1. Select a dataset from the BLUE drop-down Menu.
  2. Click the blue 'Load Dataset' button.
  3. The 'Start Exploring' button will appear when data is loaded.
  4. (Optional) come back to this page to load another dataset.

Select a Dataset

About Us

This interactive web application is designed for lay scientists to explore complex single-cell RNA datasets from human and mouse cardiovascular tissues.

PlaqView is officially published at Atherosclerosis!

Please cite PlaqView if you find us helpful. Submit your data to us at

Details of Single- Cell Dataset and IDs

Gene Expression Explorer

Use this module to query specific gene(s) and generate graphs for publications.

Gene names will be converted to correct format based on species. For a single gene, feature and ridge plots work well. For multiple genes, use dot plots.

Acronyms: chondrocytes (CH) common myeloid progenitor (CMP), dendritic cells (DC), endothelial cells (EC), fibroblasts (FB), granulocyte/monocyte progenitors (GMP), Macrophages (Mø), Monocytes (Mono), Natural Killer (NK), stem cells (SC), smooth muscle cells (SMC).

You must restart query if you change database Download Complete Pathway Enrichment Data

Annotation Explorer

Use this module to compare different cell annotation techniques.

By default, Plaqview uses Seurat v4 (Butler et al.) with the Tabula sapiens reference. Alternatively we found SingleR (Aran et al.), which uses reference-based algorithms to deduce cell identity of individual cells, to work better in some cases. Lastly, users can download differential gene expression lists created by Seurat v4 and manually define cells by clusters.

Note: scCATCH is no longer supported due to its underperformance in cardiac tissues. Automatic labeling methods may not work well for mouse data, we are actively working on this and we hope to bring a new solution soon.

Differential Expression by Cluster

Numbered Only (Unlabeled) Author Supplied (Manual) This will download a .csv of differentially expressed genes by cluster

Differential Expression by Cell Type

SingleR Seurat + Tabula sapiens This will download a .csv of differentially expressed genes as identified by individual cells

CIPR: Cluster Identity Predictor

Use this module to compare our annotation against CIPR references.


  1. Choose the PlaqView Annotation Method.
  2. Choose the CIPR Reference Database.
  3. Choose the CIPR Method (Calculates Cell Identity Score).
  4. Click Run CIPR.
  5. Optional: Drag a Box on CIPR Graph and Select Clusters for More Details.

CIPR provides a graphical representation of identity score of unknown or labeled cell clusters against known references.

This is an additional tool to help users further asset the quality of the current annotation or to further explore differential gene expressions.

We acknowledge and thank Ekiz et al. (2020) for their publication and development of CIPR. By default,PlaqView will run CIPR with the LogFC Dot Product Method (LogFC values of the matching genes are mutliplied and added together to yield an aggregate identity score). For details about CIPR, definitions, and references, please see their website.


  • Cluster: Cluster annotation provided by PlaqView.
  • Reference: Broad classification of the reference cell type.
  • Ref.ID: Shortened unique identifier for reference cell type.
  • Full.Name: Human-readable long name of the reference cell type.
  • Description: Additional details about this cell type.
  • Identity.Score: Identity score for the given reference cell type calculated via logFC dot product or correlation methods (see CIPR documents for more details).
  • Percent.Pos.Cor.: The percentage of the differentially expressed genes in PlaqView clusters is also differentially expressed in a similar fashion in the reference cell subsets (e.g. both upregulated and downregulated).

Metadata Explorer (experimental)

Use this module to query specific gene(s) in the context of the raw metadata associated with the original dataset. Please note: scRNA-seq metadata formatting has no common convention, and PlaqView may not be able to render completely. Furthermore, some datasets may not have sufficient metadata for graphing, and will generate errors.

When available, Factor-Type metadata are generally categories like gender, sex, etc., whereas Continuous-Type metadata are generally quality control (QC) metrics. You can change how the QC metrics are displayed in relation to the Factor-Type metadata.

Acronyms: chondrocytes (CH) common myeloid progenitor (CMP), dendritic cells (DC), endothelial cells (EC), fibroblasts (FB), granulocyte/monocyte progenitors (GMP), Macrophages (Mø), Monocytes (Mono), Natural Killer (NK), stem cells (SC), smooth muscle cells (SMC).

Common Metadata Annotations: nCount/nFeatures- Seurat's internal metric for QC control. Prediction.scores- SingleR's metric for QC. For other abbreviations not listed, please refer to the original publication of the respective dataset for explanations.

Trajectory Explorer

Use this module to explore RNA velocity and inferred cell differentiation paths.

Plaqview uses Monocle3 (Trapnell Lab) for cell RNA trajectory inference. We found that Monocle3 was the most flexible in handling complex datasets with multiple cell origins (as opposed to purified/cultured samples).

To further explore subsets of the data, use the right modules to select and re-cluster cells. It may take up to 10mins to recalculate depending on number of cells and complexity, please be patient.


  1. Highlight points by clicking and dragging.
  2. Click the 'Select' button.
  3. Repeat until all of the desired cells are black.
  4. Click 'Calculate'.


  • To start over, click 'Clear'
  • You can also choose/unchoose specific cells by clicking on them directly.
  • You can un-select a cluster of cells by clicking 'Choose' again.
Calculation may take up to 10mins. Results will appear below.

Original Full Trajecotry

Trajectory of Selected Cells

Druggable Genome Explorer

Use this module to explore potential drug targeting of specific cell types.

If you have a gene of interest, you can enter that gene to find known drugs or compounds that interact with it. A feature map of the gene of interest is provided from your selected dataset. The drug interaction data is derived from and made available through the rDGIDB package.

For details about each variables, click here. For details about interaction types, click here. For information about each databases and versions, click here.

This is not for use in diagnosis or treatment in the clinical setting, only for scientific research.

You must restart query if you change database. PubMed ID and citations of interactions are available in full download file.