Cell Type Annotation Research Articles

Rediscovering publicly available single-cell data with the DISCO platform.

Single-cell RNA sequencing (scRNA-seq) has emerged as the key technique for studying transcriptomics at the single-cell level. In our previous work, we presented the DISCO database (https://www.immunesinglecell.org/) that integrates publicly available human scRNA-seq data. We now introduce an enhanced version of DISCO, which has expanded fourfold to include >100 million cells from >17 thousand samples. It provides uniformly realigned read count tables, curated metadata, integrated tissue and phenotype specific atlases, and harmonized cell type annotations. It also hosts a single-cell enhanced knowledgebase of cell type ontology and gene signatures relating to cell types and phenotypes. Lastly, it offers a suite of tools for data retrieval, integration, annotation, and mapping, allowing users to construct customized atlases and perform integrated analysis with their own data. These tools are also available in a standalone R package for offline analysis.

EPCO-11. BIASES IN HEALTHY BRAIN SINGLE-CELL DATASETS: FINDING THE BEST COMPARISON FOR GLIOBLASTOMA STUDIES

Abstract Tumor cells undergo metabolic alterations that support their proliferation and simultaneously modify immune populations, resulting in an immunosuppressive tumor microenvironment. This study aims to compare the metabolic expression of immune cells within the tumor microenvironment to those in a healthy environment. However, there is a lack of single-cell transcriptomic data for normal brain cells, necessitating an analysis of different methods to compare immune cells in tumors with those in a healthy brain. To address this, we explored cell-to-cell metabolic variability using public single-cell RNA-seq (scRNA-seq) data of glioblastoma (n=338564), and from immune cells in the brains of post-mortem individuals (n=14177), fetuses (n=40000), and epilepsy patients (n=466), as well as from immune blood cells (n=329762). We employed ssGSEA to generate activity scores for 70 key metabolic pathways from KEGG for each individual cell. Known marker genes were utilized to identify normal cell populations (macrophages, T cells, oligodendrocytes) in datasets lacking cell type annotations. Our results revealed that blood cells exhibited high metabolic activity in macrophages but showed moderate activity in other cell types. Immune blood cells also demonstrated high oxidative phosphorylation (OXPHOS) but low fatty acid activity. Fetal brain cells displayed a lack of heterogeneity and showed high fatty acid activity with low OXPHOS activity. Brain cells from post-mortem individuals displayed heterogeneity but exhibited low metabolic activity across nearly all cells and pathways. Additionally, the hierarchy of pathway activity differed from the other datasets. Epilepsy data revealed a metabolic activity pattern similar to tumor-infiltrating immune cells, with comparable activity hierarchy and heterogeneity, although the sample size was limited for detailed comparison. Understanding the differences between types of datasets is important to choose the most appropriate comparison, allowing us to observe the metabolic alterations in immune cells caused by glioblastoma and to develop better immunotherapy strategies.

Knowledge-based inductive bias and domain adaptation for cell type annotation.

Measurement techniques often result in domain gaps among batches of cellular data from a specific modality. The effectiveness of cross-batch annotation methods is influenced by inductive bias, which refers to a set of assumptions that describe the behavior of model predictions. Different annotation methods possess distinct inductive biases, leading to varying degrees of generalizability and interpretability. Given that certain cell types exhibit unique functional patterns, we hypothesize that the inductive biases of cell annotation methods should align with these biological patterns to produce meaningful predictions. In this study, we propose KIDA, Knowledge-based Inductive bias and Domain Adaptation. The knowledge-based inductive bias constrains the prediction rules learned from the reference dataset, composed of multiple batches, to functional patterns relevant to biology, thereby enhancing the generalization of the model to unseen batches. Since the query dataset also contains gaps from multiple batches, KIDA's domain adaptation employs pseudo labels for self-knowledge distillation, effectively narrowing the distribution gap between model predictions and the query dataset. Benchmark experiments demonstrate that KIDA is capable of achieving accurate cross-batch cell type annotation.

Cytometry masked autoencoder: An accurate and interpretable automated immunophenotyper.

Single-cell cytometry data are crucial for understanding the role of the immune system in diseases and responses to treatment. However, traditional methods for annotating cytometry data face challenges in scalability, robustness, and accuracy. We propose a cytometry masked autoencoder (cyMAE), which automates immunophenotyping tasks including cell type annotation. The model upholds user-defined cell type definitions, facilitating interpretability and cross-study comparisons. The training of cyMAE has a self-supervised phase, which leverages large amounts of unlabeled data, followed by fine-tuning on specialized tasks using smaller amounts of annotated data. The cost of training a new model is amortized over repeated inferences on new datasets using the same panel. Through validation across multiple studies using the same panel, we demonstrate that cyMAE delivers accurate and interpretable cellular immunophenotyping and improves the prediction of subject-level metadata. This proof of concept marks a significant step forward for large-scale immunology studies.

Single-cell transcriptomics of rectal organoids from individuals with perianal fistulizing Crohn’s disease reveals patient-specific signatures

Perianal fistulizing Crohn’s disease (CD) is a severe gastrointestinal disorder causing extensive mucosal damage with limited treatment options. Severe manifestations of the disease appear at higher rates in non-Europeans but the genetic and cellular mechanisms driving the disease phenotypes remain poorly understood. Herein, we tested whether pathologic determinants in the epithelial stem cell compartment could be detected at the transcript level in rectal organoids derived from a diverse patient population. Rectal organoid and mucosal cells from endoscopic biopsies of each patient having perianal fistulizing CD or no disease controls were prepared for and sequenced at the single cell level. After cell type annotations based on expressed marker genes, samples were analyzed by principal components, for differential transcript expression, cell type proportions, and pathway enrichment. After QC, we produced 77,044 rectal organoid cells (n = 13 patients; 8 CD, 5 controls) with high quality sequences that identified 10 distinct epithelial subtypes, that we compared to 141,367 mucosal epithelial cells (n = 29 patients; 18 CD, 11 controls). Consistent with mucosal epithelial cells, rectal organoids prominently displayed disease signatures represented by the stem and transit amplifying regions of the rectal crypt, including alterations in transcriptional signatures of metabolic, epigenetic, and proliferating pathways. Organoids also retained their gender- and ancestral-specific gene expression signatures. However, they lacked many of the inflammatory signatures observed in epithelial cells from diseased mucosa. Perianal CD patient derived rectal organoids reflect gene expression signatures related to disease, gender, and ancestry, suggesting they harbor inherent properties amenable to further patient-specific, disease-related experimentation.

Single-cell RNA-seq analysis revealed the stemness of a specific cluster of B cells in acute lymphoblastic leukemia progression.

Childhood acute lymphoblastic leukemia (ALL) is a common pediatric cancer. The heterogeneous characterization of B cells in ALL progression poses new challenges to researchers. We used single-cell sequencing to explore the critical role of B cells in regulating the ALL immune microenvironment. We collected the single cell (sc) RNA-seq data of ALL and health sample from the gene expression omnibus (GEO) database, the "Seurat" and "harmony" R package was used for quality control and scRNA-seq analysis, in which the CellMarker2.0 database was used for cell type annotation. Subsequently, the FindAllMarkers function was used to identify the differentially expressed genes (DEGs) among various cell types and the DAVID database was applied for the biological process of DEGs. Then, the "inferCNV" package was used for copy number variation, regulons and cell communication were performed by SCENIC tool and CellChat package. The role of the target gene in regulating ALL progression was assessed using RT-qPCR, Transwell and scratch healing assays. We identified nine mainly cell clusters after scRNA-seq analysis, in which the B cells had higher infiltration proportion in the ALL samples and were sub-clustered into five cell sub-groups. The B cells 1 is closely associated with cell proliferation and stemness (TNFAIP3 and KDM5B), and the significant CNV of amplification occurred on chr6 and chr21 that supported stemness of B cells1. RXRB is a key transcription factor mediated the proliferation of B cells 1, which in turn suppressed hematopoietic stem cells (HSCs) proliferation and promoted cytotoxic NK/T cells activation through diverse cell communication ways. One of the key regulators of B cells is MYC, which promotes the migration and invasive ability of cell line leukemia cell lines. This study reveals the stemness characteristics of B cells and their critical role in ALL progression, a finding that provides new potential directions for the development of targeted therapies against ALL.

Whole-brain annotation and multi-connectome cell typing of Drosophila

The fruit fly Drosophila melanogaster has emerged as a key model organism in neuroscience, in large part due to the concentration of collaboratively generated molecular, genetic and digital resources available for it. Here we complement the approximately 140,000 neuron FlyWire whole-brain connectome1 with a systematic and hierarchical annotation of neuronal classes, cell types and developmental units (hemilineages). Of 8,453 annotated cell types, 3,643 were previously proposed in the partial hemibrain connectome2, and 4,581 are new types, mostly from brain regions outside the hemibrain subvolume. Although nearly all hemibrain neurons could be matched morphologically in FlyWire, about one-third of cell types proposed for the hemibrain could not be reliably reidentified. We therefore propose a new definition of cell type as groups of cells that are each quantitatively more similar to cells in a different brain than to any other cell in the same brain, and we validate this definition through joint analysis of FlyWire and hemibrain connectomes. Further analysis defined simple heuristics for the reliability of connections between brains, revealed broad stereotypy and occasional variability in neuron count and connectivity, and provided evidence for functional homeostasis in the mushroom body through adjustments of the absolute amount of excitatory input while maintaining the excitation/inhibition ratio. Our work defines a consensus cell type atlas for the fly brain and provides both an intellectual framework and open-source toolchain for brain-scale comparative connectomics.

SpotGF: Denoising spatially resolved transcriptomics data using an optimal transport-based gene filtering algorithm

Spatially resolved transcriptomics (SRT) combines gene expression profiles with the physical locations of cells in their native states but suffers from unpredictable spatial noise due to cell damage during cryosectioning and exposure to reagents for staining and mRNA release. To address this noise, we developed SpotGF, an algorithm for denoising SRT data using optimal transport-based gene filtering. SpotGF quantifies diffusion patterns numerically, distinguishing widespread expression genes from aggregated expression genes and filtering out the former as noise. Unlike conventional denoising methods, SpotGF preserves raw sequencing data, thereby avoiding false positives that can arise from imputation. Additionally, SpotGF demonstrates superior performance in cell clustering, identifying potential marker genes, and annotating cell types. Overall, SpotGF has the potential to become a crucial preprocessing step in the downstream analysis of SRT data. The SpotGF software is freely available at GitHub. A record of this paper's transparent peer review process is included in the supplemental information.

Identification of kidney cell types in scRNA-seq and snRNA-seq data using machine learning algorithms

IntroductionSingle-cell RNA sequencing (scRNA-seq) and single-nucleus RNA sequencing (snRNA-seq) provide valuable insights into the cellular states of kidney cells. However, the annotation of cell types often requires extensive domain expertise and time-consuming manual curation, limiting scalability and generalizability. To facilitate this process, we tested the performance of five supervised classification methods for automatic cell type annotation. ResultsWe analyzed publicly available sc/snRNA-seq datasets from five expert-annotated studies, comprising 62,120 cells from 79 kidney biopsy samples. Datasets were integrated by harmonizing cell type annotations across studies. Five different supervised machine learning algorithms (support vector machines, random forests, multilayer perceptrons, k-nearest neighbors, and extreme gradient boosting) were applied to automatically annotate cell types using four training datasets and one testing dataset. Performance metrics, including accuracy (F1 score) and rejection rates, were evaluated. All five machine learning algorithms demonstrated high accuracies, with a median F1 score of 0.94 and a median rejection rate of 1.8 %. The algorithms performed equally well across different datasets and successfully rejected cell types that were not present in the training data. However, F1 scores were lower when models trained primarily on scRNA-seq data were tested on snRNA-seq data. ConclusionsDespite limitations including the number of biopsy samples, our findings demonstrate that machine learning algorithms can accurately annotate a wide range of adult kidney cell types in scRNA-seq/snRNA-seq data. This approach has the potential to standardize cell type annotation and facilitate further research on cellular mechanisms underlying kidney disease.

ScGAA: a general gated axial-attention model for accurate cell-type annotation of single-cell RNA-seq data

Single-cell RNA sequencing (scRNA-seq) is a key technology for investigating cell development and analysing cell diversity across various diseases. However, the high dimensionality and extreme sparsity of scRNA-seq data pose great challenges for accurate cell type annotation. To address this, we developed a new cell-type annotation model called scGAA (general gated axial-attention model for accurate cell-type annotation of scRNA-seq). Based on the transformer framework, the model decomposes the traditional self-attention mechanism into horizontal and vertical attention, considerably improving computational efficiency. This axial attention mechanism can process high-dimensional data more efficiently while maintaining reasonable model complexity. Additionally, the gated unit was integrated into the model to enhance the capture of relationships between genes, which is crucial for achieving an accurate cell type annotation. The results revealed that our improved transformer model is a promising tool for practical applications. This theoretical innovation increased the model performance and provided new insights into analytical tools for scRNA-seq data.

Vitessce: integrative visualization of multimodal and spatially resolved single-cell data.

Multiomics technologies with single-cell and spatial resolution make it possible to measure thousands of features across millions of cells. However, visual analysis of high-dimensional transcriptomic, proteomic, genome-mapped and imaging data types simultaneously remains a challenge. Here we describe Vitessce, an interactive web-based visualization framework for exploration of multimodal and spatially resolved single-cell data. We demonstrate integrative visualization of millions of data points, including cell-type annotations, gene expression quantities, spatially resolved transcripts and cell segmentations, across multiple coordinated views. The open-source software is available at http://vitessce.io .

ScEMB: Learning context representation of genes based on large-scale single-cell transcriptomics.

The rapid advancement of single-cell transcriptomic technologies has led to the curation of millions of cellular profiles, providing unprecedented insights into cellular heterogeneity across various tissues and developmental stages. This growing wealth of data presents an opportunity to uncover complex gene-gene relationships, yet also poses significant computational challenges. We present scEMB, a transformer-based deep learning model developed to capture context-aware gene embeddings from large-scale single-cell transcriptomics data. Trained on over 30 million single-cell transcriptomes, scEMB utilizes an innovative binning strategy that integrates data across multiple platforms, effectively preserving both gene expression hierarchies and cell-type specificity. In downstream tasks such as batch integration, clustering, and cell type annotation, scEMB demonstrates superior performance compared to existing models like scGPT and Geneformer. Notably, scEMB excels in silico correlation analysis, accurately predicting gene perturbation effects in CRISPR-edited datasets and microglia state transition, identifying a few known Alzheimer's disease (AD) risks genes in top gene list. Additionally, scEMB offers robust fine-tuning capabilities for domain-specific applications, making it a versatile tool for tackling diverse biological problems such as therapeutic target discovery and disease modeling. scEMB represents a powerful tool for extracting biologically meaningful insights from complex gene expression data. Its ability to model in silico perturbation effects and conduct correlation analyses in the embedding space highlights its potential to accelerate discoveries in precision medicine and therapeutic development.

Definition of regulatory elements and transcription factors controlling porcine immune cell gene expression at single cell resolution using single nucleus ATAC-seq

The transcriptome of porcine peripheral blood mononuclear cells (PBMC) at single cell (sc) resolution is well described, but little is understood about the cis-regulatory mechanism behind scPBMC gene expression. Here, we profiled the open chromatin landscape of porcine PBMC that define cis-regulatory elements and mechanism contributing to the transcription using single nucleus ATAC sequencing (snATAC-seq). Approximately 22 % of the identified peaks overlapped with annotated transcription start sites (TSS). Using clustering based on open chromatin pattern similarity, we demonstrate that cell type annotations using snATAC-seq are highly concordant to that reported for sc RNA sequencing (scRNA-seq). The differentially accessible peaks (DAPs) for each cell type were characterized and the pattern of accessibility of the DAPs near cell type markers across cell types was similar to that of the average gene expression level of corresponding marker genes. Additionally, we found that peaks identified in snATAC-seq have the potential power to predict the cell type specific transcription starting site (TSS). We identified both transcription factors (TFs) whose binding motif were enriched in cell type DAPs of multiple cell types and cell type specific TFs by conducting transcription factor binding motif (TFBM) analysis. Furthermore, we identified the putative enhancer or promoter regions bound by TFs for each differentially expressed gene (DEG) with a DAP that overlapped with its TSS by generating cis-co-accessibility networks (CCAN). To predict the regulators of such DEGs, TFBM analysis was performed for each CCAN. The regulator TF-target DEG pairs predicted in this way were largely consistent with the results reported in the ENCODE Transcription Factor Targets Dataset (TFTD). This snATAC-seq approach provides insights into the regulation of chromatin accessibility landscape of porcine PBMCs and enables discovery of TFs predicted to control DEG through binding regulatory elements whose chromatin accessibility correlates with the DEG promoter region.

Robust self-supervised learning strategy to tackle the inherent sparsity in single-cell RNA-seq data.

Single-cell RNA sequencing (scRNA-seq) is a powerful tool for elucidating cellular heterogeneity and tissue function in various biological contexts. However, the sparsity in scRNA-seq data limits the accuracy of cell type annotation and transcriptomic analysis due to information loss. To address this limitation, we present scRobust, a robust self-supervised learning strategy to tackle the inherent sparsity of scRNA-seq data. Built upon the Transformer architecture, scRobust employs a novel self-supervised learning strategy comprising contrastive learning and gene expression prediction tasks. We demonstrated the effectiveness of scRobust using nine benchmarks, additional dropout scenarios, and combined datasets. scRobust outperformed recent methods in cell-type annotation tasks and generated cell embeddings that capture multi-faceted clustering information (e.g. cell types and HbA1c levels). In addition, cell embeddings of scRobust were useful for detecting specific marker genes related to drug tolerance stages. Furthermore, when we applied scRobust to scATAC-seq data, high-quality cell embedding vectors were generated. These results demonstrate the representational power of scRobust.

Flexible and robust cell type annotation for highly multiplexed tissue images.

Identifying cell types in highly multiplexed images is essential for understanding tissue spatial organization. Current cell type annotation methods often rely on extensive reference images and manual adjustments. In this work, we present a tool, Robust Image-Based Cell Annotator (RIBCA), that enables accurate, automated, unbiased, and fine-grained cell type annotation for images with a wide range of antibody panels, without requiring additional model training or human intervention. Our tool has successfully annotated over 1 million cells, revealing the spatial organization of various cell types across more than 40 different human tissues. It is open-source and features a modular design, allowing for easy extension to additional cell types.

AtML: An Arabidopsis thaliana root cell identity recognition tool for medicinal ingredient accumulation

Arabidopsis thaliana synthesizes various medicinal compounds, and serves as a model plant for medicinal plant research. Single-cell transcriptomics technologies are essential for understanding the developmental trajectory of plant roots, facilitating the analysis of synthesis and accumulation patterns of medicinal compounds in different cell subpopulations. Although methods for interpreting single-cell transcriptomics data are rapidly advancing in Arabidopsis, challenges remain in precisely annotating cell identity due to the lack of marker genes for certain cell types. In this work, we trained a machine learning system, AtML, using sequencing datasets from six cell subpopulations, comprising a total of 6000 cells, to predict Arabidopsis root cell stages and identify biomarkers through complete model interpretability. Performance testing using an external dataset revealed that AtML achieved 96.50% accuracy and 96.51% recall. Through the interpretability provided by AtML, our model identified 160 important marker genes, contributing to the understanding of cell type annotations. In conclusion, we trained AtML to efficiently identify Arabidopsis root cell stages, providing a new tool for elucidating the mechanisms of medicinal compound accumulation in Arabidopsis roots.

A natural language processing system for the efficient extraction of cell markers

Single-cell RNA sequencing (scRNA-seq) has emerged as a pivotal tool for exploring cellular landscapes across diverse species and tissues. Precise annotation of cell types is essential for understanding these landscapes, relying heavily on empirical knowledge and curated cell marker databases. In this study, we introduce MarkerGeneBERT, a natural language processing (NLP) system designed to extract critical information from the literature regarding species, tissues, cell types, and cell marker genes in the context of single-cell sequencing studies. Leveraging MarkerGeneBERT, we systematically parsed full-text articles from 3702 single-cell sequencing-related studies, yielding a comprehensive collection of 7901 cell markers representing 1606 cell types across 425 human tissues/subtissues, and 8223 cell markers representing 1674 cell types across 482 mouse tissues/subtissues. Comparative analysis against manually curated databases demonstrated that our approach achieved 76% completeness and 75% accuracy, while also unveiling 89 cell types and 183 marker genes absent from existing databases. Furthermore, we successfully applied the compiled brain tissue marker gene list from MarkerGeneBERT to annotate scRNA-seq data, yielding results consistent with original studies. Conclusions: Our findings underscore the efficacy of NLP-based methods in expediting and augmenting the annotation and interpretation of scRNA-seq data, providing a systematic demonstration of the transformative potential of this approach. The 27323 manual reviewed sentences for training MarkerGeneBERT and the source code are hosted at https://github.com/chengpeng1116/MarkerGeneBERT.

Integrating large-scale single-cell RNA sequencing in central nervous system disease using self-supervised contrastive learning

The central nervous system (CNS) comprises a diverse range of brain cell types with distinct functions and gene expression profiles. Although single-cell RNA sequencing (scRNA-seq) provides new insights into the brain cell atlases, integrating large-scale CNS scRNA-seq data still encounters challenges due to the complexity and heterogeneity among CNS cell types/subtypes. In this study, we introduce a self-supervised contrastive learning method, called scCM, for integrating large-scale CNS scRNA-seq data. scCM brings functionally related cells close together while simultaneously pushing apart dissimilar cells by comparing the variations of gene expression, effectively revealing the heterogeneous relationships within the CNS cell types/subtypes. The effectiveness of scCM is evaluated on 20 CNS datasets covering 4 species and 10 CNS diseases. Leveraging these strengths, we successfully integrate the collected human CNS datasets into a large-scale reference to annotate cell types and subtypes in neural tissues. Results demonstrate that scCM provides an accurate annotation, along with rich spatial information of cell state. In summary, scCM is a robust and promising method for integrating large-scale CNS scRNA-seq data, enabling researchers to gain insights into the cellular and molecular mechanisms underlying CNS functions and diseases.

Single-cell DNA methylation analysis tool Amethyst reveals distinct noncanonical methylation patterns in human glial cells.

Single-cell sequencing technologies have revolutionized biomedical research by enabling deconvolution of cell type-specific properties in highly heterogeneous tissue. While robust tools have been developed to handle bioinformatic challenges posed by single-cell RNA and ATAC data, options for emergent modalities such as methylation are much more limited, impeding the utility of results. Here we present Amethyst, a comprehensive R package for atlas-scale single-cell methylation sequencing data analysis. Amethyst begins with base-level methylation calls and expedites batch integration, doublet detection, dimensionality reduction, clustering, cell type annotation, differentially methylated region calling, and interpretation of results, facilitating rapid data interaction in a local environment. We introduce the workflow using published single-cell methylation human peripheral blood mononuclear cell (PBMC) and human cortex data. We further leverage Amethyst on an atlas-scale brain dataset to describe a noncanonical methylation pattern in human astrocytes and oligodendrocytes, challenging the notion that this form of methylation is principally relevant to neurons in the brain. Tools such as Amethyst will increase accessibility to single-cell methylation data analysis, catalyzing research progress across diverse contexts.

VICTOR: Validation and inspection of cell type annotation through optimal regression

Single-cell RNA sequencing provides unprecedent opportunities to explore the heterogeneity and dynamics inherent in cellular biology. An essential step in the data analysis involves the automatic annotation of cells. Despite development of numerous tools for automated cell annotation, assessing the reliability of predicted annotations remains challenging, particularly for rare and unknown cell types. Here, we introduce VICTOR: Validation and inspection of cell type annotation through optimal regression. VICTOR aims to gauge the confidence of cell annotations by an elastic-net regularized regression with optimal thresholds. We demonstrated that VICTOR performed well in identifying inaccurate annotations, surpassing existing methods in diagnostic ability across various single-cell datasets, including within-platform, cross-platform, cross-studies, and cross-omics settings.