C identifies plastic chromosome conformations underlying the gastrulation enhancer landscape

Embryonic development involves massive proliferation and differentiation of cell lineages. This must be supported by chromosome replication and epigenetic reprogramming, but how proliferation and cell fate acquisition are balanced in this process is not well understood. Here we use single cell Hi-C to map chromosomal conformations in post-gastrulation mouse embryo cells and study their distributions and correlations with matching embryonic transcriptional atlases. We find that embryonic chromosomes show a remarkably strong cell cycle signature. Despite that, replication timing, chromosome compartment structure, topological associated domains (TADs) and promoter-enhancer contacts are shown to be variable between distinct epigenetic states. About 10% of the nuclei are identified as primitive erythrocytes, showing exceptionally compact and organized compartment structure. The remaining cells are broadly associated with ectoderm and mesoderm identities, showing only mild differentiation of TADs and compartment structures, but more specific localized contacts in hundreds of ectoderm and mesoderm promoter-enhancer pairs. The data suggest that while fully committed embryonic lineages can rapidly acquire specific chromosomal conformations, most embryonic cells are showing plastic signatures driven by complex and intermixed enhancer landscapes.

hdWGCNA identifies co-expression networks in high-dimensional transcriptomics data

Biological systems are immensely complex, organized into a multi-scale hierarchy of functional units based on tightly regulated interactions between distinct molecules, cells, organs, and organisms. While experimental methods enable transcriptome-wide measurements across millions of cells, popular bioinformatic tools do not support systems-level analysis. Here we present hdWGCNA, a comprehensive framework for analyzing co-expression networks in high-dimensional transcriptomics data such as single-cell and spatial RNA sequencing (RNA-seq). hdWGCNA provides functions for network inference, gene module identification, gene enrichment analysis, statistical tests, and data visualization. Beyond conventional single-cell RNA-seq, hdWGCNA is capable of performing isoform-level network analysis using long-read single-cell data. We showcase hdWGCNA using data from autism spectrum disorder and Alzheimer's disease brain samples, identifying disease-relevant co-expression network modules. hdWGCNA is directly compatible with Seurat, a widely used R package for single-cell and spatial transcriptomics analysis, and we demonstrate the scalability of hdWGCNA by analyzing a dataset containing nearly 1 million cells.

Time-aligned hourglass gastrulation models in rabbit and mouse

The hourglass model describes the convergence of species within the same phylum to a similar body plan during development; however, the molecular mechanisms underlying this phenomenon in mammals remain poorly described. Here, we compare rabbit and mouse time-resolved differentiation trajectories to revisit this model at single-cell resolution. We modeled gastrulation dynamics using hundreds of embryos sampled between gestation days 6.0 and 8.5 and compared the species using a framework for time-resolved single-cell differentiation-flows analysis. We find convergence toward similar cell-state compositions at E7.5, supported by the quantitatively conserved expression of 76 transcription factors, despite divergence in surrounding trophoblast and hypoblast signaling. However, we observed noticeable changes in specification timing of some lineages and divergence of primordial germ cell programs, which in the rabbit do not activate mesoderm genes. Comparative analysis of temporal differentiation models provides a basis for studying the evolution of gastrulation dynamics across mammals.

Inflammatory signals from fatty bone marrow support DNMT3A driven clonal hematopoiesis

Both fatty bone marrow (FBM) and somatic mutations in hematopoietic stem cells (HSCs), also termed clonal hematopoiesis (CH) accumulate with human aging. However it remains unclear whether FBM can modify the evolution of CH. To address this question, we herein present the interaction between CH and FBM in two preclinical male mouse models: after sub-lethal irradiation or after castration. An adipogenesis inhibitor (PPARγ inhibitor) is used in both models as a control. A significant increase in self-renewal can be detected in both human and rodent DNMT3A^Mut-HSCs when exposed to FBM. DNMT3A^Mut-HSCs derived from older mice interacting with FBM have even higher self-renewal in comparison to DNMT3A^Mut-HSCs derived from younger mice. Single cell RNA-sequencing on rodent HSCs after exposing them to FBM reveal a 6-10 fold increase in DNMT3A^Mut-HSCs and an activated inflammatory signaling. Cytokine analysis of BM fluid and BM derived adipocytes grown in vitro demonstrates an increased IL-6 levels under FBM conditions. Anti-IL-6 neutralizing antibodies significantly reduce the selective advantage of DNMT3A^Mut-HSCs exposed to FBM. Overall, paracrine FBM inflammatory signals promote DNMT3A-driven clonal hematopoiesis, which can be inhibited by blocking the IL-6 pathway.

Mouse embryo model derived exclusively from embryonic stem cells undergoes neurulation and heart development

Several in vitro models have been developed to recapitulate mouse embryogenesis solely from embryonic stem cells (ESCs). Despite mimicking many aspects of early development, they fail to capture the interactions between embryonic and extraembryonic tissues. To overcome this difficulty, we have developed a mouse ESC-based in vitro model that reconstitutes the pluripotent ESC lineage and the two extraembryonic lineages of the post-implantation embryo by transcription-factor-mediated induction. This unified model recapitulates developmental events from embryonic day 5.5 to 8.5, including gastrulation; formation of the anterior-posterior axis, brain, and a beating heart structure; and the development of extraembryonic tissues, including yolk sac and chorion. Comparing single-cell RNA sequencing from individual structures with time-matched natural embryos identified remarkably similar transcriptional programs across lineages but also showed when and where the model diverges from the natural program. Our findings demonstrate an extraordinary plasticity of ESCs to self-organize and generate a whole-embryo-like structure.