Single-cell atlases: shared and tissue-specific cell types across human organs

Nuclei Isolation from Ocular Tissues of the Embryonic Chicken for Single-Nucleus Profiling

The generation of quality data from a single-nucleus profiling experiment requires nuclei to be isolated from tissues in a gentle and efficient manner. Nuclei isolation must be carefully optimized across tissue types to preserve nuclear architecture, prevent nucleic acid degradation, and remove unwanted contaminants. Here, we present an optimized workflow for generating a single-nucleus suspension from ocular tissues of the embryonic chicken that is compatible with various downstream workflows. The described protocol enables the rapid isolation of a high yield of aggregate-free nuclei from the embryonic chicken eye without compromising nucleic acid integrity, and the nuclei suspension is compatible with single-nucleus RNA and ATAC sequencing. We detail several stopping points, either via cryopreservation or fixation, to enhance workflow adaptability. Further, we provide a guide through multiple QC points and demonstrate proof-of-principle using two commercially available kits. Finally, we demonstrate that existing in silico genotyping methods can be adopted to computationally derive biological replicates from a single pool of chicken nuclei, greatly reducing the cost of biological replication and allowing researchers to consider sex as a variable during analysis. Together, this tutorial represents a cost-effective, simple, and effective approach to single-nucleus profiling of embryonic chicken eye tissues and is likely to be easily modified to be compatible with similar tissue types.

Cross-tissue human fibroblast atlas reveals myofibroblast subtypes with distinct roles in immune modulation

Fibroblasts, known for their functional diversity, play crucial roles in inflammation and cancer. In this study, we conduct comprehensive single-cell RNA sequencing analyses on fibroblast cells from 517 human samples, spanning 11 tissue types and diverse pathological states. We identify distinct fibroblast subpopulations with universal and tissue-specific characteristics. Pathological conditions lead to significant shifts in fibroblast compositions, including the expansion of immune-modulating fibroblasts during inflammation and tissue-remodeling myofibroblasts in cancer. Within the myofibroblast category, we identify four transcriptionally distinct subpopulations originating from different developmental origins, with LRRC15+ myofibroblasts displaying terminally differentiated features. Both LRRC15+ and MMP1+ myofibroblasts demonstrate pro-tumor potential that contribute to the immune-excluded and immune-suppressive tumor microenvironments (TMEs), whereas PI16+ fibroblasts show potential anti-tumor functions in adjacent non-cancerous regions. Fibroblast-subtype compositions define patient subtypes with distinct clinical outcomes. This study advances our understanding of fibroblast biology and suggests potential therapeutic strategies for targeting specific fibroblast subsets in cancer treatment.

Universal and tissue-specific fibroblasts in chronic inflammation and cancer

In this issue of Cancer Cell, Gao et al. map fibroblast diversity across tumors and chronic inflammatory tissues. The authors uncover universal fibroblast subtypes such as LRRC15+ and MMP1+ myofibroblasts along with specialized tissue-specific subtypes. They reveal cellular roles of fibroblasts in immunosuppression through stromal niches and cell-cell interactions.

New frameworks for hematopoiesis derived from single-cell genomics

ABSTRACT

Recent advancements in single-cell genomics have enriched our understanding of hematopoiesis, providing intricate details about hematopoietic stem cell biology, differentiation, and lineage commitment. Technological advancements have highlighted extensive heterogeneity of cell populations and continuity of differentiation routes. Nevertheless, intermediate "attractor" states signify structure in stem and progenitor populations that link state transition dynamics to fate potential. We discuss how innovative model systems quantify lineage bias and how stress accelerates differentiation, thereby reducing fate plasticity compared with native hematopoiesis. We conclude by offering our perspective on the current model of hematopoiesis and discuss how a more precise understanding can translate to strategies that extend healthy hematopoiesis and prevent disease.

Building a learnable universal coordinate system for single-cell atlas with a joint-VAE model

A universal coordinate system that can ensemble the huge number of cells and capture their heterogeneities is of vital importance for constructing large-scale cell atlases as references for molecular and cellular studies. Studies have shown that cells exhibit multifaceted heterogeneities in their transcriptomic features at multiple resolutions. This nature of complexity makes it hard to design a fixed coordinate system through a combination of known features. It is desirable to build a learnable universal coordinate model that can capture major heterogeneities and serve as a controlled generative model for data augmentation. We developed UniCoord, a specially-tuned joint-VAE model to represent single-cell transcriptomic data in a lower-dimensional latent space with high interpretability. Each latent dimension can represent either discrete or continuous feature, and either supervised by prior knowledge or unsupervised. The latent dimensions can be easily reconfigured to generate pseudo transcriptomic profiles with desired properties. UniCoord can also be used as a pre-trained model to analyze new data with unseen cell types and thus can serve as a feasible framework for cell annotation and comparison. UniCoord provides a prototype for a learnable universal coordinate framework to enable better analysis and generation of cells with highly orchestrated functions and heterogeneities.

Scalable spatial single-cell transcriptomics and translatomics in 3D thick tissue blocks

Characterizing the transcriptional and translational gene expression patterns at the single-cell level within their three-dimensional (3D) tissue context is essential for revealing how genes shape tissue structure and function in health and disease. However, most existing spatial profiling techniques are limited to 5-20 μm thin tissue sections. Here, we developed Deep-STARmap and Deep-RIBOmap, which enable 3D in situ quantification of thousands of gene transcripts and their corresponding translation activities, respectively, within 200-μm thick tissue blocks. This is achieved through scalable probe synthesis, hydrogel embedding with efficient probe anchoring, and robust cDNA crosslinking. We first utilized Deep-STARmap in combination with multicolor fluorescent protein imaging for simultaneous molecular cell typing and 3D neuron morphology tracing in the mouse brain. We also demonstrate that 3D spatial profiling facilitates comprehensive and quantitative analysis of tumor-immune interactions in human skin cancer.

aKNNO: single-cell and spatial transcriptomics clustering with an optimized adaptive k-nearest neighbor graph

Typical clustering methods for single-cell and spatial transcriptomics struggle to identify rare cell types, while approaches tailored to detect rare cell types gain this ability at the cost of poorer performance for grouping abundant ones. Here, we develop aKNNO to simultaneously identify abundant and rare cell types based on an adaptive k-nearest neighbor graph with optimization. Benchmarking on 38 simulated and 20 single-cell and spatial transcriptomics datasets demonstrates that aKNNO identifies both abundant and rare cell types more accurately than general and specialized methods. Using only gene expression aKNNO maps abundant and rare cells more precisely compared to integrative approaches.

A systematic overview of single-cell transcriptomics databases, their use cases, and limitations

Rapid advancements in high-throughput single-cell RNA-seq (scRNA-seq) technologies and experimental protocols have led to the generation of vast amounts of transcriptomic data that populates several online databases and repositories. Here, we systematically examined large-scale scRNA-seq databases, categorizing them based on their scope and purpose such as general, tissue-specific databases, disease-specific databases, cancer-focused databases, and cell type-focused databases. Next, we discuss the technical and methodological challenges associated with curating large-scale scRNA-seq databases, along with current computational solutions. We argue that understanding scRNA-seq databases, including their limitations and assumptions, is crucial for effectively utilizing this data to make robust discoveries and identify novel biological insights. Such platforms can help bridge the gap between computational and wet lab scientists through user-friendly web-based interfaces needed for democratizing access to single-cell data. These platforms would facilitate interdisciplinary research, enabling researchers from various disciplines to collaborate effectively. This review underscores the importance of leveraging computational approaches to unravel the complexities of single-cell data and offers a promising direction for future research in the field.

Aging atlas reveals cell-type-specific effects of pro-longevity strategies

Organismal aging involves functional declines in both somatic and reproductive tissues. Multiple strategies have been discovered to extend lifespan across species. However, how age-related molecular changes differ among various tissues and how those lifespan-extending strategies slow tissue aging in distinct manners remain unclear. Here we generated the transcriptomic Cell Atlas of Worm Aging (CAWA, http://mengwanglab.org/atlas ) of wild-type and long-lived strains. We discovered cell-specific, age-related molecular and functional signatures across all somatic and germ cell types. We developed transcriptomic aging clocks for different tissues and quantitatively determined how three different pro-longevity strategies slow tissue aging distinctively. Furthermore, through genome-wide profiling of alternative polyadenylation (APA) events in different tissues, we discovered cell-type-specific APA changes during aging and revealed how these changes are differentially affected by the pro-longevity strategies. Together, this study offers fundamental molecular insights into both somatic and reproductive aging and provides a valuable resource for in-depth understanding of the diversity of pro-longevity mechanisms.

Deciphering metabolic crosstalk in context: lessons from inflammatory diseases

Metabolism plays a crucial role in regulating the function of immune cells in both health and disease, with altered metabolism contributing to the pathogenesis of cancer and many inflammatory diseases. The local microenvironment has a profound impact on the metabolism of immune cells. Therefore, immunological and metabolic heterogeneity as well as the spatial organization of cells in tissues should be taken into account when studying immunometabolism. Here, we highlight challenges of investigating metabolic communication. Additionally, we review the capabilities and limitations of current technologies for studying metabolism in inflamed microenvironments, including single-cell omics techniques, flow cytometry-based methods (Met-Flow, single-cell energetic metabolism by profiling translation inhibition (SCENITH)), cytometry by time of flight (CyTOF), cellular indexing of transcriptomes and epitopes by sequencing (CITE-Seq), and mass spectrometry imaging. Considering the importance of metabolism in regulating immune cells in diseased states, we also discuss the applications of metabolomics in clinical research, as well as some hurdles to overcome to implement these techniques in standard clinical practice. Finally, we provide a flowchart to assist scientists in designing effective strategies to unravel immunometabolism in disease-relevant contexts.

Rat Models of Hormone Receptor-Positive Breast Cancer

Hormone receptor-positive (HR+) breast cancer (BC) is the most common type of breast cancer among women worldwide, accounting for 70-80% of all invasive cases. Patients with HR+ BC are commonly treated with endocrine therapy, but intrinsic or acquired resistance is a frequent problem, making HR+ BC a focal point of intense research. Despite this, the malignancy still lacks adequate in vitro and in vivo models for the study of its initiation and progression as well as response and resistance to endocrine therapy. No mouse models that fully mimic the human disease are available, however rat mammary tumor models pose a promising alternative to overcome this limitation. Compared to mice, rats are more similar to humans in terms of mammary gland architecture, ductal origin of neoplastic lesions and hormone dependency status. Moreover, rats can develop spontaneous or induced mammary tumors that resemble human HR+ BC. To date, six different types of rat models of HR+ BC have been established. These include the spontaneous, carcinogen-induced, transplantation, hormone-induced, radiation-induced and genetically engineered rat mammary tumor models. Each model has distinct advantages, disadvantages and utility for studying HR+ BC. This review provides a comprehensive overview of all published models to date.

Machine learning integrative approaches to advance computational immunology

The study of immunology, traditionally reliant on proteomics to evaluate individual immune cells, has been revolutionized by single-cell RNA sequencing. Computational immunologists play a crucial role in analysing these datasets, moving beyond traditional protein marker identification to encompass a more detailed view of cellular phenotypes and their functional roles. Recent technological advancements allow the simultaneous measurements of multiple cellular components-transcriptome, proteome, chromatin, epigenetic modifications and metabolites-within single cells, including in spatial contexts within tissues. This has led to the generation of complex multiscale datasets that can include multimodal measurements from the same cells or a mix of paired and unpaired modalities. Modern machine learning (ML) techniques allow for the integration of multiple "omics" data without the need for extensive independent modelling of each modality. This review focuses on recent advancements in ML integrative approaches applied to immunological studies. We highlight the importance of these methods in creating a unified representation of multiscale data collections, particularly for single-cell and spatial profiling technologies. Finally, we discuss the challenges of these holistic approaches and how they will be instrumental in the development of a common coordinate framework for multiscale studies, thereby accelerating research and enabling discoveries in the computational immunology field.

A Molecular Voyage: Multiomics Insights into Circulating Tumor Cells

Circulating tumor cells (CTCs) play a pivotal role in metastasis, the leading cause of cancer-associated death. Recent improvements of CTC isolation tools, coupled with a steady development of multiomics technologies at single-cell resolution, have enabled an extensive exploration of CTC biology, unlocking insights into their molecular profiles. A detailed molecular portrait requires CTC interrogation across various levels encompassing genomic, epigenetic, transcriptomic, proteomic and metabolic features. Here, we review how state-of-the-art multiomics applied to CTCs are shedding light on how cancer spreads. Further, we highlight the potential implications of CTC profiling for clinical applications aimed at enhancing cancer diagnosis and treatment.

SIGNIFICANCE

Exploring the complexity of cancer progression through cutting-edge multiomics studies holds the promise of uncovering novel aspects of cancer biology and identifying therapeutic vulnerabilities to suppress metastasis.

Deep generative models in single-cell omics

Deep Generative Models (DGMs) are becoming instrumental for inferring probability distributions inherent to complex processes, such as most questions in biomedical research. For many years, there was a lack of mathematical methods that would allow this inference in the scarce data scenario of biomedical research. The advent of single-cell omics has finally made square the so-called "skinny matrix", allowing to apply mathematical methods already extensively used in other areas. Moreover, it is now possible to integrate data at different molecular levels in thousands or even millions of samples, thanks to the number of single-cell atlases being collaboratively generated. Additionally, DGMs have proven useful in other frequent tasks in single-cell analysis pipelines, from dimensionality reduction, cell type annotation to RNA velocity inference. In spite of its promise, DGMs need to be used with caution in biomedical research, paying special attention to its use to answer the right questions and the definition of appropriate error metrics and validation check points that confirm not only its correct use but also its relevance. All in all, DGMs provide an exciting tool that opens a bright future for the integrative analysis of single-cell -omics to understand health and disease.

Characterizing efficient feature selection for single-cell expression analysis

Unsupervised feature selection is a critical step for efficient and accurate analysis of single-cell RNA-seq data. Previous benchmarks used two different criteria to compare feature selection methods: (i) proportion of ground-truth marker genes included in the selected features and (ii) accuracy of cell clustering using ground-truth cell types. Here, we systematically compare the performance of 11 feature selection methods for both criteria. We first demonstrate the discordance between these criteria and suggest using the latter. We then compare the distribution of selected genes in their means between feature selection methods. We show that lowly expressed genes exhibit seriously high coefficients of variation and are mostly excluded by high-performance methods. In particular, high-deviation- and high-expression-based methods outperform the widely used in Seurat package in clustering cells and data visualization. We further show they also enable a clear separation of the same cell type from different tissues as well as accurate estimation of cell trajectories.

Stochasticity in genetics and gene regulation

Development from fertilized egg to functioning multi-cellular organism requires precision. There is no precision, and often no survival, without plasticity. Plasticity is conferred partly by stochastic variation, present inherently in all biological systems. Gene expression levels fluctuate ubiquitously through transcription, alternative splicing, translation and turnover. Small differences in gene expression are exploited to trigger early differentiation, conferring distinct function on selected individual cells and setting in motion regulatory interactions. Non-selected cells then acquire new functions along the spatio-temporal developmental trajectory. The differentiation process has many stochastic components. Meiotic segregation, mitochondrial partitioning, X-inactivation and the dynamic DNA binding of transcription factor assemblies-all exhibit randomness. Non-random X-inactivation generally signals deleterious X-linked mutations. Correct neural wiring, such as retina to brain, arises through repeated confirmatory activity of connections made randomly. In immune system development, both B-cell antibody generation and the emergence of balanced T-cell categories begin through stochastic trial and error followed by functional selection. Aberrant selection processes lead to immune dysfunction. DNA sequence variants also arise through stochastic events: some involving environmental fluctuation (radiation or presence of pollutants), or genetic repair system malfunction. The phenotypic outcome of mutations is also fluid. Mutations may be advantageous in some circumstances, deleterious in others. This article is part of a discussion meeting issue 'Causes and consequences of stochastic processes in development and disease'.

Single Cell Atlas: a single-cell multi-omics human cell encyclopedia

Single-cell sequencing datasets are key in biology and medicine for unraveling insights into heterogeneous cell populations with unprecedented resolution. Here, we construct a single-cell multi-omics map of human tissues through in-depth characterizations of datasets from five single-cell omics, spatial transcriptomics, and two bulk omics across 125 healthy adult and fetal tissues. We construct its complement web-based platform, the Single Cell Atlas (SCA, www.singlecellatlas.org ), to enable vast interactive data exploration of deep multi-omics signatures across human fetal and adult tissues. The atlas resources and database queries aspire to serve as a one-stop, comprehensive, and time-effective resource for various omics studies.

Prediction of protein-RNA interactions from single-cell transcriptomic data

Proteins are crucial in regulating every aspect of RNA life, yet understanding their interactions with coding and noncoding RNAs remains limited. Experimental studies are typically restricted to a small number of cell lines and a limited set of RNA-binding proteins (RBPs). Although computational methods based on physico-chemical principles can predict protein-RNA interactions accurately, they often lack the ability to consider cell-type-specific gene expression and the broader context of gene regulatory networks (GRNs). Here, we assess the performance of several GRN inference algorithms in predicting protein-RNA interactions from single-cell transcriptomic data, and propose a pipeline, called scRAPID (single-cell transcriptomic-based RnA Protein Interaction Detection), that integrates these methods with the catRAPID algorithm, which can identify direct physical interactions between RBPs and RNA molecules. Our approach demonstrates that RBP-RNA interactions can be predicted from single-cell transcriptomic data, with performances comparable or superior to those achieved for the well-established task of inferring transcription factor-target interactions. The incorporation of catRAPID significantly enhances the accuracy of identifying interactions, particularly with long noncoding RNAs, and enables the identification of hub RBPs and RNAs. Additionally, we show that interactions between RBPs can be detected based on their inferred RNA targets. The software is freely available at https://github.com/tartaglialabIIT/scRAPID.

scButterfly: a versatile single-cell cross-modality translation method via dual-aligned variational autoencoders

Recent advancements for simultaneously profiling multi-omics modalities within individual cells have enabled the interrogation of cellular heterogeneity and molecular hierarchy. However, technical limitations lead to highly noisy multi-modal data and substantial costs. Although computational methods have been proposed to translate single-cell data across modalities, broad applications of the methods still remain impeded by formidable challenges. Here, we propose scButterfly, a versatile single-cell cross-modality translation method based on dual-aligned variational autoencoders and data augmentation schemes. With comprehensive experiments on multiple datasets, we provide compelling evidence of scButterfly's superiority over baseline methods in preserving cellular heterogeneity while translating datasets of various contexts and in revealing cell type-specific biological insights. Besides, we demonstrate the extensive applications of scButterfly for integrative multi-omics analysis of single-modality data, data enhancement of poor-quality single-cell multi-omics, and automatic cell type annotation of scATAC-seq data. Moreover, scButterfly can be generalized to unpaired data training, perturbation-response analysis, and consecutive translation.

Overcoming barriers to single-cell RNA sequencing adoption in low- and middle-income countries

The advent of single-cell resolution sequencing and spatial transcriptomics has enabled the delivery of cellular and molecular atlases of tissues and organs, providing new insights into tissue health and disease. However, if the full potential of these technologies is to be equitably realised, ancestrally inclusivity is paramount. Such a goal requires greater inclusion of both researchers and donors in low- and middle-income countries (LMICs). In this perspective, we describe the current landscape of ancestral inclusivity in genomic and single-cell transcriptomic studies. We discuss the collaborative efforts needed to scale the barriers to establishing, expanding, and adopting single-cell sequencing research in LMICs and to enable globally impactful outcomes of these technologies.

Technology-enabled great leap in deciphering plant genomes

Plant genomes provide essential and vital basic resources for studying many aspects of plant biology and applications (for example, breeding). From 2000 to 2020, 1,144 genomes of 782 plant species were sequenced. In the past three years (2021-2023), 2,373 genomes of 1,031 plant species, including 793 newly sequenced species, have been assembled, representing a great leap. The 2,373 newly assembled genomes, of which 63 are telomere-to-telomere assemblies and 921 have been generated in pan-genome projects, cover the major phylogenetic clades. Substantial advances in read length, throughput, accuracy and cost-effectiveness have notably simplified the achievement of high-quality assemblies. Moreover, the development of multiple software tools using different algorithms offers the opportunity to generate more complete and complex assemblies. A database named N3: plants, genomes, technologies has been developed to accommodate the metadata associated with the 3,517 genomes that have been sequenced from 1,575 plant species since 2000. We also provide an outlook for emerging opportunities in plant genome sequencing.

Cellular heterogeneity and dynamics of the human uterus in healthy premenopausal women

The human uterus is a complex and dynamic organ whose lining grows, remodels, and regenerates in every menstrual cycle or upon tissue damage. Here we applied single-cell RNA sequencing to profile more the 50,000 uterine cells from both the endometrium and myometrium of 5 healthy premenopausal individuals, and jointly analyzed the data with a previously published dataset from 15 subjects. The resulting normal uterus cell atlas contains more than 167K cells representing the lymphatic endothelium, blood endothelium, stromal, ciliated epithelium, unciliated epithelium, and immune cell populations. Focused analyses within each major cell type and comparisons with subtype labels from prior studies allowed us to document supporting evidence, resolve naming conflicts, and to propose a consensus annotation system of 39 subtypes. We release their gene expression centroids, differentially expressed genes, and mRNA patterns of literature-based markers as a shared community resource. We find many subtypes show dynamic changes over different phases of the cycle and identify multiple potential progenitor cells: compartment-wide progenitors for each major cell type, transitional cells that are upstream of other subtypes, and potential cross-lineage multipotent stromal progenitors that may be capable of replenishing the epithelial, stromal, and endothelial compartments. When compared to the healthy premenopausal samples, a postpartum and a postmenopausal uterus sample revealed substantially altered tissue composition, involving the rise or fall of stromal, endothelial, and immune cells. The cell taxonomy and molecular markers we report here are expected to inform studies of both basic biology of uterine function and its disorders.

SIGNIFICANCE

We present single-cell RNA sequencing data from seven individuals (five healthy pre-menopausal women, one post-menopausal woman, and one postpartum) and perform an integrated analysis of this data alongside 15 previously published scRNA-seq datasets. We identified 39 distinct cell subtypes across four major cell types in the uterus. By using RNA velocity analysis and centroid-centroid comparisons we identify multiple computationally predicted progenitor populations for each of the major cell compartments, as well as potential cross-compartment, multi-potent progenitors. While the function and interactions of these cell populations remain to be validated through future experiments, the markers and their "dual characteristics" that we describe will serve as a rich resource to the scientific community. Importantly, we address a significant challenge in the field: reconciling multiple uterine cell taxonomies being proposed. To achieve this, we focused on integrating historical and contemporary knowledge across multiple studies. By providing detailed evidence used for cell classification we lay the groundwork for establishing a stable, consensus cell atlas of the human uterus.

Principled and interpretable alignability testing and integration of single-cell data

Single-cell data integration can provide a comprehensive molecular view of cells, and many algorithms have been developed to remove unwanted technical or biological variations and integrate heterogeneous single-cell datasets. Despite their wide usage, existing methods suffer from several fundamental limitations. In particular, we lack a rigorous statistical test for whether two high-dimensional single-cell datasets are alignable (and therefore should even be aligned). Moreover, popular methods can substantially distort the data during alignment, making the aligned data and downstream analysis difficult to interpret. To overcome these limitations, we present a spectral manifold alignment and inference (SMAI) framework, which enables principled and interpretable alignability testing and structure-preserving integration of single-cell data with the same type of features. SMAI provides a statistical test to robustly assess the alignability between datasets to avoid misleading inference and is justified by high-dimensional statistical theory. On a diverse range of real and simulated benchmark datasets, it outperforms commonly used alignment methods. Moreover, we show that SMAI improves various downstream analyses such as identification of differentially expressed genes and imputation of single-cell spatial transcriptomics, providing further biological insights. SMAI's interpretability also enables quantification and a deeper understanding of the sources of technical confounders in single-cell data.

Uncertainty-aware single-cell annotation with a hierarchical reject option

MOTIVATION

Automatic cell type annotation methods assign cell type labels to new datasets by extracting relationships from a reference RNA-seq dataset. However, due to the limited resolution of gene expression features, there is always uncertainty present in the label assignment. To enhance the reliability and robustness of annotation, most machine learning methods address this uncertainty by providing a full reject option, i.e. when the predicted confidence score of a cell type label falls below a user-defined threshold, no label is assigned and no prediction is made. As a better alternative, some methods deploy hierarchical models and consider a so-called partial rejection by returning internal nodes of the hierarchy as label assignment. However, because a detailed experimental analysis of various rejection approaches is missing in the literature, there is currently no consensus on best practices.

RESULTS

We evaluate three annotation approaches (i) full rejection, (ii) partial rejection, and (iii) no rejection for both flat and hierarchical probabilistic classifiers. Our findings indicate that hierarchical classifiers are superior when rejection is applied, with partial rejection being the preferred rejection approach, as it preserves a significant amount of label information. For optimal rejection implementation, the rejection threshold should be determined through careful examination of a method's rejection behavior. Without rejection, flat and hierarchical annotation perform equally well, as long as the cell type hierarchy accurately captures transcriptomic relationships.

AVAILABILITY AND IMPLEMENTATION

Code is freely available at https://github.com/Latheuni/Hierarchical_reject and https://doi.org/10.5281/zenodo.10697468.

Tutorial on survival modeling with applications to omics data

MOTIVATION

Identification of genomic, molecular and clinical markers prognostic of patient survival is important for developing personalized disease prevention, diagnostic and treatment approaches. Modern omics technologies have made it possible to investigate the prognostic impact of markers at multiple molecular levels, including genomics, epigenomics, transcriptomics, proteomics and metabolomics, and how these potential risk factors complement clinical characterization of patient outcomes for survival prognosis. However, the massive sizes of the omics datasets, along with their correlation structures, pose challenges for studying relationships between the molecular information and patients' survival outcomes.

RESULTS

We present a general workflow for survival analysis that is applicable to high-dimensional omics data as inputs when identifying survival-associated features and validating survival models. In particular, we focus on the commonly used Cox-type penalized regressions and hierarchical Bayesian models for feature selection in survival analysis, which are especially useful for high-dimensional data, but the framework is applicable more generally.

AVAILABILITY AND IMPLEMENTATION

A step-by-step R tutorial using The Cancer Genome Atlas survival and omics data for the execution and evaluation of survival models has been made available at https://ocbe-uio.github.io/survomics.

Gene panel selection for targeted spatial transcriptomics

Targeted spatial transcriptomics hold particular promise in analyzing complex tissues. Most such methods, however, measure only a limited panel of transcripts, which need to be selected in advance to inform on the cell types or processes being studied. A limitation of existing gene selection methods is their reliance on scRNA-seq data, ignoring platform effects between technologies. Here we describe gpsFISH, a computational method performing gene selection through optimizing detection of known cell types. By modeling and adjusting for platform effects, gpsFISH outperforms other methods. Furthermore, gpsFISH can incorporate cell type hierarchies and custom gene preferences to accommodate diverse design requirements.

Deciphering Drug Targets and Actions with Single-Cell and Spatial Resolution

Recent advances in chemical, molecular, and genetic approaches have provided us with an unprecedented capacity to identify drug-target interactions across the whole proteome and genome. Meanwhile, rapid developments of single-cell and spatial omics technologies are revolutionizing our understanding of the molecular architecture of biological systems. However, a significant gap remains in how we align our understanding of drug actions, traditionally based on molecular affinities, with the in vivo cellular and spatial tissue heterogeneity revealed by these newer techniques. Here, we review state-of-the-art methods for profiling drug-target interactions and emerging multiomics tools to delineate the tissue heterogeneity at single-cell resolution. Highlighting the recent technical advances enabling high-resolution, multiplexable in situ small-molecule drug imaging (clearing-assisted tissue click chemistry, or CATCH), we foresee the integration of single-cell and spatial omics platforms, data, and concepts into the future framework of defining and understanding in vivo drug-target interactions and mechanisms of actions.

Building a two-way street between cell biology and machine learning

With biomedical sciences quickly outgrowing many other application areas in terms of data generation, there is a unique opportunity for life sciences to become one of the greatest beneficiaries of research in machine learning and AI, and also inspire foundational developments in it.

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.

A critical assessment of clustering algorithms to improve cell clustering and identification in single-cell transcriptome study

Cell clustering is typically the initial step in single-cell RNA sequencing (scRNA-seq) analyses. The performance of clustering considerably impacts the validity and reproducibility of cell identification. A variety of clustering algorithms have been developed for scRNA-seq data. These algorithms generate cell label sets that assign each cell to a cluster. However, different algorithms usually yield different label sets, which can introduce variations in cell-type identification based on the generated label sets. Currently, the performance of these algorithms has not been systematically evaluated in single-cell transcriptome studies. Herein, we performed a critical assessment of seven state-of-the-art clustering algorithms including four deep learning-based clustering algorithms and commonly used methods Seurat, Cosine-based Tanimoto similarity-refined graph for community detection using Leiden's algorithm (CosTaL) and Single-cell consensus clustering (SC3). We used diverse evaluation indices based on 10 different scRNA-seq benchmarks to systematically evaluate their clustering performance. Our results show that CosTaL, Seurat, Deep Embedding for Single-cell Clustering (DESC) and SC3 consistently outperformed Single-Cell Clustering Assessment Framework and scDeepCluster based on nine effectiveness scores. Notably, CosTaL and DESC demonstrated superior performance in clustering specific cell types. The performance of the single-cell Variational Inference tools varied across different datasets, suggesting its sensitivity to certain dataset characteristics. Notably, DESC exhibited promising results for cell subtype identification and capturing cellular heterogeneity. In addition, SC3 requires more memory and exhibits slower computation speed compared to other algorithms for the same dataset. In sum, this study provides useful guidance for selecting appropriate clustering methods in scRNA-seq data analysis.

Pathogenetic mechanisms and treatment targets in cerebral malaria

Malaria, the most prevalent mosquito-borne infectious disease worldwide, has accompanied humanity for millennia and remains an important public health issue despite advances in its prevention and treatment. Most infections are asymptomatic, but a small percentage of individuals with a heavy parasite burden develop severe malaria, a group of clinical syndromes attributable to organ dysfunction. Cerebral malaria is an infrequent but life-threatening complication of severe malaria that presents as an acute cerebrovascular encephalopathy characterized by unarousable coma. Despite effective antiparasite drug treatment, 20% of patients with cerebral malaria die from this disease, and many survivors of cerebral malaria have neurocognitive impairment. Thus, an important unmet clinical need is to rapidly identify people with malaria who are at risk of developing cerebral malaria and to develop preventive, adjunctive and neuroprotective treatments for cerebral malaria. This Review describes important advances in the understanding of cerebral malaria over the past two decades and discusses how these mechanistic insights could be translated into new therapies.

Single-cell transcriptome profiling reveals immune and stromal cell heterogeneity in primary Sjögren's syndrome

Primary Sjögren's syndrome (pSS) is a complex autoimmune disease characterized by lymphocytic infiltration and exocrine dysfunction, particularly affecting the salivary gland (SG). We employed single-cell RNA sequencing to investigate cellular heterogeneity in 11 patients with pSS and 5 non-SS controls. Notably, patients with pSS exhibited downregulated SOX9 in myoepithelial cells, potentially associated with impaired epithelial regeneration. An expanded ACKR1+ endothelial subpopulation in patients with pSS suggested a role in facilitating lymphocyte transendothelial migration. Our analysis of immune cells revealed expanded IGHD+ naive B cells in peripheral blood from patients with pSS. Pseudotime trajectory analysis outlined a bifurcated differentiation pathway for peripheral B cells, enriching three subtypes (VPREB3+ B, BANK1+ B, CD83+ B cells) within SGs in patients with pSS. Fibroblasts emerged as pivotal components in a stromal-immune interaction network, potentially driving extracellular matrix disruption, epithelial regeneration impairment, and inflammation. Our study illuminates immune and stromal cell heterogeneity in patients with pSS, offering insights into therapeutic strategies.

Single-cell and Spatial Transcriptomics Clustering with an Optimized Adaptive K-Nearest Neighbor Graph

Single-cell and spatial transcriptomics have been widely used to characterize cellular landscape in complex tissues. To understand cellular heterogeneity, one essential step is to define cell types through unsupervised clustering. While typical clustering methods have difficulty in identifying rare cell types, approaches specifically tailored to detect rare cell types gain their ability at the cost of poorer performance for grouping abundant ones. Here, we developed aKNNO, a method to identify abundant and rare cell types simultaneously based on an adaptive k-nearest neighbor graph with optimization. Benchmarked on 38 simulated and 20 single-cell and spatial transcriptomics datasets, aKNNO identified both abundant and rare cell types accurately. Without sacrificing performance for clustering abundant cell types, aKNNO discovered known and novel rare cell types that those typical and even specifically tailored methods failed to detect. aKNNO, using transcriptome alone, stereotyped fine-grained anatomical structures more precisely than those integrative approaches combining expression with spatial locations and histology image.

Decoding Human Biology and Disease Using Single-cell Omics Technologies

Over the past decade, advances in single-cell omics (SCO) technologies have enabled the investigation of cellular heterogeneity at an unprecedented resolution and scale, opening a new avenue for understanding human biology and disease. In this review, we summarize the developments of sequencing-based SCO technologies and computational methods, and focus on considerable insights acquired from SCO sequencing studies to understand normal and diseased properties, with a particular emphasis on cancer research. We also discuss the technological improvements of SCO and its possible contribution to fundamental research of the human, as well as its great potential in clinical diagnoses and personalized therapies of human disease.

The technological landscape and applications of single-cell multi-omics

Single-cell multi-omics technologies and methods characterize cell states and activities by simultaneously integrating various single-modality omics methods that profile the transcriptome, genome, epigenome, epitranscriptome, proteome, metabolome and other (emerging) omics. Collectively, these methods are revolutionizing molecular cell biology research. In this comprehensive Review, we discuss established multi-omics technologies as well as cutting-edge and state-of-the-art methods in the field. We discuss how multi-omics technologies have been adapted and improved over the past decade using a framework characterized by optimization of throughput and resolution, modality integration, uniqueness and accuracy, and we also discuss multi-omics limitations. We highlight the impact that single-cell multi-omics technologies have had in cell lineage tracing, tissue-specific and cell-specific atlas production, tumour immunology and cancer genetics, and in mapping of cellular spatial information in fundamental and translational research. Finally, we discuss bioinformatics tools that have been developed to link different omics modalities and elucidate functionality through the use of better mathematical modelling and computational methods.

Spatio-temporal heterogeneity in cancer evolution and tumor microenvironment of renal cell carcinoma with tumor thrombus

Metastasis is the most fatal aspect of cancer, often preceded by a tumor thrombus (TT) which forms within the vascular system. Renal cell carcinoma (RCC), the predominant form of kidney cancer, witnesses a venous system invasion in 4-10% of cases, resulting in venous tumor thrombus (RCC-TT). This variant represents a formidable clinical challenge due to its escalated surgical complexity, heightened risk of progression and metastasis, and an adverse prognosis. However, recent trials addressing RCC-TT face significant barriers stemming from the profound inter- and intra-tumoral heterogeneity, patient-specific treatment variations, and distinct therapeutic resistance patterns between the primary tumor (PT) and the TT. This review delves into the unique evolutionary pathway of RCC-TT, the relationship between the staging and grading of RCC-TT invasion patterns, and the spatial molecular profiling of RCC-TT. Additionally, we assess the temporal heterogeneity among TT, PT, and distant metastases, as well as the functional phenotypes of TME components. An outlook for future research on RCC-TT is also provided.

Single-cell transcriptome analysis reveals cellular heterogeneity and highlights Fstl1-regulated alveolar myofibroblasts in mouse lung at birth

The matricellular protein, follistatin-like 1 (FSTL1), regulates lung development and saccular formation. Here, we employed single-cell RNA sequencing (scRNA-seq) to construct a transcriptomic atlas of 22,774 individual cells from wild-type (WT) and Fstl1-/- lung (E18.5) samples and identified 27 cell subtypes. We observed abnormal population sizes and gene expression profiles in diverse cell subtypes in Fstl1-/- lung samples. We identified Pdgfra and Tgfbi as genetic markers specifically expressed in postnatal myofibroblasts (MyoFBs). Fstl1 deletion decreased the number of MyoFB cells and downregulated their roles in ECM organization and muscle tissue/vasculature development, partly through the TGF-β1/BMP4 signaling pathway. Our data provide a single-cell view of the cellular heterogeneity and the molecular mechanisms underlying abnormal saccular formation and atelectatic lungs in Fstl1-/- mice.

A deep generative framework with embedded vector arithmetic and classifier for sample generation, label transfer, and clustering of single-cell data

Multiple-source single-cell datasets have accumulated quickly and need computational methods to integrate and decompose into meaningful components. Here, we present inClust (integrated clustering), a flexible deep generative framework that enables embedding auxiliary information, latent space vector arithmetic, and clustering. All functional parts are relatively modular, independent in implementation but interrelated at runtime, resulting in an all-in general framework that could work in supervised, semi-supervised, or unsupervised mode. We show that inClust is superior to most data integration methods in benchmark datasets. Then, we demonstrate the capability of inClust in the tasks of conditional out-of-distribution generation in supervised mode, label transfer in semi-supervised mode, and spatial domain identification in unsupervised mode. In these examples, inClust could accurately express the effect of each covariate, distinguish the query-specific cell types, or segment spatial domains. The results support that inClust is an excellent general framework for multiple-task harmonization and data decomposition.

How single-cell techniques help us look into lung cancer heterogeneity and immunotherapy

Lung cancer patients tend to have strong intratumoral and intertumoral heterogeneity and complex tumor microenvironment, which are major contributors to the efficacy of and drug resistance to immunotherapy. From a new perspective, single-cell techniques offer an innovative way to look at the intricate cellular interactions between tumors and the immune system and help us gain insights into lung cancer and its response to immunotherapy. This article reviews the application of single-cell techniques in lung cancer, with focuses directed on the heterogeneity of lung cancer and the efficacy of immunotherapy. This review provides both theoretical and experimental information for the future development of immunotherapy and personalized treatment for the management of lung cancer.

Robust collection and processing for label-free single voxel proteomics

With advanced mass spectrometry (MS)-based proteomics, genome-scale proteome coverage can be achieved from bulk tissues. However, such bulk measurement lacks spatial resolution and obscures important tissue heterogeneity, which make it impossible for proteome mapping of tissue microenvironment. Here we report an integrated wet collection of single tissue voxel and Surfactant-assisted One-Pot voxel processing method termed wcSOP for robust label-free single voxel proteomics. wcSOP capitalizes on buffer droplet-assisted wet collection of single tissue voxel dissected by LCM into the PCR tube cap and MS-compatible surfactant-assisted one-pot voxel processing in the collection cap. This convenient method allows reproducible label-free quantification of ∼900 and ∼4,600 proteins for single voxel from fresh frozen human spleen tissue at 20 μm × 20 μm × 10 μm (close to single cells) and 200 μm × 200 μm × 10 μm (∼100 cells), respectively. 100s-1000s of protein signatures with differential expression levels were identified to be spatially resolved between spleen red and white pulp regions depending on the voxel size. Region-specific signaling pathways were enriched from single voxel proteomics data. Antibody-based CODEX imaging was used to validate label-free MS quantitation for single voxel analysis. The wcSOP-MS method paves the way for routine robust single voxel proteomics and spatial proteomics.

Methods and applications for single-cell and spatial multi-omics

The joint analysis of the genome, epigenome, transcriptome, proteome and/or metabolome from single cells is transforming our understanding of cell biology in health and disease. In less than a decade, the field has seen tremendous technological revolutions that enable crucial new insights into the interplay between intracellular and intercellular molecular mechanisms that govern development, physiology and pathogenesis. In this Review, we highlight advances in the fast-developing field of single-cell and spatial multi-omics technologies (also known as multimodal omics approaches), and the computational strategies needed to integrate information across these molecular layers. We demonstrate their impact on fundamental cell biology and translational research, discuss current challenges and provide an outlook to the future.

Recurrent composite markers of cell types and states

Determining concise sets of genomic markers that identify cell types and states within tissue ecosystems remains challenging. To address this challenge, we developed Recurrent Composite Markers for Biological Identities with Neighborhood Enrichment (RECOMBINE). Validations of RECOMBINE with simulation and transcriptomics data in bulk, single-cell and spatial resolutions demonstrated the method's ability for unbiased selection of composite markers that characterize biological subpopulations. RECOMBINE captured markers of mouse visual cortex from single-cell RNA sequencing data and provided a gene panel for targeted spatial transcriptomics profiling. RECOMBINE identified composite markers of CD8 T cell states including GZMK ⁺ HAVCR2 ^- effector memory cells associated with anti-PD1 therapy response. The method outperformed differential gene expression analysis in characterizing a rare cell subpopulation within mouse intestine. Using RECOMBINE, we uncovered hierarchical gene programs of inter- and intra-tumoral heterogeneity in breast and skin tumors. In conclusion, RECOMBINE offers a data-driven approach for unbiased selection of composite markers, resulting in improved interpretation, discovery, and validation of cell types and states.

Defining Interactions Between the Genome, Epigenome, and the Environment in Inflammatory Bowel Disease: Progress and Prospects

Recent advances in our understanding of the pathogenesis of inflammatory bowel disease (IBD) have highlighted the complex interplay between the genome, the epigenome, and the environment. Despite the exciting advances in genomics that have enabled the identification of over 200 susceptibility loci, these only account for a small proportion of the disease variance and the estimated heritability in IBD. It is likely that gene-environment (GxE) interactions contribute to "missing heritability" and these may act through epigenetic mechanisms. Several environmental factors, such as the microbiome, nutrition, and tobacco smoking, induce alterations in the epigenome of children and adults, which may impact disease susceptibility. Other mechanisms for GxE interactions are also directly pertinent in early life. We discuss a model in which environmental factors imprint disease risk in a window of susceptibility during infancy that may contribute to later disease onset, whereas other elements of the exposome act later in life and contribute directly to the pathogenesis and course of the disease. Understanding the mechanisms underlying GxE interactions may provide the basis for new therapeutic targets or preventative strategies for IBD.

Learning consistent subcellular landmarks to quantify changes in multiplexed protein maps

Highly multiplexed imaging holds enormous promise for understanding how spatial context shapes the activity of the genome and its products at multiple length scales. Here, we introduce a deep learning framework called CAMPA (Conditional Autoencoder for Multiplexed Pixel Analysis), which uses a conditional variational autoencoder to learn representations of molecular pixel profiles that are consistent across heterogeneous cell populations and experimental perturbations. Clustering these pixel-level representations identifies consistent subcellular landmarks, which can be quantitatively compared in terms of their size, shape, molecular composition and relative spatial organization. Using high-resolution multiplexed immunofluorescence, this reveals how subcellular organization changes upon perturbation of RNA synthesis, RNA processing or cell size, and uncovers links between the molecular composition of membraneless organelles and cell-to-cell variability in bulk RNA synthesis rates. By capturing interpretable cellular phenotypes, we anticipate that CAMPA will greatly accelerate the systematic mapping of multiscale atlases of biological organization to identify the rules by which context shapes physiology and disease.

High sensitivity top-down proteomics captures single muscle cell heterogeneity in large proteoforms

Single-cell proteomics has emerged as a powerful method to characterize cellular phenotypic heterogeneity and the cell-specific functional networks underlying biological processes. However, significant challenges remain in single-cell proteomics for the analysis of proteoforms arising from genetic mutations, alternative splicing, and post-translational modifications. Herein, we have developed a highly sensitive functionally integrated top-down proteomics method for the comprehensive analysis of proteoforms from single cells. We applied this method to single muscle fibers (SMFs) to resolve their heterogeneous functional and proteomic properties at the single-cell level. Notably, we have detected single-cell heterogeneity in large proteoforms (>200 kDa) from the SMFs. Using SMFs obtained from three functionally distinct muscles, we found fiber-to-fiber heterogeneity among the sarcomeric proteoforms which can be related to the functional heterogeneity. Importantly, we detected multiple isoforms of myosin heavy chain (~223 kDa), a motor protein that drives muscle contraction, with high reproducibility to enable the classification of individual fiber types. This study reveals single muscle cell heterogeneity in large proteoforms and establishes a direct relationship between sarcomeric proteoforms and muscle fiber types, highlighting the potential of top-down proteomics for uncovering the molecular underpinnings of cell-to-cell variation in complex systems.

Single-cell transcriptomics for the assessment of cardiac disease

Cardiovascular disease is the leading cause of death globally. An advanced understanding of cardiovascular disease mechanisms is required to improve therapeutic strategies and patient risk stratification. State-of-the-art, large-scale, single-cell and single-nucleus transcriptomics facilitate the exploration of the cardiac cellular landscape at an unprecedented level, beyond its descriptive features, and can further our understanding of the mechanisms of disease and guide functional studies. In this Review, we provide an overview of the technical challenges in the experimental design of single-cell and single-nucleus transcriptomics studies, as well as a discussion of the type of inferences that can be made from the data derived from these studies. Furthermore, we describe novel findings derived from transcriptomics studies for each major cardiac cell type in both health and disease, and from development to adulthood. This Review also provides a guide to interpreting the exhaustive list of newly identified cardiac cell types and states, and highlights the consensus and discordances in annotation, indicating an urgent need for standardization. We describe advanced applications such as integration of single-cell data with spatial transcriptomics to map genes and cells on tissue and define cellular microenvironments that regulate homeostasis and disease progression. Finally, we discuss current and future translational and clinical implications of novel transcriptomics approaches, and provide an outlook of how these technologies will change the way we diagnose and treat heart disease.

Gene panel selection for targeted spatial transcriptomics

Targeted spatial transcriptomics hold particular promise in analysis of complex tissues. Most such methods, however, measure only a limited panel of transcripts, which need to be selected in advance to inform on the cell types or processes being studied. A limitation of existing gene selection methods is that they rely on scRNA-seq data, ignoring platform effects between technologies. Here we describe gpsFISH, a computational method to perform gene selection through optimizing detection of known cell types. By modeling and adjusting for platform effects, gpsFISH outperforms other methods. Furthermore, gpsFISH can incorporate cell type hierarchies and custom gene preferences to accommodate diverse design requirements.

Aging Atlas Reveals Cell-Type-Specific Regulation of Pro-longevity Strategies

Organism aging occurs at the multicellular level; however, how pro-longevity mechanisms slow down aging in different cell types remains unclear. We generated single-cell transcriptomic atlases across the lifespan of Caenorhabditis elegans under different pro-longevity conditions (http://mengwanglab.org/atlas). We found cell-specific, age-related changes across somatic and germ cell types and developed transcriptomic aging clocks for different tissues. These clocks enabled us to determine tissue-specific aging-slowing effects of different pro-longevity mechanisms, and identify major cell types sensitive to these regulations. Additionally, we provided a systemic view of alternative polyadenylation events in different cell types, as well as their cell-type-specific changes during aging and under different pro-longevity conditions. Together, this study provides molecular insights into how aging occurs in different cell types and how they respond to pro-longevity strategies.

Vascular endothelial cell development and diversity

Vascular endothelial cells form the inner layer of blood vessels where they have a key role in the development and maintenance of the functional circulatory system and provide paracrine support to surrounding non-vascular cells. Technical advances in the past 5 years in single-cell genomics and in in vivo genetic labelling have facilitated greater insights into endothelial cell development, plasticity and heterogeneity. These advances have also contributed to a new understanding of the timing of endothelial cell subtype differentiation and its relationship to the cell cycle. Identification of novel tissue-specific gene expression patterns in endothelial cells has led to the discovery of crucial signalling pathways and new interactions with other cell types that have key roles in both tissue maintenance and disease pathology. In this Review, we describe the latest findings in vascular endothelial cell development and diversity, which are often supported by large-scale, single-cell studies, and discuss the implications of these findings for vascular medicine. In addition, we highlight how techniques such as single-cell multimodal omics, which have become increasingly sophisticated over the past 2 years, are being utilized to study normal vascular physiology as well as functional perturbations in disease.