This crosswalk maps the cell type annotations from Azimuth https://azimuth.hubmapconsortium.org/ to the latest revision of Cell Ontology Version IRI: http://purl.obolibrary.org/obo/cl/releases/2024-05-15/cl.owl in which many new cell types have been added since the original cell type annotations were mapped for Human - PBMC, Human - Pancreas, Human - Kidney, and Human - Bone Marrow. New Azimuth references for Human - Lung v2 (HLCA), Human - Adipose, Human - Heart, Human - Liver and Human - Tonsil v2 were mapped for the first time in this crosswalk.

This crosswalk maps the cell type annotations from Azimuth (Hao et al. 2021) to Cell Ontology Version IRI: http://purl.obolibrary.org/obo/cl/releases/2025-04-10/cl.owl. More information on Azimuth is available at https://azimuth.hubmapconsortium.org/.

The Human Reference Atlas regularly adds new cell types to Cell Ontology; an overview and current numbers are available under "Ontologies Extended" at https://apps.humanatlas.io/dashboard/data.

Bibliography:

• Hao, Yuhan, Stephanie Hao, Erica Andersen-Nissen, William M. Mauck, Shiwei Zheng, Andrew Butler, Maddie J. Lee, et al. 2021. “Integrated Analysis of Multimodal Single-Cell Data.” Cell 184 (13): 3573-3587.e29. https://doi.org/10.1016/j.cell.2021.04.048.

Supplementary Table Annotation Markers Azimuth and Triana Thesis Robbe Fonteyn

When analyzing scRNA-seq data with clustering algorithms, annotating the clusters with cell types is an essential step toward biological interpretation of the data. Annotations can be performed manually using known cell type marker genes. Annotations can also be automated using knowledge-driven or data-driven machine learning algorithms. Majority of cell type annotation algorithms are designed to predict cell types for individual cells in a new dataset. Since biological interpretation of scRNA-seq data is often made on cell clusters rather than individual cells, several algorithms have been developed to annotate cell clusters. In this study, we compared five cell type annotation algorithms, Azimuth, SingleR, Garnett, scCATCH, and SCSA, which cover the spectrum of knowledge-driven and data-driven approaches to annotate either individual cells or cell clusters. We applied these five algorithms to two scRNA-seq datasets of peripheral blood mononuclear cells (PBMC) samples from COVID-19 patients and healthy controls, and evaluated their annotation performance. From this comparison, we observed that methods for annotating individual cells outperformed methods for annotation cell clusters. We applied the cell-based annotation algorithm Azimuth to the two scRNA-seq datasets to examine the immune response during COVID-19 infection. Both datasets presented significant depletion of plasmacytoid dendritic cells (pDCs), where differential expression in this cell type and pathway analysis revealed strong activation of type I interferon signaling pathway in response to the infection.

Single-cell RNA sequencing has revolutionized the study of cellular heterogeneity, yet accurate cell type annotation remains a significant challenge. Inconsistent labels, technological variability, and limitations in transferring annotations from reference datasets hinder precise annotation. This study presents a novel approach for accurate cell type annotation in scRNA-seq data using unique marker gene sets. By manually curating cell type names and markers from 280 publications, we verified marker expression profiles across these datasets and unified nomenclatures to consistently identify 166 cell types and subtypes. Our customized algorithm, which builds on the AUCell method, achieves accurate cell labeling at single-cell resolution and surpasses the performance of reference-based tools like Azimuth, especially in distinguishing closely related subtypes. To enhance accessibility and practical utility for researchers, we have also developed a user-friendly application that automates the cell typing process, enabling efficient verification and supporting comprehensive downstream analyses. The desktop application can be accessed at https://omnibusx.com.

When analyzing scRNA-seq data with clustering algorithms, annotating the clusters with cell types is an essential step toward biological interpretation of the data. Annotations can be performed manually using known cell type marker genes. Annotations can also be automated using knowledge-driven or data-driven machine learning algorithms. Majority of cell type annotation algorithms are designed to predict cell types for individual cells in a new dataset. Since biological interpretation of scRNA-seq data is often made on cell clusters rather than individual cells, several algorithms have been developed to annotate cell clusters. In this study, we compared five cell type annotation algorithms, Azimuth, SingleR, Garnett, scCATCH, and SCSA, which cover the spectrum of knowledge-driven and data-driven approaches to annotate either individual cells or cell clusters. We applied these five algorithms to two scRNA-seq datasets of peripheral blood mononuclear cells (PBMC) samples from COVID-19 patients and healthy controls, and evaluated their annotation performance. From this comparison, we observed that methods for annotating individual cells outperformed methods for annotation cell clusters. We applied the cell-based annotation algorithm Azimuth to the two scRNA-seq datasets to examine the immune response during COVID-19 infection. Both datasets presented significant depletion of plasmacytoid dendritic cells (pDCs), where differential expression in this cell type and pathway analysis revealed strong activation of type I interferon signaling pathway in response to the infection.

The graph representation of the cell type annotations for Azimuth dataset.

Additional file 2: Table S1. Summary of scRNA-seq and scATAC-seq dataset. Table S2. Marker genes of each cell type. Table S3. Cattle PBMC cell type annotation under three resolutions using Azimuth. Table S4. Cattle PBMC cell type annotation using SingleR. Table S5. The expression of 93 cell cycle-related genes in each cell. Table S6. The summary information of TF identified by SCENIC. Table S7. Single cell’s pseudotime value obtained from Monocle2. Table S8. Gene list of each module identified by WGCNA. Table S9. Enrichment results of each module identified by WGCNA.

Additional file 3: Table S10. Differentially expressed genes with each cell cluster between CO and T1 identified by edgeR. Table S11. Differentially expressed genes with each cell cluster between CO and T2 identified by edgeR. Table S12. Differentially expressed genes with each cell cluster between CO and T3 identified by edgeR. Table S13. Differentially expressed genes with each cell cluster between T1 and T2 identified by edgeR. Table S14. Differentially expressed genes with each cell cluster between T1 and T3 identified by edgeR. Table S15. Differentially expressed genes with each cell cluster between T2 and T3 identified by edgeR. Table S16. Human and cattle PBMC cell type annotation under three resolutions using Azimuth.